Villalobos Fereres 2016 Copy Corrected 130222

Francisco J.
Villalobos · Elias Fereres

Editors
Principles of
Agronomy for
Sustainable
Agriculture
Principles of Agronomy for Sustainable Agriculture
Francisco J. Villalobos • Elias Fereres
Editors
Principles of Agronomy for

Sustainable Agriculture
Editors
Francisco J. Villalobos Elias Fereres
Instituto de Agricultura Instituto de Agricultura
Sostenible (CSIC) & Universidad Sostenible (CSIC) & Universidad
de Cordoba, Spain de Cordoba, Spain
Cordoba, Spain Cordoba, Spain
ISBN 978-3-319-46115-1 ISBN 978-3-319-46116-8 (eBook)

DOI 10.1007/978-3-319-46116-8
Library of Congress Control Number: 2016958826
© Springer International Publishing AG 2016

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or
dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained
herein or for any errors or omissions that may have been made.
Printed on acid-free paper
This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
This textbook is the result of a long experience teaching general agronomy at the
University of Cordoba (Spain). After many years of teaching the subject to agron-
omy engineering students in Spanish, we now offer a separate class, taught in
English, and this book reflects the organization and materials used in the class.
The book reflects our vision of agronomy as a complex, integrative subject at the
crossroads of many disciplines (crop ecology, agrometeorology, soil science, agri-
cultural engineering) with a strong emphasis on providing quantitative answers to
specific problems. Our experience has been primarily with water-limited agricul-
ture; hence, there is an emphasis throughout the book on the role of water in the
agronomy of agricultural systems. We also seek to leave behind artificial bound-
aries that have been created in the past among crop production areas such as
horticulture, pomology, and field crops that have led to separate journals and
professional careers in the past. In this book, we cover all common aspects of
crop management and productivity that should concern anyone dealing with the
management of agricultural systems, and we provide relevant examples from
different cropping systems, from herbaceous to woody crops.
Our quantitative approach is based on providing the ideas and concepts needed
as foundations in all the quantitative assessments required for making informed,
technical decisions in farm management. Farmers operate along the philosophy of
learning by doing (adaptive management), and agronomists should also follow the
same path, but they should have the knowledge and tools that are needed to first
correctly interpret the complex responses of the system to change and then provide
reasonable options for subsequent actions. This book does not fall in the category of
those that focus on providing prescriptive agronomic recommendations or blue-
prints that cannot be generalized because of their empirical nature. Rather, we have
tried to concentrate on the analysis of crop productivity processes which lead to
identifying the main factors affecting management decisions and on how to get
quantitative answers to agronomic problems in the context of making current
agricultural systems more sustainable.
v
vi Preface
From a teaching perspective, the book includes two short blocks on the envi-
ronment and crop productivity that could serve as an introduction for students with
no background in soil science, crop ecology, or agrometeorology. The third, larger
block, is devoted to specific crop production techniques (sowing, soil management,
irrigation, fertilizers, etc.). A number of our colleagues have contributed to the
writing, all with the aim of providing future agronomists and practitioners with the
quantitative tools required to calculate the adequate level of inputs (such as water,
nutrients, or energy) for sustainable crop production and to assess the yield
responses as a function of climate and soil conditions and of management options.
Cordoba, Spain Francisco J. Villalobos

Elias Fereres
Acknowledgements
We greatly appreciate the assistance of Carmen Ruz and Jose Luis Vazquez in
preparing the figures and tables of the book. We also would like to thank the
University of Cordoba and the Institute for Sustainable Agriculture (IAS-CSIC)
for supporting our teaching and research activities which formed the basis for this
book. Thanks are due to our colleagues that have contributed to the writing of
specific chapters. The feedback from hundreds of our students with their questions
and suggestions greatly helped in refining its structure and content. We have been
extremely lucky for the interactions with so many of our colleagues from all around
the world that have inspired many of the ideas presented here.
Finally we dedicate this book to our families, whose love and support has been
the engine moving us forward.
vii
Contents
1 Agriculture and Agricultural Systems . . . . . . . . . . . . . . . . . . . . . . . 1

Elias Fereres and Francisco J. Villalobos
Part I The Crop Environment

2 The Soil. Physical, Chemical and Biological Properties . . . . . . . . . . 15
Antonio Delgado and José A. Gómez
3 The Radiation Balance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Francisco J. Villalobos, José Paulo De Melo-Abreu, Luciano Mateos,
and Elias Fereres
4 Wind and Turbulent Transport . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Francisco J. Villalobos, Elias Fereres, and Luca Testi
5 Air Temperature and Humidity . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Francisco J. Villalobos, Luciano Mateos, Luca Testi,
and Elias Fereres
6 Soil Temperature and Soil Heat Flux . . . . . . . . . . . . . . . . . . . . . . . 69
Francisco J. Villalobos, Luca Testi, Luciano Mateos,
and Elias Fereres
7 The Energy Balance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Francisco J. Villalobos, Luca Testi, Luciano Mateos,
and Elias Fereres
8 The Water Budget . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Francisco J. Villalobos, Luciano Mateos, Francisco Orgaz,
and Elias Fereres
9 The Components of Evapotranspiration . . . . . . . . . . . . . . . . . . . . . 107
Francisco J. Villalobos, Luca Testi, and Elias Fereres
ix
x Contents
10 Calculation of Evapotranspiration and Crop Water

Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Part II Crop Productivity

11 Crop Development and Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Victor O. Sadras, Francisco J. Villalobos, and Elias Fereres
12 Plant Density and Competition . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
Francisco J. Villalobos, Victor O. Sadras, and Elias Fereres
13 Radiation Interception, Radiation Use Efficiency and Crop
Productivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
14 Effects of Water Stress on Crop Production . . . . . . . . . . . . . . . . . . 189
Victor O. Sadras, Francisco J. Villalobos, Francisco Orgaz,
and Elias Fereres
15 Limitations to Crop Productivity . . . . . . . . . . . . . . . . . . . . . . . . . . 205
Part III Crop Management

16 Sowing and Planting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
Francisco J. Villalobos, Francisco Orgaz, and Elias Fereres
17 Tillage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
José A. Gómez, Francisco Orgaz, Helena Gómez-Macpherson,
Francisco J. Villalobos, and Elias Fereres
18 Soil Conservation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
Helena Gómez-Macpherson, José A. Gómez, Francisco Orgaz,
19 Irrigation Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
Luciano Mateos
20 Irrigation Scheduling Using the Water Balance . . . . . . . . . . . . . . . 269
Francisco J. Villalobos, Luciano Mateos, and Elias Fereres
21 Deficit Irrigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
22 Control of Salinity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
Francisco J. Villalobos, Luciano Mateos, Miguel Quemada,
Antonio Delgado, and Elias Fereres
23 Fertilizers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321
Antonio Delgado, Miguel Quemada, and Francisco J. Villalobos
Contents xi
24 Nitrogen Fertilization I: The Nitrogen Balance . . . . . . . . . . . . . . . . 341

Miguel Quemada, Antonio Delgado, Luciano Mateos,
and Francisco J. Villalobos
25 Nitrogen Fertilization II: Fertilizer Requirements . . . . . . . . . . . . . 369
26 Fertilization with Phosphorus, Potassium and Other Nutrients . . . 381
Antonio Delgado, Miguel Quemada, Francisco J. Villalobos,
and Luciano Mateos
27 Fertigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407
Francisco J. Villalobos, Miguel Quemada, and Antonio Delgado
28 Manipulating the Crop Environment . . . . . . . . . . . . . . . . . . . . . . . 425
Francisco J. Villalobos, Luca Testi, and Luciano Mateos
29 Frost Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443
José Paulo De Melo-Abreu, Francisco J. Villalobos,
and Luciano Mateos
30 Control of Weeds and Other Biotic Factors . . . . . . . . . . . . . . . . . . 459
31 Application of Herbicides and Other Biotic Control Agents . . . . . . 471
Francisco J. Villalobos and Elias Fereres
32 Harvest and Conservation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 487
33 New Tools and Methods in Agronomy . . . . . . . . . . . . . . . . . . . . . . 503
Pablo J. Zarco-Tejada, Luciano Mateos, Elias Fereres,
34 Cropping and Farming Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . 515
Helena Gómez-Macpherson, Francisco J. Villalobos,
and Elias Fereres
35 Agronomy and the Sustainability of Crop Production . . . . . . . . . . 527
Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 543
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 547
Contributors
José Paulo De Melo-Abreu DCEB, ISA & LEAF, ULisboa, Lisboa, Portugal
Antonio Delgado Dpto. Ciencias Agroforestales, ETSIA, Universidad de Sevilla,
Sevilla, Spain
Elias Fereres Instituto de Agricultura Sostenible (CSIC) & Universidad de
Cordoba, Spain, Cordoba, Spain
José A. Gómez Instituto de Agricultura Sostenible (CSIC), Cordoba, Spain
Helena Gómez-Macpherson Instituto de Agricultura Sostenible (CSIC),
Cordoba, Spain
Luciano Mateos Instituto de Agricultura Sostenible (CSIC), Cordoba, Spain
Francisco Orgaz Instituto de Agricultura Sostenible (CSIC), Cordoba, Spain
Miguel Quemada ETS de Ingenierı́a Agronómica, Alimentaria y de Biosistemas,
Universidad Politécnica de Madrid, Madrid, Spain
Victor O. Sadras South Australian Research and Development Institute, Waite
Campus, Urrbrae, Australia
Luca Testi Instituto de Agricultura Sostenible (CSIC), Cordoba, Spain
Francisco J. Villalobos Instituto de Agricultura Sostenible (CSIC) & Universidad
de Cordoba, Spain, Cordoba, Spain
Pablo J. Zarco-Tejada Instituto de Agricultura Sostenible (CSIC), Cordoba,
Spain
xiii
Chapter 1
Agriculture and Agricultural Systems
Abstract Crop Ecology deals with agricultural ecosystems that are manipulated by
man to funnel the maximum energy into usable products (food and raw materials).
Agricultural ecosystems show normally low biodiversity, low autonomy and a short
trophic chain. The main features of farming systems are productivity, stability,
resilience, and sustainability, the latter indicating the ability to maintain a certain
level of production indefinitely. Production of agricultural systems requires inputs
of matter, energy and information. Normally the economic optimum provision of
inputs is below that necessary to achieve maximum production. Various parameters
have been defined to characterize the productivity of agricultural systems (potential
yield, attainable yield, actual yield).
Agricultural activity is characterized by uncertainty due to numerous environ-
mental and economic factors. Faced with uncertainty, farmers’ decisions are
focused on avoiding risk and that may lead to losing opportunities. To make rational
decisions the farmer has access to many sources of information, ranging from their
own experience to research/technology transfer. The current trend is to improve the
acquisition, sources, and the use of information on the agricultural system for
improved decision-making.
1.1 Introduction
According to recent FAO statistics, agriculture occupies 28 % of the land area of the
Earth, with 30 % devoted to crops and 70 % to pastures. Broadly, the cultivated area
is less than 10 % of the total land area, encompassing around 1500 million ha
largely unchanged since 1960, as expansion of new cultivated land has been offset
E. Fereres (*) • F.J. Villalobos

Instituto de Agricultura Sostenible (CSIC) & Universidad de Cordoba, Spain, Alameda del
Obispo s/n, 14004 Cordoba, Spain
e-mail: [email protected]; [email protected]
© Springer International Publishing AG 2016 1

F.J. Villalobos, E. Fereres (eds.), Principles of Agronomy for Sustainable
Agriculture, DOI 10.1007/978-3-319-46116-8_1
2 E. Fereres and F.J. Villalobos
by the disappearance of arable land due mostly to urbanization and to irreversible

soil degradation. Of this area, 90 % is devoted to annual crops and 10 % to perennial
crops. Irrigated lands occupy more than 300 Mha, representing 17 % of the culti-
vated area. Cereals are the dominant crops representing nearly 60 % of the total
cropped area. In fact, although more than 7000 plant species are used in agriculture,
only about 120 are considered important, and more than 90 % of the calories
consumed by humans today come from less than 30 species. This does not mean
that agricultural biodiversity is low, because there are germplasm collections
harboring many thousands of different genotypes for all the main crop species.
Farming allows man to produce food and other products by managing and
manipulating the trophic webs of ecosystems (Box 1.1). Agriculture is a set of
human interventions that alter ecosystems to maximize the yield of the desired
product and minimize energy losses along trophic chains. The science and technol-
ogy of producing and using plants for human use is Agronomy. It deals with the
exploitation by man of terrestrial ecosystems and has, therefore, its roots in
ecology. The ecosystems modified by agriculture are called agroecosystems and
the science that deals with their study is Crop Ecology. An agroecosystem is an
ecosystem managed by man with the ultimate goal of producing food and other
goods and services derived from agriculture. Population pressure has reduced the
area of ecosystems free of human intervention. However, there are reserves, forests
and other areas that may be called natural ecosystems, which are generally charac-
terized by higher biodiversity, longer trophic chains and higher autonomy than
agroecosystems.
The agroecosystem is characterized by the presence of a lower number of
species than the natural ecosystem. This lower diversity is a result of the need to
reduce energy losses along trophic chains in agricultural ecosystems, which aim to
remove all unwanted energy transfers (to parasites, pathogens or plants which
compete with the crop) and is usually associated with a shortening of the trophic
chain. The energy autonomy of agroecosystems is relatively low because they
depend on inputs of materials, energy and information provided by humans.
The unit of study in Crop Ecology is the field or plot. A community of plants,
along with management practices (e.g. tillage method, rotation, etc.) located on a
field is called cropping system. At this level one can analyze the production
processes of plants, their relationships with the soil and their dependence on the
aerial environment. By observing the same plot for several years we can analyze the
effects of rotation, tillage practices or crop residue management on soil properties
and resulting yields, as they are affected by the use of resources such as water and
nutrients. Economic analyses or determining manpower needs are often made at the
plot scale. A farm represents a single management unit constituted by a number of
fields or plots.
At a higher level of organization, the cropping system is part of a farming system
where other elements (e.g. livestock) are also managed by the farmer providing
inputs to the crops or using crop products. The different crops and management
practices prevalent in a given area, are called agricultural systems at a regional
scale.
1 Agriculture and Agricultural Systems 3
Agriculture, like all human activities, has known successes and failures through-
out its history. Today agriculture produces enough food for the vast majority of the
world population, despite the unprecedented population growth experienced over
the last 50 years. However, it has also negative environmental impacts such as soil
degradation and water pollution from the use of fertilizers and pesticides, excessive
use of water resources, and the reduction of biodiversity. Furthermore, other sectors
of society are very sensitive to a diverse set of problems created by agriculture,
notably those related to food safety and to the threats to natural ecosystems.
Box 1.1 Ecology and Ecosystems

Ecology is the science that studies the relationships between organisms and
their environment. The environment is the set of biotic and abiotic factors
affecting growth, reproduction or mortality of organisms. Environmental
factors may also be divided into resources when the factor is directly con-
sumed by the organism (e.g. a nutrient) or regulators, when they affect the
rate of use of resources (e.g. temperature).
An ecosystem is a group of organisms and the environment that coincide in
time and space. The ecosystem is the fundamental unit of study of ecology.
The organisms in an ecosystem are interrelated by flows of energy and
materials. The dimensional characterization of the energy flows in the eco-
system is called the trophic chain. The primary source of energy is solar
radiation, the driving force of life on Earth. Primary producers fix solar
energy through photosynthesis, transforming it into usable energy that will
move along the trophic chains. Over millions of years, the man has obtained
the energy required for living from different trophic chains, by hunting and by
gathering edible plants until 10,000 years ago, when agriculture was invented.
1.2 Characteristics of Agricultural Systems
The primary objective of farming is the production of sufficient food and other
goods and services so that the farm stays viable. Therefore, a key feature of
farming systems is their productivity, defined as output per unit of resource used,
commonly referred to the cultivated area, which is the primary limiting factor of
agriculture. Thus, productivity is defined as the yield of usable product per unit
area but can be applied to other natural or artificial inputs as radiation, water,
nutrients or labor, which are also typically measured per unit of area. The
productivity level further serves as an indirect measure of the efficiency with
which these inputs are used.
When characterizing agricultural systems, the term efficiency is often used to
define the ratios of crop productivity and certain inputs. For example, efficiency of
water use is defined as the ratio of yield to the volume of water used, but it would be
more correct to speak of the productivity of water or nutrients, expressed as kg/m3

water or kg/kg nutrient. In engineering, efficiency is the ratio between the output
and the input of any entity in a system, for example, the energy supplied to an
engine.
Besides productivity and efficiency there are other important properties of
agricultural systems. Yields may vary from year to year by weather and other
causes. The term stability refers to the magnitude of these oscillations. The lack of
stability causes fluctuations in production that threaten the persistence of agricul-
tural systems. This is particularly true when there are sequences of successive
years of low yields that may have a catastrophic effect on their economic viability.
Related to the fluctuations in productivity, there is another feature termed resil-
ience which is defined as the capacity of the system to recover from a catastrophic
event, for example a drought. High resilience is a desirable property of
agroecosystems.
Another feature of farming systems is their sustainability which indicates the
ability to maintain a certain level of production indefinitely. This feature stems
from the concept of sustainable development, a development model that proposes
economic growth without adversely affecting the opportunities of future genera-
tions. A farming system is considered sustainable when it is economically viable
and socially acceptable, however, one must define the time frame, because what is
feasible and acceptable today may not be so in the future. Thus in agricultural
systems it would be more correct to speak of the degree of sustainability: a system
will be more sustainable when its exploitation does not degrade the quality of
water and soil resources, and when current management practices do not affect the
productivity and viability of the system in the future. The improvement the
sustainability should be based on two objectives: reducing or eliminating, if
possible, the negative environmental effects of agriculture while maintaining
high productivity. Decades of intensive production in many agricultural systems
have caused negative environmental effects and have created awareness of the
need to focus on the sustainability of the agricultural systems, leading to a debate
about developing new forms of agriculture that ensure economic and ecological
sustainability.
1.3 Management of Agricultural Systems
The strategy of agriculture is to manipulate the environment and the plant commu-
nity to optimize the yield of goods useful to mankind. This involves establishing
communities (crops or pastures) dominated by species that distribute a large
proportion of the primary production to usable organs or materials. In addition,
the farmer tries to minimize system losses due to insects, diseases and weeds.
Farmers have numerous management tools to control their crops, such as tillage
for weed removal and seedbed preparation, choice of species and cultivars, sowing
date and sowing density, application of fertilizers and pesticides, etc. External
factors such as climate and markets are difficult to predict so the flexibility in
managing the crop is very important to minimize the risk of crop failure or of
economic losses in the farm. For example, an application of fertilizer may be
reduced or waived if the rainfall is very low or if the expected price of the product
is very low.
In general, for many resources the response curve of yield versus input level is
curvilinear and the maximum profit is obtained at a level of resources below (but
not far from) that required for maximum yield. This is because of the synergies that
occur among inputs and of the addition of fixed costs, which make low input
strategies generally inefficient. The more productive and more profitable farms
are those that use resource levels which are commensurate with the production
target, without any input clearly limiting yield. For example, there is little point to
provide additional water as irrigation if the additional quantities of fertilizer
required to realize the targeted yield are not provided.
Example 1.1 The response of a wheat crop to N fertilizer in a rainfed Medi-

terranean area is shown in Table 1.1. In this case the highest yield is achieved
using 250 kg of fertilizer N per hectare. However, the economic optimum is
achieved with an application of 200 kg N/ha. We can also see that a very
limited use of resources leads to worse economic performance than overuse.
Table 1.1 Analysis of the response of a cereal crop to N fertilizer applied. The selling
price is 0.25 €/kg, the fertilizer cost is 0.80 €/kg and the fixed cost is 200 €/ha
N applied Yield Income N cost Income N cost Net profit
kg N/ha kg/ha €/ha
0 1200 300 0 300 100
50 1929 482 40 442 242
100 2329 582 80 502 302
150 2558 640 120 520 320
200 2883 721 160 561 361
250 3020 755 200 555 355
The criteria for managing agricultural systems must take into account many
factors that are affected by farmer decisions. Not only plant and animal production
processes are important, as are economic objectives, but also the effects on soils,
water, animal welfare and human health, landscape and biodiversity, among others,
have to be considered. All these items have a different weight depending on the
farming system under consideration, although, as in any other business, when the
farm is not dedicated to the subsistence of the owner, it is handled essentially based
on economic criteria. Nevertheless, there are many facets to the management of

farming systems. In areas where the ratio population/arable land and input prices
are low (e.g. USA and Australia), the emphasis is on maximizing profit per unit of
labor. In Northern and Central Europe and in Japan, where arable land is the
limiting factor and input prices and wages are high, farmers tend to maximize
productivity per unit area. Similar goals are pursued in the agriculture of China and
India due to the limited availability of land per farm. These situations contrast with
those of many poor countries where labor is abundant and access to inputs and
capital is scarce.
Crop yields are close to their maximum potential only in a few areas (as in farms
of Japan and Northern Europe), so that the average yields of agricultural systems
are generally poor indicators of potential productivity. Actual yields lie in a broad
interval from zero (crop failure) to a maximum attainable level which is only
limited by the aerial environment (solar radiation and temperature regime), this
yield level is called potential yield. Actual yield is defined as the average yield of a
cultivar in all the fields of a farm or of a specific region, and represents the state of
the climate and soil and the ability of farmers to apply successfully the technology
available to them.
The potential yield of a species in an area is achieved when the technology is not
limiting, that is, when all inputs are used optimally. Strictly, this concept applies to
the yield of a well fit cultivar with no limitations due to water or nutrients and full
control of weeds, pests and diseases. In general, the potential yield is calculated
using theoretical models based on climate and other environmental factors and on
the morphological and physiological characteristics of the crop in question. In
practice, these estimates of potential yield should be contrasted against record
yields obtained by the best farmers in the same geographical area.
There is a considerable gap between actual and potential yield in most agricul-
tural systems, so sometimes other indices are defined for diagnostic purposes For
example, attainable yield is defined as the yield achieved within environmental
constraints of climate and soil of the area, using the best technology available today.
The yields obtained by the best farmers and research stations in the area are an
indicator of attainable yield. The attainable yield in particularly favorable years,
result in record yields.
The concepts of potential and actual yield (and to some extent attainable and
record yields) are very useful for the evaluation of farming systems and the
identification of possible improvements that will help in closing the gap between
them. These concepts are also used to define cultivation intensity. In the intensive
agriculture of Japan and Northern Europe, actual yields are close to potential
yields and the yield gap is small. As yields approach potential levels, there is little
incentive for farmers to further intensify production and there is always some gap
between actual and potential production in all systems. As the difference between
actual and potential yield increases, so do the opportunities to increase
productivity.
1.4 Types of Agricultural Systems
Agricultural systems can be classified according to various criteria. An ecologically

based approach is based on the type of trophic chain. The shorter chain is one in
which crops are directly consumed by humans. In other chains, crops or pastures are
eaten by livestock, which in turn is consumed by humans. The energy efficiency of
a system is lower the greater the number of levels of the trophic chain. On average
each transfer in a food chain has a net efficiency of about 10 %. Thus, for a net
primary productivity of 100 units, if it is consumed directly (vegetarian diet), the
transfer of energy is close to 100. If cattle are employed to transfer the energy to
humans only 10 would be recovered. This does not imply that animal husbandry
should be abandoned. On the one hand, animals are the only choice for exploiting
marginal areas where crop production is not possible (see below). On the other,
some animals can use materials not digestible by humans (e.g. cellulose in crop
residues) or not suitable for food (e.g. food leftovers, residues from industrial
processing).
Farming aims at minimizing energy flows through undesired routes (weeds,
insects, etc.) that end up in the level of decomposers. As we have seen, a short
food chain (crop ! humans) is the most efficient from an energy transfer stand-
point. However in many agricultural systems environmental conditions (for
instance, very shallow/poor soils) prevent obtaining products for direct use by
humans (e.g. grain production) and only pastures may be grown. There is also the
case of areas with semi-permanent flooding or the very arid areas. In all of these
situations, cattle allow the conversion of primary production to other usable forms
by man, even at the cost of lower efficiency.
Agricultural systems may be characterized also according to their position
within an interval that goes from subsistence agriculture to intensive agriculture.
In subsistence farming, many species are used, cultivars are adapted to the specific
environments, yield potential is low and actual yields are low but stable. They are
also very labor-intensive and livestock is a main component in nutrient manage-
ment. This leads to high energy efficiency. At the opposite extreme, intensive
agriculture is characterized by lower genetic diversity (both in terms of species
and cultivars) in search for high yields, greater use of machinery replacing labor, as
well as high use of fertilizers and pesticides, resulting in high productivity but often
with low efficiency.
Historically agriculture in developed countries has undergone a transition from
subsistence farming to intensive agriculture with a continuous increase in produc-
tivity and a gradual decline in energy efficiency. The routes differ depending on
how land use has evolved in the different countries: For instance, Canada, Australia
and large parts of the USA and Argentina, have not intensified their agriculture as
much as it has occurred in Northern and Central Europe and in Japan. In many
Asian countries, a very intensive agriculture is practiced with high use of certain
inputs and low yield gaps. Therefore in some developed countries we may find
extensive systems with low inputs, but high level of mechanization, that require
large areas for the farm to be economically viable, while in other countries (mostly
developing) highly productive systems with high use of labor may coexist with
subsistence agricultural systems.
The intensification of agriculture in many countries has led to major pollution
episodes due to excesses in the use of inputs such as fertilizers and pesticides
and, in some cases, to the production of agricultural surpluses due to
ill-conceived subsidies. In some cases, food safety incidents have been related
by the public opinion of these countries to agricultural intensification. This has
led to proposals to develop alternative agricultural systems, some based on
avoiding the use of mineral fertilizers and synthetic pesticides, such as in the
different forms of biological or ecological agriculture, called organic farming.
Other alternatives have proposed to adopt agricultural practices that are envi-
ronmentally friendly and that ensure the quality and safety of food. The term
“sustainable agriculture” refers to farming practices that allow the indefinite
maintenance (sustainability) of agricultural systems, which requires the conser-
vation of resources and the maintenance of economically viable farms. Some
experts speak of a transition from traditional agriculture (low input, low control)
to intensive agriculture (high input, low control), from which we must move to
an agriculture which is more sustainable (inputs optimized, high control), where
resources are used only in the appropriate amounts for each system and where
there is a better control of the environment and the crop.
1.5 Decision Making in Agriculture
Farmers must combine a number of biological, physical and economic factors when
making their decisions. The success of a farmer’s activity can be measured by
several variables (e.g. net income, yield, minimum risk, etc.), But not only is the
average value of the variable important, also are its statistical distribution and
extreme values. Agricultural activity is characterized by the uncertainty of a system
that depends on the weather, which is highly variable and on relatively
unpredictable biotic factors (pests and diseases). Therefore, the same agricultural
practices can lead to different yields in different years. One can therefore assume
that a set of agricultural practices will result in a frequency distribution of the
variable considered (e.g. yield). Knowledge of this distribution would be necessary
for the farmer to make decisions rationally. For example, a set of agricultural
practices can result in a high average yield, but very low yields in certain years,
which would have catastrophic effects on the economic viability of the farm. A
farmer may choose to get a lower average yield in exchange for avoiding those
years of very low yield.
Example 1.2 The result of two different nitrogen fertilization strategies

(A ¼ no fertilizer, B ¼ 50 kg N/ha) over 30 years in a cereal crop is shown in
Fig. 1.1. The average yield is higher for strategy B, but strategy A has a lower
standard deviation (217 vs 583 kg ha1). Strategy B (apply fertilizer) implies
a higher risk (lower yields in the worst years). If the farmer cannot stand a
single year of low yield, strategy A will be preferred, although it involves
lower average yield.
1.0
0.9
0.8
CUMULATIVE FREQUENCY
0.7
0.6
0.5
0.4
0.3
0.2 0N
0.1 50 kg N/ha
0.0
800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 3000 3200
WHEAT YIELD (kg/ha)
Fig. 1.1 Cumulative frequency of wheat yield with zero N fertilizer or 50 kg N/ha in a very
dry area. The mean and standard deviations are 1447 and 217 kg/ha for zero N and 1606 and
583 for 50 kg N/ha, respectively
The uncertainty of farming is not just the result of climate variability and the
possible occurrence of pests or diseases. Prices of agricultural products and inputs
can deviate substantially from the expected prices for the farmer, which further
hinders the process of decision making. The uncertainty of markets is proverbial,
and more recently, the volatility of grain prices has caused food crises in several
countries. The large price fluctuations have a very negative influence in the
sustainability of farming, particularly in the case of fruits and vegetables, where
there are many intermediaries between the producer and the consumer.
The general historical trend in agricultural systems around the world has been to
develop management practices that reduce the risk, that is, to ensure sufficient
yields in adverse years, but that do not fully exploit the potential in the most
favorable years, even by sacrificing some yield in average years, thus not achieving
the maximum average yield. In the past, when farmers did not have access to
insurance or subsidies, a sequence of several years of bad harvests put in serious
jeopardy the very existence of farming and the farmer. This has meant that avoiding
risk is a priority in the strategic decisions of agriculture in many areas, particularly
in rainfed systems. Examples would be the adoption of low tree density in rainfed
olive orchards so that the tree always has enough water available, even in the worst
droughty years, or the use of fallow in cereal rotations to store rainwater in the soil
during the fallow year for the next crop. This tendency to avoid risk partly explains
the slow adoption of new technologies in many agricultural systems, as compared
to other productive sectors.
Decisions on a farm can be classified into four types (operational, tactical,
strategic and structural) that correspond to different temporal scales. Operational
decisions are made during the growing season (e.g. irrigation dates, amounts of
fertilizer, date of application of an insecticide). Tactical decisions are made only
once for each crop (crop choice and sowing date, target yield, etc.). Strategic and
structural decisions have an impact on a number of crops (e.g. farm production
orientation, investment in machinery, infrastructure improvements). Obviously, if
we deal with multiannual crops (e.g. orchards), the temporary classification is
changed, as in this case, the tactical decision affects a number of crop seasons.
1.6 Sources of Information for Decision Making

in Agriculture
The farmer needs to know how the crop responds to different agricultural practices
in a particular environment (soil and climate). Also, in order to make operational
decisions, information is needed on the status of the crop and of the soil throughout
the season. The sources of information available to the farmer to make decisions are
quite diverse in terms of quality of information and of the cost associated with its
acquisition. Today we tend to consider information as a production factor, abso-
lutely necessary for efficient agriculture. The different sources of information
available on how crops respond to different management practices are discussed
below, with the exception of new technologies and the concept of Site Specific or
Precision Agriculture which are presented in Chap. 33.
1.6.1 Farmer’s Experience
The experience of the farmer is the traditional way on which all agricultural activity
is based, and it may be the best source of information on agricultural systems that
vary little over time. Local knowledge has developed over many generations and it
integrates the multiple features of the environment and the society as they affect
agriculture. It represents the human capital of a rural area which needs to be
protected and preserved. While traditional knowledge is always useful, sometimes
it presents difficulties for adopting innovations or to adapt to new situations (crop or

variety changes, new technologies, appearance of new pests and diseases, etc.).
The complexity of the agricultural system in which many biotic and abiotic
factors interact, makes it difficult to correctly interpret the observed responses and
to achieve a good understanding of system performance based solely on experience.
Often, a particular phenomenon may be due to causes that have nothing to do with
the apparent causes.
An additional problem is the adoption of farming practices, based on the
“collective experience”, which may be detrimental in the long term, even if they
do not cause any apparent injury, thus remaining unchanged over time. As an
example, we can cite the excessive tillage used for decades in many agricultural
systems. Finally, another drawback of using only experience as the basis for
decision making is the difficulty to detect processes that damage natural resources
over the long term. A classic example is the loss of soil by water erosion, difficult to
detect except when torrential rains create gullies that are obvious. However,
corrective measures, once gullies occur, are no longer effective. Other problems
such as salinization or acidification are very difficult to detect just by experience
until the problem is severe and difficult to overcome.
1.6.2 Research, Experimentation and Technology Transfer
Research leads to new knowledge, new processes or new products. Research and
experimentation are the only ways to produce new knowledge about the manage-
ment of agricultural systems. There has been great emphasis in many developed
countries for significant investment in agricultural research since the
mid-nineteenth century. It can be said that these investments were the engine of
economic development in these countries until the early decades of the twentieth
century. Subsequently, investments in agricultural research have been the basis on
which they have founded the notable increases in agricultural productivity since
1950, to the point that all the impact studies show that these investments stand as
one of the best business of the public sector of all times. The success of agricultural
research in the developed countries led to the creation around 1960 of a network of
international agricultural research centers located in developing countries such as
Mexico, the Philippines, India, Nigeria, etc. These centers are managed and coor-
dinated through the Consultative Group on International Agricultural Research,
which brings together more than 50 countries and international organizations and
have been responsible for the worldwide development and dissemination of new
varieties for the major crops, and for the introduction of management techniques to
intensify production in a more sustainable fashion.
For scientific knowledge to reach the farmer and to adapt it to its needs,
institutions were needed to transfer the new knowledge in parallel to those dedi-
cated to research. These institutions are called agricultural extension services. The
prestige and usefulness of extension services have been highly variable in the
different countries, according to the investment, its tradition and the various forms
of organization adopted. The growing use of the Internet as a source of information
and of technology transfer has also taken place in agricultural extension. Extension
services of U.S. universities often maintain pages with plenty of information for
farmers in different states. An example is the page about horticulture at the
University of California at Davis (http://virc.ucdavis.edu).
Agricultural research is often based on field experiments, usually performed in
experiment stations, which are farms devoted to research and technology transfer.
Results of experimentation have a limited validity in agriculture. As said before, a
set of agricultural practices could have different results in different years. The same
applies to the results of one experiment. Thus agricultural experimentation is slow
and expensive (needs to be replicated for a number of years). Adopting the
experimental results of a single year can lead to significant errors. It is necessary,
therefore, to consider the results of several years and yet, there is uncertainty in
extrapolating the results to other environments. This limitation highlights the need
to use other tools to complement experimentation in decision-making.
1.6.3 Commercial Information
Many of the inputs needed in farming are commercialized and the private sector has
made significant investments in research, particularly in recent years, and is very
active in technology transfer in the agricultural sector. For products and services
offered by the private sector, technology transfer is very effective. However, there
are issues of agricultural production systems, for example in the area of natural
resources management, where there are many stakeholders and where societal
interests may not always compatible with the interests of the private sector.
Furthermore, there are no economic incentives for the private sector to generate
all the information that is required for the sustainable management of natural
resources. While commercial information can be very useful for the farmer, often
it is promoted in such a way that tends to overestimate the benefits of the products.
Examples include the indiscriminate use of foliar fertilization and of some soil
amendments.
Bibliography
Connor, D. J., Loomis, R. S., & Cassman, K. G. (2011). Crop ecology: Productivity and
management in agricultural systems (2nd ed.). Cambridge: Cambridge University Press.
Sadras, V. O., Cassman, K. G. G., Grassini, P., Hall, A. J., Bastiaanssen, W. G. M., Laborte, A. G.,
Milne, A. E., Sileshi, G., & Steduto, P. (2015). Yield gap analysis of field crops – Methods and
case studies (Water reports no. 41). Rome: Food and Agriculture Organization.
Chapter 2
The Soil. Physical, Chemical and Biological
Properties
Antonio Delgado and José A. Gómez
Abstract This chapter provides a basic description of soil properties and pro-
cesses, stressing the concept that the soil is a dynamic entity where complex
interactions among its biological, chemical and physical components take place.
All these components and properties determine the functioning of the soil for
different purposes; this functioning is included in the concept of “soil quality”.
One of the most used definitions of soil quality is the capacity of a soil to function
within ecosystem boundaries to sustain biological productivity, maintain environ-
mental quality, and promote plant and animal health (https://www.soils.org/publi
cations/soils-glossary). Land use and management can have a profound impact on
many soil properties, thus indirectly affecting soil quality which can result in
improvements or constraints for productivity of agricultural lands and for agricul-
tural sustainability in the long term.
2.1 Introduction
From the point of view of agriculture, the soil offers support to plants and acts as a
reservoir of water and nutrients. However, in addition to being a physical medium,
the soil may be considered a living system, vital for producing the food and fiber
that humans need and for maintaining the ecosystems on which all life ultimately
depends. Soils directly and indirectly affect agricultural productivity, water quality,
and the global climate through its function as a medium for plant growth, and as
regulator of water flow and nutrient cycling. The soil structure should be suitable
for the germination of the seeds and the growth of the roots, and must have
A. Delgado (*)
Dpto. Ciencias Agroforestales, ETSIA, Universidad de Sevilla, Ctra Utrera km 1, 41013
Sevilla, Spain
e-mail: [email protected]
J.A. Gómez
Instituto de Agricultura Sostenible (CSIC), Alameda del Obispo s/n, 14004 Cordoba, Spain

Agriculture, DOI 10.1007/978-3-319-46116-8_2
16 A. Delgado and J.A. Gómez
characteristics that enhance the storage and supply of water, nutrients, gases and
heat to the crop. Soil chemistry is dominated by the interaction between its solid
components (primarily the insoluble compounds of silica, calcium and aluminum)
and its water phase. Understanding soil chemistry is of paramount importance,
since it is the basis of soil fertility and provides the needed knowledge to understand
the differences in fertility among different soils and their response to fertilization.
Sometimes soil chemistry can have a direct impact on soil physical conditions as in
the case of sodic soils with high exchangeable sodium content. The soil also hosts a
complex fauna and microbial web involved in many different biological processes,
which also affects its physical and chemical properties, and ultimately the produc-
tivity of agricultural ecosystems.
For a given soil, its properties depend on the history of the soil formation
(Fig. 2.1) and can be substantially modified by human intervention (e.g. through
agricultural practices). A proper understanding of soil characteristics and adequate
interpretation of the magnitudes of its properties, both combined under the broader
term of soil quality (Table 2.1), is required for proper management of agricultural
soils.
Fig. 2.1 Soil profiles showing two different degrees of development. Shallow Calcaric Cambisol
(left) and deeper Vertic soil (right)
Table 2.1 Some soil properties normally used in evaluating soil quality
Soil property
Physical Soil texture Bulk density Infiltration rate
Chemical Cation exchange capacity Organic carbon concentration Soil pH
Biological Soil respiration Earthworms presence Microbial
biodiversity
2 The Soil. Physical, Chemical and Biological Properties 17
2.2 Dynamics of Soil Formation and Soil Loss
Soil genesis refers to the developmental processes that the soil, as a natural entity,
has undertaken over long time periods as the result of the complex interactions of
physical, chemical and biological processes, as described in Fig. 2.2. Soil forming
processes usually refer to the results of the interaction of these processes of
different nature, such as the accumulation of soil components (e.g. organic matter),
formation on site of new ones (e.g. clay minerals or oxides), transport within the
soil profile (e.g. clay, carbonate or soluble salts), or changes in the aggregation state
of soil particles (e.g. formation of a structure). As mentioned in Sect. 2.1., these
processes will define the soil type and can strongly affect soil quality.
Available soil depth for plant growth (the depth of the soil profile that can be
explored by plant roots also termed rootable soil depth), a determining factor in
agronomy since it strongly affects overall crop development and soil productivity,
is the result of the balance between soil formation and erosion rates. Soil formation
rates are extremely low and mostly related to geology (bedrock properties) and
climate conditions. It is usually less than 5 mm per century (although rates range
from 0.01 to 40 mm per century). In landscapes that are not under quick geological
transformations, eg. alpine uplifting, these soil formation rates tend to be in
equilibrium with the erosion rates under natural vegetation. Natural erosion rates
range between 0.005 and 60 mm per century, and are mostly the result of water and
wind erosion and mass movement by gravitational forces.
SOIL FORMATION PROCESSES
Minerals, nutrientes:
Ions in soil solution
Oxides or alumina & iron

Wind
Silica Rain
Cold Hail
Heat Ice
Frost
2-layer 3-layer
clay clay
Swell, Dry, Mechanical
Espontaneous weathering
temperate, Tropical, weathering
bogs, Parent
arable
moist material
Silt Fine parent

material
Sand/
quartz
Chemical weadering:
acids, moisture
Fig. 2.2 Description of key processes in soil formation

Human interventions mainly by removing the protective plant cover can result in
accelerated erosion rates under inappropriate land use or soil management prac-
tices. These accelerated erosion rates can reach up to 50 mm per year, resulting in a
reduction of the soil profile depth and its degradation. Achieving sustainable
erosion rates is a major goal of soil conservation practices. Such rates are defined
as those which are either close to the soil formation rates or, at least below a given
safe rate (customarily below 10–100 mm per century) that extends far into the
future the impact of the imbalance between soil formation and soil erosion rates.
The use of soil conservation techniques aim at reducing erosion rates within the
range of 0.003–60 mm per century for achieving a more sustainable agriculture.
2.3 Soil Physical Properties: Texture and Structure
Soil physical properties determine many key soil processes (Fig. 2.3), and thus the
agronomical potential of a soil. Soil texture, which is a description of the size
distribution of the mineral soil particles composing the solid fraction of the soil
(from clay <2 μm to coarse particles >2000 μm) is perhaps the most important,
since it determines many other physical properties (such as infiltration rate) and
some chemical properties (such as cation exchange capacity). Clay mineralogy
influences the physical and chemical properties of soils, one of them the swelling-
shrinking behavior of the soil, e.g. vertisols, if the clay is an expansive type. Soil
structure describes the arrangement of mineral particles and organic matter in the
soil, and particularly the arrangement of pores among these particles, and also the
stability of this arrangement under external forces such as traffic or rainfall drops. In
contrast to texture, soil structure can be substantially modified by soil management.
Distribution of pore space and texture determines soil water retention properties
(see Chap. 8) which are characterized by the relationship between soil water content
Plant processes
• Water uptake
Soil texture &
• Root growth
structure Clay mineralogy
Soil physical properties Soil processes & management
• Hydraulic conductivity • Infiltration, runoff, waterlogging

• Water retention and porosity • Leaching, salinization, acidification
• Mechanical properties • Trafficability and erosion
(e.g. penetration resistance)
Fig. 2.3 Description of key soil physical properties and related soil processes and management
issues (Adapted from Geeves et al. (2000))
and soil water potential (tension). This relation is determined by soil structure and
pore size distribution when the soil is at low water tension (wet) and mostly by soil
texture at high water tension (dry soil). Bulk density, the ratio between soil dry mass
and volume, is a very important soil property influencing soil water retention,
aeration, trafficability, and infiltration rate, and is extremely sensible to soil man-
agement. Average soil porosity (calculated as P ¼ 1 – bulk density/particle density
[taken usually as 2.65 t/m3]) is a useful parameter. Soil mechanical resistance
reflects the resistance encountered in the soil to penetration and is directly related
to soil compaction. Mechanical resistance of the soil increases sharply as the soil
dries and is used to complement the information provided by bulk density.
Soil permeability is a broad term used to define the ability of the soil for
transmitting water. It is important to understand the water dynamics and the
water balance of the soil (Chap. 8) and it must be known for accurate management
of irrigation (Chaps. 19 and 20). It is determined partly by texture, with sandy soils
having high permeability as compared to clay soils and it can be altered by soil
management (e.g. tillage, Chap. 17). Other parameters that reflect the water trans-
mission properties of the soil are the infiltration rate, i.e. the rate of water flow
through the soil surface, and the hydraulic conductivity, i.e. the ability of a soil to
conduct water, a parameter extremely sensitive to soil water content.
Soil particles and the void spaces with their continuity and sizes are all arranged
in clusters giving way to a certain structure. Soil physical, chemical and biological
properties all influence soil structure by providing means that help held together soil
aggregates. Structure affects many soil properties that are relevant in agronomy.
The penetration of plant roots, the movement and storage of soil water, the aeration
and the mechanical resistance of a soil are some of the more relevant properties
influenced by the way soil aggregates are clustered together in a structure. Common
management practices such as tillage can change soil structure very rapidly. Such
short-term changes are reversible but the long-term degradation of soil structure is a
serious problem as it is associated with decreased water infiltration and increased
erosion risks. Organic matter plays an important role in facilitating aggregate
formation and its long-term decline contributes to the loss of soil structural stability.
2.4 Soil Chemical Properties
2.4.1 pH
Soil pH is that of the soil solution that is in equilibrium with protons (H+) retained
by soil colloids (clays, organic matter, oxides). The soil pH is determined in the
laboratory as the pH of soil suspensions in water or salt solutions (usually 0.1 M
CaCl2 or 1 MKCl). The degree of acidity or alkalinity of a soil is a very relevant
property affecting many other physicochemical and biological properties. Problems
derived from acidic soils or acidification of agricultural soils can be overcome by
increasing base saturation and pH with soil amendments (liming). Basic or alkaline
soils are the consequence of the buffering of soil pH by base elements or by the
presence of buffering compounds such as carbonates. Calcareous soils are those
with an appreciable concentration of CaCO3 which buffers soil pH near 8.5; the
presence of other carbonates (Mg or Na in sodic soils) can buffer soil pH well above
8.5. The pH of a calcareous soil cannot be changed due to its high buffering
capacity and its limitations for agricultural use, mainly related to restrictions in
nutrient uptake and in plant nutrition, may be overcome with special fertilizer
products and fertilization strategies.
Some of the soil fertility features affected by soil pH include:
(a) Availability of mineral elements to plants in the soil. At low pH, the risks of
deficiency of base nutrients (Ca, Mg, and K) increases due to their low content;
also the solubility of Mo and P compounds is decreased, thus decreasing its
availability. On the contrary, Al concentration is increased (usually at pH <5.5)
and thus its toxicity effects; the concentration of Fe and Mn, essential nutrients
for plants, can be high enough at low pH as to cause toxicity. At high pH, the
solubility of many metals and trace elements is decreased, including essential
nutrients for plants such as Fe, Mn, Cu or Zn. Deficiency of Fe, known as iron
chlorosis, is frequent in basic soils (typically in calcareous ones).
(b) Biological properties: extreme pH values decrease microbial activity in soils,
which affects many soil processes (for instance, soil organic matter decompo-
sition, nitrification, and biological N2 fixation under acidic conditions, see
Chap. 24).
(c) Physical properties: low Ca concentration in acidic soils is usually related to an
increased dispersion of colloids if Al is not present at high concentration. Thus,
acidic soils can have poor soil physical properties, including poor structural
stability or low permeability.
2.4.2 Redox Status
The redox status of a soil is determined by the availability of electrons which can
participate in redox reactions (pE, – logarithm of the activity of electrons) and it is
controlled by physical conditions (water content and porosity) and biological
activity. It affects the solubility and speciation of elements with different redox
states, such as N, S, Fe, Mn, some toxic trace elements (e.g. As, Se), and even
C. Reducing conditions in agricultural soils usually occur at very high water
contents (saturation) since, under these conditions, oxygen is quickly consumed
by biological activity. Reducing conditions increase the solubility of Fe and Mn
compounds, enhancing the uptake of these nutrients by plants (which can become
toxic) and of elements adsorbed on Fe and Mn oxides (e.g. P and heavy metals).
2.4.3 Ion Retention in Soils
Ions can be retained in soils by precipitation and adsorption processes. Precipita-

tion means the formation of a new solid phase, e.g. when P fertilizer is applied to a
soil with a high Ca concentration, new crystals of Ca phosphates can be formed.
Adsorption is the accumulation of chemical species (sorbate) on the surfaces of an
existing solid in the soil (sorbent). Precipitated and adsorbed species are in equi-
librium with the soil solution (precipitation/dissolution and adsorption/desorption
equilibria).
Adsorption can be the consequence of chemical reactions with functional groups
of sorbent surface which is sorbate specific (e.g. P on hydroxilated surfaces), or
electrostatic attraction by sorbent surface which is not sorbate specific. Charge
associated with mineral and organic surfaces can be permanent and variable.
Permanent charge arises from isomorphic substitution within clay minerals. Vari-
able charge is the result of unsatisfied bonds at the edge of minerals and organic
matter and is pH dependent.
2.4.3.1 Exchange Capacity
Exchangeable ions are those weakly adsorbed by soil particles that can be displaced
from sorption sites by other ions in the solution. Exchangeable ions are essential for
maintaining plant nutrient reserves in the soil.
Cation exchange capacity (CEC) is measured as the amount of cations (equiv-
alents or moles of charge) which can be extracted by a high concentrated cation
solution (usually, 1 M K+ or NH4+). The CEC is usually dominated by Ca, Mg, Na,
K, Al, and protons. The selectivity or relative affinity of cation by sorbent surfaces
is based on the ion’s charge and size: the smaller the hydrated radius (cation + water
molecules strongly interacting by ion-dipole interaction) the greater the affinity
(ions with small dehydrated radius have large hydrated radius), and the higher the
valence the greater the exchanger preference for the cation; the affinity scale for
dominant cations in soils can be summarized:
Al3þ > Ca2þ > Mg2þ > NH4 þ > Kþ > Naþ
Base saturation is defined as ratio of base exchangeable cations (Ca, Mg, K, and
Na) to total CEC, which decreases at decreased pH in the soil. Ca, Mg, and K are
nutrients for plants; thus a high base saturation means a greater nutrient reserve than
a low base saturation for the same CEC. Low base saturation related to soil acidity
can determine Ca deficiency for crops. In order to guarantee good physical soil
properties (soil aggregation, structure stability, good aeration, and drainage) and
nutrition for crops, Ca must be the dominant cation in the exchange complex
(ideally >50 % of CEC); also it is desirable that the Ca/Mg ratio would be 5–10
and the K/Mg ratio 0.2–0.3 in order to avoid nutritional disorders (antagonisms) for
plants which can lead to a deficiency of a nutrient promoted by a high level of the
antagonistic nutrient.
2.4.4 Salinity and Sodicity
Salinity is defined as a high concentration of soluble salts (more soluble than

gypsum) in soils. A saline soil has a soluble salt concentration high enough to
negatively affect the growth and development of most cultivated plants. Classifi-
cation of saline soils and the assessment of the negative effects of salinity on crops
are based on the electrical conductivity (EC) of the saturation extract of the soil. If
the EC of the soil is higher than 4 dS/m it is defined as saline. There is ample
variation in the responses to salinity among different crops (Chap. 22). Crops highly
sensitive to salinity (e.g. carrot, bean, strawberry) are affected by EC values slightly
above 1 dS/m. On the opposite side tolerant crops such as barley and sugar beet
among others, can tolerate EC levels above 4 dS/m. Impact of salinity on plant
growth is caused by osmotic effects (decreased water potential in soil), and from
specific toxicity, typically due to high Cl or Na concentrations.
Sodicity is referred to a high exchangeable Na concentration in soils. Since Na
salts are common in saline soils, both problems are usually related. Na is a
monovalent cation with a big hydrated radius. Hence, high contents of Na adsorbed
on soil colloids promote their dispersion, thus negatively affecting soil physical
properties. A soil is classified as sodic if exchangeable Na accounts for more of
15 % of the CEC (Exchange Na percentage –ESP– >15). However, crops sensitive
to Na toxicity are affected at ESP >7 (e.g. peach, citrus, strawberry). Problems in
crops tolerant to Na toxicity (e.g. cotton or rye) usually are derived from physical
degradation of soil. Soils with EC > 4 dS/m and ESP >15 are classified as saline-
sodic. Problems derived from sodic soils can also be related to their very high pH
values (usually >8.5 if the soil is not saline).
Chapter 22 expands on the salinity problem in agriculture and describes the
approaches for its management and control.
2.5 Soil Biological Properties
Soils host a complex web of organisms (Fig. 2.4) which can influence soil
evolution and specific soil physical and chemical properties. For instance earth-
worms activity increases infiltration rate, or microbial activity decreases soil
organic matter due to mineralization.
Soil biological properties are also interconnected with other soil physical and
chemical properties; e.g. aeration, soil organic matter or pH affect the activity of
many microorganisms in soils which in turn perform relevant activities in carbon
and nutrients cycling. Examples of this interconnection were given in Sect. 2.4.
Roots Nematodes 1 Arthropods 1 Arthropods 2
Animals
Fungi Nematodes 2 Nematodes 3

Organic
Matter
Bacteria Protozoa
Fig. 2.4 Soil food web
Thus, changes in soil properties due to management can significantly affect bio-
logical properties in soils, some of them being extremely sensitive to soil manage-
ment; e.g. soil microbial activity can be greatly increased by improved drainage,
liming or organic amendments. That is why some soil biological properties can be
used as indirect indicators of appropriate soil management and good soil quality,
like soil respiration rate or some enzymatic activities that can be derived from
living organisms in soil.
Soil organic matter is a key factor affecting biological activity in soils. It is the
carbon source for many organisms, including soil microbiota. Not only the amount,
but also the type of organic compounds in the soil determines its biological activity;
e.g., microbial activity is greatly increased by incorporating fresh organic residues
(such as green manure or crop residues), which can be readily mineralized by
microbes. On the other hand, stable forms of organic matter (humic and fulvic
compounds), which constitutes most of the organic matter of soils in temperate
regions, is not a very suitable carbon source for soil microbiota, which explains the
long half-life of these compounds in soils (usually >1000 years); thus, stable
organic compounds do not contribute significantly to soil microbial activity but
constitutes an stabilized stored soil C pool which is very relevant to the C global
cycle, partially buffering the consequences of increasing C emissions to the
atmosphere.
The rhizosphere is the volume of soil altered by the root system and is the part
of the soil profile where the concentration of suitable C sources for many microor-
ganisms is greatest. Organic compounds exuded by plant roots (including organic
anions of low molecular weight) alter soil chemical properties and greatly increase
the biological activity in comparison to the bulk soil. The rhizosphere is a space of
intense interaction of plant roots with soil microorganisms. Rhizospheric microor-
ganisms can significantly affect plant development through the production of
growth regulators, by decreasing the incidence of plant diseases, and by increasing
nutrient availability to plants.
Table 2.2 Some soil biological properties

Property Comments
Respiration rate CO2 evolution under standard laboratory conditions or at the field
Potential N or C Increase in mineral N or C content under standard laboratory
mineralization conditions
Earthworms Density of earthworms
Bacterial biomass Total bacterial biomass for a given soil mass
Bacterial diversity It can be determined by functional groups, or describing genetic
diversity
Presence of pathogens By different pathology techniques, from cultures to DNA
profiling
Understanding soil biological properties is important for soil management but

also for prevention and control of crop pests and diseases. Many of the properties
indicated in Table 2.2 are a description of the diversity and activity of parts of the
soil food web, or of related properties such as soil respiration rate or organic matter
content.
2.6 Nutrient Cycles and Balances in the Soil
Nutrient in soils are present in different chemical forms, which can remain in
solution or bound to soil particles. Exchange of nutrients between different forms
or “soil pools” is governed by physical, chemical, or biological processes. All these
processes are included in the concept of “nutrient cycle” in soils. Since the soil is
not a “closed system”, gains or losses of nutrients from the soils occur to/from the
atmosphere or water courses (leaching or erosion), which links the “soil nutrient
cycle” with the “global nutrient cycle” in the Earth crust. The soil and global
nutrient cycles are affected by human activities. In agricultural soils, fertilization
clearly alters the cycle, introducing nutrients in the system. Without this supply, the
natural input of nutrients in soils would be much lower than typical crop extrac-
tions, thus inducing a “negative balance” which would cause a progressive deple-
tion of nutrients and thus a progressive loss of soil fertility.
A general nutrient cycle is represented in Fig. 2.5. The flux of nutrients to plant
roots comes from the soil solution, mainly as dissolved ions. The “labile nutrient
pool” is that readily equilibrated with the solution, as adsorbed ions described in
Sect. 2.4.3.1, those precipitated as soluble salts, or those present in organic com-
pounds which are readily mineralized. The “available pool” of nutrients is the
amount in solution plus that readily equilibrated with the solution (“labile
forms”); for a given nutrient it can be considered the amount that can be extracted
by successive crops until severe deficiency of this nutrient appears in crop.
Atmosphere
(N, C)
Fertilizer Crop
Uptake
Desorption Labile pool

Dissolution Adsorbed (exchangeable)
Mineralization
Nutient in Precipitated (soluble)
solution Organic (readily mineralizable)
Adsorption
Precipitation
Leaching Capilary Immobilization
Erosion rising
Non-labile pool
Adsorbed (non-exchangeable)
Precipitated (insoluble)
Water Organic (stable)
Primary minerals
Fig. 2.5 General cycle of nutrients in soil. In italics physical, chemical or biological processes
involved in nutrient cycle. Residue incorporation to soil involves nutrient recycling: not only in
soluble forms (e.g. K); most in organic forms or organic bound forms that can become part of the
labile or non-labile pool. Exchange between labile and non-labile forms implies the same
processes that those involved in the equilibria between labile forms and solution
Chemical (e.g. adsorption/desorption or precipitation/solubilization) and biological

(immobilization/mineralization) interactions affecting nutrient equilibria and
exchange rates between the labile fraction of the soil solid and solution phases,
ultimately determine the solution ionic activities and the transport of nutrients to
plant roots.
Accurate estimation of fertilizer requirements in modern agriculture is based on
the knowledge of nutrient cycles and the precise estimation of available nutrients
pools in soils through chemical methods. Mobile nutrients are considered those
which are not bound to soil particles. Nitrogen, in spite of ammonium being
adsorbed, it is readily transformed to nitrate, which is not adsorbed to soil particles.
In agricultural systems, where the contribution to available nutrient pool by organic
matter mineralization can be low, the major contributors to the available pool of
mobile nutrients are usually inorganic ions in the soil solution. Immobile nutrients
are those which are bound to soil particles through adsorption or precipitation
processes, being in this case the labile pool the major contributor to the available
pool. Immobile nutrients, such as P, K, Ca, or Mg, are less susceptible of loss
through leaching; on the other hand, the nature of chemical reactions involved in
their retention cause that only part of the nutrients supplied as fertilizer are
available to plants.
Bibliography
European Soils Portal (http://eusoils.jrc.ec.europa.eu/).

Ewing, S. A., & Singer, M. J. (2012). Soil quality. In P. M. Huang, Y. Li, & M. E. Sumner (Eds.),
Handbook of soil sciences: Resource management and environmental impacts (2nd ed., p. 26).
Boca Raton: CRC Press.
FAO Soils Portal (http://www.fao.org/soils-portal/en/).
Geeves, G. W., Craze, B., & Hamilton, G. J. (2000). Soil physical properties. In: P. E. V. Charman, &
B. W. Murphy (Eds.), Soils: Their properties and management (2nd ed.). Oxford: Oxford
University Press.
Heil, D., & Sposito, G. (1997). Chemical attributes and processes affecting soil quality. In E. G.
Gregorich & M. R. Carter (Eds.), Soil quality for crop production and ecosystem health.
Amsterdam: Elsevier Scientific.
Mass, E. V. (1996). Crop salt tolerance. In K. K. Tanji (Ed.), Agricultural salinity assessment and
management. New York: American Society of Civil Engineers.
Montgomery, D. R. (2007). Soil erosion and agricultural sustainability. Proceedings of the
National Academy of Sciences, 104, 13268–13271.
National Resources Conservation Service, (http://www.nrcs.usda.gov/wps/portal/nrcs/main/soils/
health/biology/).
Palm, C., Sánchez, P., Ahamed, S., & Awiti, A. (2007). Soils: A contemporary perspective. Annual
Review of Environment and Resources, 32, 99–129.
Rawls, W. J., & Brakensiek, D. L. (1989). Estimation of soil water retention and hydraulic
properties. Unsaturated Flow in Hydrologic Modeling NATO ASI Ser, 275, 275–300.
Thompson, A., & Goyne, K. W. (2012). Introduction to the sorption of chemical constituents in
soils. Nature Education Knowledge, 3(6), 15.
Chapter 3
The Radiation Balance
Francisco J. Villalobos, José Paulo De Melo-Abreu, Luciano Mateos,

and Elias Fereres
Abstract Solar radiation (short wave) is the energy source for photosynthesis,
warming and evaporation in agricultural systems. Its value can be calculated as a
function of latitude, time of year and cloud cover. Fifty percent of solar radiation is
available to photosynthesis and is called photosynthetically active radiation (PAR),
although only a very small fraction is actually used in this process. Net radiation is
obtained by discounting the reflected solar radiation (which depends on the albedo)
and longwave losses that depend on air temperature, humidity and cloud cover.
Plants intercept all the radiation fluxes. Shortwave radiation interception is modu-
lated by leaf angle distribution that varies with LAI and plant type. The fraction of
radiation intercepted by trees can be calculated assuming simple geometrical forms
(spheroids).
3.1 Introduction
Electromagnetic radiation is the basic physical phenomenon determining the envi-

ronment of crops. Solar radiation, which is shortwave radiation, constitutes the
primary energy source for crop production. In addition we have to consider the long
wave thermal radiation emitted by any object on the planet, including soil, crops,
water and the atmosphere. Moreover, light quality, i.e. its composition in different
F.J. Villalobos (*) • E. Fereres

Instituto de Agricultura Sostenible (CSIC) and Universidad de Cordoba, Spain, Alameda del
J.P. De Melo-Abreu
DCEB, ISA & LEAF, ULisboa, Tapada da Ajuda, 1349-017 Lisboa, Portugal
L. Mateos

Agriculture, DOI 10.1007/978-3-319-46116-8_3
28 F.J. Villalobos et al.
wavelengths, plays a key role in many developmental processes of plants, as

discussed in Chap. 11.
The total energy emitted for the whole electromagnetic spectrum is calculated by
the Stefan-Boltzmann Law:
E ¼ ε σ T4 ð3:1Þ
where ε is emissivity (or effectiveness of the body in the emission of radiation), σ is

the Stefan-Boltzmann constant (5.67 108 W/m2/K4) and T is the absolute temper-
ature of the surface of the emitting body. This Law applies to any object, either the
sun or the earth’s surface. In the wavelength interval between 8 and 14 μm, the
emissivity of plant surfaces is between 0.97 and 0.99 which is fairly close to that of
the “black body”, which by definition has unit emissivity.
3.2 Solar Radiation
The flux density of solar radiation in the limit of the atmosphere (extraterrestrial
radiation), on a surface perpendicular to the beam, when the sun and the earth are at
average distance apart, is called the “solar constant” and its value varies between
1350 and 1400 W/m2, with an average value of 1370 W/m2.
Considering a horizontal surface, if the ray and the normal to the surface are not
parallel, flux density can be calculated by the Lambert’s cosine law:
I ¼ I p cos θ ð3:2Þ
where θ is the zenith angle (angle between the radiant beam and the vertical) and Ip
is the flux density in the direction of the beam.
The zenith angle for a horizontal surface on the planet depends on the latitude
(λs), the solar declination (δs) and the time of day (expressed as hour angle, ha, that
varies from 0 to 360 , taking the value of 0 at solar noon):
cos θ ¼ sin λs sin δs þ cos λs cos δs cos ha ð3:3Þ
The solar declination ranges from +23.45 at summer solstice (Northern hemi-
sphere) to 23.45 at winter solstice and may be calculated (in degrees) as:

360 ðDOY 172Þ
δs ¼ 23:45 cos ð3:4Þ
365
where DOY is day of the year (DOY ¼ 1 for January 1 and DOY ¼ 365 for
December 31).
3 The Radiation Balance 29
Example 3.1 On February 1 (DOY 32) in Cordoba, Spain (λs ¼ 37.85 ), the
solar declination is

360 ð32 172Þ
δs ¼ 23:45 cos ¼ 23:45 cos ð138 Þ ¼ 17:4
365
At 3 h after solar noon: ha ¼ (15–12) 15 ¼ 45 , thus:
cos θ ¼ sin ð37:85Þ sin ð17:4Þ þ cos ð37:85Þ cos ð17:4Þ cos ð45Þ
¼ 0:35 and then θ ¼ arc cos ð0:35Þ ¼ 69:5 :
Legal time is obtained by adding or subtracting a certain number of hours to

standard time, plus a daylight savings time (usually 1 h) in summer. For instance in
Spain there is 1 h difference in the fall-winter period and 2 h in spring and summer
(daylight savings). To calculate actual solar time we must take into account the
longitude of the place as the path of the sun to the west has an apparent speed of
15 per hour. However we will ignore other phenomena related to the rotation of the
Earth that can change up to 16 min our predictions of solar time.
Example 3.2 Santiago de Compostela (Spain) is located at 42.9 N and

8.43 W. We will calculate solar time at 1500 h (legal time) on May 1.
As the date corresponds to spring-summer, solar time at the standard
meridian (in this case the Greenwich meridian) will be:
15–2 ¼ 13 h

Then we subtract 1 h per 15 longitude to the West:
13–8.43∙1/15 ¼ 12.44 h, which means that actual solar time is 12:26.
In crop ecology and agronomy, we are especially interested in three major bands
in the spectrum of solar radiation reaching the upper atmosphere. The infrared and
visible wavebands represent approximately 51 % and 40 % of the solar constant,
respectively, while the ultraviolet waveband is approximately 9 %. The visible
waveband, which ranges from 400 to 700 nm, is also the Photosynthetically Active
Radiation (PAR), although a very small fraction of this radiation is actually used in
this process. PAR may be expressed as radiation flux density (W m2) or as photon
flux density (mol m2 s1).
As the sun rays pass through the atmosphere, the radiation is altered in quantity,
quality and direction by the processes of absorption and scattering. The absorption,
which is a change from radiant energy to heat, results into heating of the atmosphere
and a reduction of the amount of radiant energy that reaches the ground. Absorption
is mainly due to ozone and oxygen, especially in the ultraviolet waveband, and
water vapor and carbon dioxide, in the infrared waveband. Some aerosols are also
important absorbers of shortwave radiation. The scattering occurs when photons hit
against the molecules composing the air and airborne particles and aerosols,
causing changes in the direction of radiation, but without removing energy from
the radiation. In the visible region of the spectrum, absorption by molecules of the
atmosphere is less important than scattering while in the infrared waveband the
opposite occurs. Solar radiation on the surface of the earth, measured perpendicu-
larly to the sun’s rays, rarely exceeds 75 % of the solar constant, due to absorption
and scattering.
3.3 Solar Radiation at Ground Level
As a result of atmospheric attenuation, solar radiation reaching the earth’s surface is

no longer only beam radiation as part of the radiation comes from all directions.
Beam radiation coming directly from the sun is called direct solar radiation and the
remaining part is called diffuse solar radiation, thus being the radiation scattered in
the atmosphere that reaches the surface coming from the entire sky hemisphere. The
sum of direct and diffuse solar radiation, measured on a horizontal flat surface, is
called global radiation, total radiation or simply solar radiation. On average, PAR
represents 50 % of the global radiation flux, while in the extraterrestrial radiation it
represents only about 40 %.
In general, the ratio of diffuse and direct radiation increases with latitude and
with the zenith angle as the path of the rays through the atmosphere gets longer.
This implies that both at sunrise and at sunset, the diffuse/direct ratio is higher than
at noon. Cloudiness also increases the ratio of diffuse to direct radiation, and when
the sky is completely overcast all solar radiation is diffuse. However, the absolute
maximum of diffuse radiation is reached when cloud cover is around 50 %.
Global radiation (Rs) during clear days follows a sinusoidal curve during the
daytime (Fig. 3.1) which can be measured by pyranometers or, on a daily time step,
estimated from extraterrestrial radiation (RA) and the ratio of the actual number (ns)
and the maximum possible number (Ns) of sunshine hours:

ns
Rs ¼ 0:25 þ 0:50 RA ð3:5Þ
Ns
Alternatively, when no data on sunshine duration is available we may use the

Hargreaves-Samani equation to calculate solar radiation as a function of air tem-
perature and extraterrestrial radiation as follows:
1000
900 SOLAR RADIATION
800 REFLECTED Rs
RADIATION FLUX (W m–2)
700 LONG WAVE LOSSES
600 Rn
500
400
300
200
100
0
–100
0 2 4 6 8 10 12 14 16 18 20 22 24
TIME GMT
Fig. 3.1 Daily time course of solar radiation, reflected shortwave radiation, long wave losses and
net radiation over a cotton crop in Cordoba (Spain) on June 27, 2003
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Rs ¼ K RS RA T max T min ð3:6Þ
where Tmax and Tmin are the daily maximum and minimum air temperature,
respectively.
Usually, the values of KRs vary between 0.16 and 0.19 K0.5 for interior and
coastal locations, respectively.
Daily extraterrestrial radiation (MJ/m2/day) may be calculated by integrating the
cosine law throughout the day:
h π
RA ¼ 37:4 dr sin λs sin δs hs þ cos λs cos δs sin hs ð3:7Þ
180
where hs is half the daylength (degrees):
hs ¼ arccos ½tgλs tgδs ð3:8Þ
And dr is the correction for changes in the distance between the earth and the sun,
which depends on the day of the year:

360 DOY
dr ¼ 1 þ 0:033 cos ð3:9Þ
365
From Eq. 3.7 we may deduce day length, i.e. the maximum duration of sunshine,
as:
2hs 1
Ns ¼ ¼ arc cos ½tgλs tgδs ð3:10Þ
15 7:5
On the other hand, Eq. 3.5 indicates that on clear days solar radiation is around
75 % of extraterrestrial radiation. On the average, solar radiation on overcast days is
only 25 % of extraterrestrial radiation.
Example 3.3 We will calculate daylength and solar radiation for clear days
on December 21 at Grand Rapids, Michigan (42.9 N) and South Hobart,
Australia (42.9 S).
Solar declination and the correction dr depend on the day of the year only:
December 21: DOY ¼ 355

360 ð355 172Þ
δs ¼ 23:45 cos ¼ 23:45 cos ð180Þ ¼ 23:45
365
360 355
dr ¼ 1 þ 0:033 cos ¼ 1:033
365
Grand Rapids:
hs ¼ arccos ½tg42:9 tgð23:45Þ ¼ 66:23∘
Daylength: Ns ¼ 2 hs/15 ¼ 8.82 h

h π i
RA ¼ 37:4 dr sin λs sin δs hs þ cos λs cos δs sin hs
180
¼ 11:66 MJ m2 day1
Clear day: Rs ¼ 0.75 RA ¼ 8.74 MJ m2 day1

South Hobart:
hs ¼ arc cos ½tgð42:9Þ tgð23:45Þ ¼ 113:8
Daylength: Ns ¼ 2 hs/15 ¼ 15.17 h

h π i
RA ¼ 37:4 dr sin λs sin δs hs þ cos λs cos δs sin hs
180
¼ 44:5 MJ m2 day1
Clear day: Rs ¼ 0.75 RA ¼ 33.4 MJ m2 day1
The annual time course of global radiation follows also sinusoidal patterns, with
amplitude that depends on the latitude of the site and its cloudiness. For example
Fig. 3.2 shows the annual curve of the average values of solar radiation at two sites
with very high and very low rainfall along with maximum solar radiation, calcu-
lated as 75 % of the extraterrestrial radiation. On the other hand, Fig. 3.3 shows the
35
AVERAGE MONTHLY RADIATION (MJ m–2 day–1)
30
25
20
15
10
POTENTIAL
5 KLAVER
PELOTAS
0
1 2 3 4 5 6 7 8 9 10 11 12
MONTH NUMBER
Fig. 3.2 Annual time course of the mean monthly solar radiation for two locations, Pelotas
(Brazil) with mean annual rainfall of 1395 mm and Klawer (South Africa) with mean annual
rainfall 174 mm
AVERAGE MONTHLY RADIATION (MJ m–2 day–1)
35
BOSASO
30 SHARJAH
LAS VEGAS
25
20
15
10
0
1 2 3 4 5 6 7 8 9 10 11 12
MONTH NUMBER
Fig. 3.3 Annual time course of the mean monthly solar radiation for three dry locations: Bosaso
(Somalia) (11.28 N), Sharjah (United Arab Emirates) (25.33 N) and Las Vegas (NV, USA)
(36.08 N)
annual curves of solar radiation for three dry locations differing in latitude. As we
move away from the Equator the amplitude of the annual radiation curve increases.
Mean annual solar radiation is usually between 15 and 22 MJ m2 day1 for
latitudes from 30 S to 30 N and decreases for higher latitudes (Fig. 3.4).
Once the solar radiation reaches the earth’s surface, part of the radiation is
reflected. We use the term albedo (α) to express the ratio of reflected to incident
radiation in the range of 0.3–3 μm. Some values of albedo of natural surfaces are
shown in Table 3.1.
Therefore, the short-wave radiation remaining on the surface of the earth can be
calculated as (1-α) Rs. The vegetation albedo values are usually between 0.15 and
30
POTENTIAL
25 ACTUAL
RADIATION (MJ m–2 day–1)
20
15
10
0
–80 –60 –40 –20 0 20 40 60 80
LATITUDE
Fig. 3.4 Mean annual solar radiation for different locations as a function of latitude
Table 3.1 Albedo for daily Surface Albedo

solar radiation of different
Fresh snow 0.80–0.95
surfaces
Dry sand 0.35
Soil, wet, dark when dry 0.08
Soil, dry, dark when dry 0.13
Soil, wet, light when dry 0.1
Soil, dry, light when dry 0.35
Water bodies 0.05–0.14
Annual crops 0.16–0.26
Orchards, deciduous forests 0.10–0.20
Coniferous forests 0.05–0.15
0.20 for forests and between 0.20 and 0.25 for field crops. The main factor that
determines the albedo of a soil is its color and surface water content. It is easy to see
that a dry soil gets darker after wetting. For example the soil of the Agricultural
Research Center of Cordoba (Spain) (medium texture, low in organic matter) has an
albedo of 0.16 when wet and 0.23 when dry.
3.4 Long Wave Radiation
All surfaces are emitters of long wave radiation following the law of Stefan-
Boltzmann (Eq. 3.1). Under clear skies, most of the radiation emitted by the earth’s
surface (i.e., terrestrial radiation) is absorbed by the molecules composing the
atmosphere, mainly water vapor and carbon dioxide, although nitrous oxide (N2O)
and methane (CH4) are also important absorbers. Radiation in the waveband
8–12 μm (i.e., the atmospheric window) is almost not absorbed by these gases
due to the small size of its molecules. Under cloudy skies, cloud droplets, however,
when present, contribute extensively to the absorption of the longwave radiation,
thus “closing” this atmospheric window. The remainder that is not absorbed is lost
into the extraterrestrial space. The radiation absorbed can be re-emitted to the earth
surface, thus constituting atmospheric radiation. This downward flux originates
mainly from the first kilometer of the atmosphere, from the emissions of those
constituents mentioned above, that are highly selective absorbers (thus highly
selective emitters) of long-wave radiation. For practical proposes only, since the
average temperature of the layer of air in the lower atmosphere is related to the air
temperature near the ground, although a lot colder than this air, it is possible to
apply the Stefan-Boltzmann law with a fitted apparent emissivity that take into
account the differences between the average temperature of this lower layer of the
atmosphere and air temperature near the ground, the amount of cloud cover and the
fact that the lower atmosphere is far from being a black body. Terrestrial radiation is
almost always higher that atmospheric radiation, and that results in net losses of
long-wave radiation.
Daily losses of long-wave radiation (Rb, MJ m2 day1) can be calculated as:

ns pffiffiffiffiffiffi
Rb ¼ 0:9 þ 0:1 ð0:34 0:14 ea Þ4:9 109 T 4 ð3:11Þ
Ns
where 4.9 109 is the Stefan-Boltzmann constant expressed in MJ m2 K4, ea is

the air vapor pressure (kPa), T is air temperature (K) and ns/Ns is the ratio of actual
sunshine duration (ns) and daylength (Ns).
Equation 3.11 indicates that long wave losses will be greater under clear skies
(high n/N), with lower humidity and with higher air temperature. Cloud cover has a
dramatic impact on losses of long wave radiation, as for given air temperature and
humidity, long-wave losses under an overcast sky are only 10 % of those when the
sky is clear.
3.5 Net Radiation
The net balance of radiation that remains on a surface of albedo α is expressed by

the equation:
Rn ¼ ð1 αÞRs Rb ð3:12Þ
Net radiation is thus the difference between the flux of radiation towards the surface
and from the surface of the Earth. It is therefore the energy available on the surface
for evaporation, heating of the air, the soil and the crop and to a lesser extent, for
photosynthesis.
Example 3.4 Let us calculate the net radiation over short grass (α ¼ 0.23) in
South Hobart (42.9 S) for a clear day on December 21 if the average air
temperature is 25 C and the air vapor pressure is 1.8 kPa. Solar radiation was
already calculated in Example 3.3.
Long wave loss:

15:17 pffiffiffiffiffiffiffi
Rb ¼ 0:9 þ 0:1 0:34 0:14 1:8 4:9 109 ð273 þ 25Þ4
15:17
¼ 5:88 MJ m2 day1
Net radiation:
Rn ¼ ð1 0:23Þ33:4 5:88 ¼ 19:8 MJ m2 day1
Figure 3.1 represents the daily time course of solar radiation, net radiation and
reflected solar radiation on a summer day on a cotton field in Cordoba, Spain.
Curves of Rs and Rn have similar shape but while the solar radiation flux is always
positive during the day and nil during the night, the net radiation is negative at
night.
The daily values of net radiation in summer are usually positive and become
very small as the nights get longer in the fall. At higher latitudes daily net radiation
reaches negative values during winter. In Cordoba, for example, solar and net
radiation values are highest in July although extraterrestrial radiation peaks in
June. This is explained by the higher average cloudiness of June as compared
to July.
3.6 Intercepted Radiation
Leaf area is a good indicator of the ability of the crop to intercept radiation. To
characterize the leaf area of a crop canopy we use the leaf area index (LAI), defined
as the ratio of total green leaf surface area (one side) and the ground surface.
According to Monsi and Saeki radiation transmitted through the canopy is an
exponential function of LAI:
I ¼ I 0 ek LAI
ð3:13Þ
where I0 and I are the flux densities above and below the canopy, respectively, and
k is the extinction coefficient.
Intercepted radiation will be the difference between incoming radiation and that
reaching the soil surface. Therefore:
I 0 I ¼ I 0 1 ek LAI ð3:14Þ
The above equations may be applied to any type of radiation in terms of wavelength
(e.g. PAR or Near Infrared, NIR), directional properties (direct or diffuse) or time
scale (instantaneous or daily) by taking the appropriate extinction coefficient.
The extinction coefficient depends on the angle of elevation of the sun and the
leaf angle distribution. The most useful approach is given by the ellipsoidal leaf
angle distribution of Campbell. For daily time step computations, extinction coef-
ficient is a parameter that may be fixed for the whole growing season or for specific
phenological phases. Example values of these parameters are given in Table 3.2.
The extinction coefficient may be related to the inclination angle of the leaves as
shown in Fig. 3.5. Vertical leaves (parameter of leaf inclination ¼ 0) have extinc-
tion coefficients around 0.4 while horizontal leaves approach 1.
The leaf angle distribution has agronomic and ecological implications. Small
plants with horizontal leaf distribution (i.e., higher k values) result in higher
radiation interception than those with more erect leaves. The drawback is that
when LAI is high the light distribution is very unequal, the lower leaves receive
too little light, which usually accelerates their senescence. On the contrary, more
vertical leaf angle distributions (i.e., lower k values) may be advantageous to
intercept radiation when the zenith angle is large (winter or high latitudes) and
leads to a more homogeneous distribution of radiation within the canopy when LAI
is high. Therefore the maximum LAI that may be sustained will be higher for low
extinction coefficient values. Ideally, for optimal radiation interception, the upper
leaves should be more erect and the lower ones more horizontal. In fact, modern
maize canopies have such leaf angle distribution.
Leaf level photosynthesis saturates with high irradiance (Chap. 13), but irradi-
ance decreases as we move down into the canopy, so that a large fraction of the
leaves will be below the irradiance saturation level. This results in crop carbon
Table 3.2 Daily extinctions coefficients (k) of PAR and X parameter of the ellipsoidal inclination
angle distribution for some crop canopies
Crop k (PAR) X parameter Source
Beans, pea 0.4–0.5 2, 3
Bell pepper 0.72 2.5–2.9 7
Cassava, peanut, cotton 0.80–0.87 5, 6
Forage and pasture (legumes) 0.8–0.9 1.5–3.3 3
Maize, sorghum, millet 0.57–0.70 1.40 1, 4, 5, 6
Oil palm 0.48 5
Oilseed rape 0.84 1.9–2.1 1, 2, 6
Potato 0.64 1.7–2.5 1, 6
Soybean, cowpea, pigeon pea 0.7–0.8 3, 5, 6
Sugar beet 0.68 1.5–1.9 1, 2, 6
Sugar cane 0.46 6
Sunflower 0.90 1.8–4.1 1, 6
Sweet potato 0.60 6
Wheat, barley, rice 0.44–0.52 1–1.2 1, 2, 6
Notes: (1) Campbell and Norman, 1998. (2) Hough, 1990. Eur. Commission Report EUR 13039
EN. (3) Jeuffroy and Ney, 1997. Field Crop Res. 53:3–16. (4) Kanton and Dennett, 2008. West
Afric. J. Appl. Ecol. 13:55–66. (5) Squire 1990. The Physiology of Tropical Crop Production.
CAB Int. (6) van Heemst, 1988. Simulation Report CABO-TT 17, Wageningen. (7) Vieira et al.
2009. Sci. Horticul. 121:404–409
1.2
1.0
EXTINCTION COEFFICIENT
0.8
0.6
0.4
Beam+Diffuse
0.2 Diffuse
0.0
0 1 2 3 4 5 6
PARAMETER OF LEAF INCLINATION
Fig. 3.5 Extinction coefficients for daily solar radiation using the ellipsoidal model of Campbell
assimilation increasing linearly with irradiance at the canopy level, as will be

shown in Chap. 13.
When leaf area cannot be determined, the degree of intercepted radiation may be
estimated by calculating the fraction of the ground area covered by the crop canopy
(horizontal projection). This can be easily assessed in the field using photography
and the appropriate software and is expressed as a percentage of full cover (100 %).
The ground cover is a useful parameter for characterizing canopy size from remote
sensing.
3.7 Radiation Interception of Trees
We consider the tree crown as a spheroid with constant Leaf Area Density
(μl, m 2 m3), horizontal radius r and height ht.
For any isolated tree radiation interception for rays with zenith angle θ is the
product of incoming radiation flux in the direction of the beam (Ip), Projected
Envelope Area in the θ direction (PEA(θ)) and the mean interception over PEA:
I i ¼ I p PEAðθÞ½1 tc ðθÞ ð3:15Þ
where tc(θ) is the mean transmissivity of the crown in the direction of the sun rays.
For a spheroid we have:
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
!2 ffi
u
u
PEAðθÞ ¼ πr2 tcos2 ðθÞ þ
ht
sin2 ðθÞ ð3:16Þ
2r
Note that PEA is related to the area of the shadow envelope (Ss) projected by the
tree on the horizontal plane (PEA¼Ss cos(θ)). The average transmissivity of the
spheroid may be calculated as:
1 ð1 þ AÞeA
tc ðθÞ ¼ 2 ð3:17Þ
A2
where
3 μl V
A ¼ G ðθ Þ ð3:18Þ
2 PEAðθÞ
where G(θ) is the projection function in the θ direction and V is the tree
volume (m3). Similar equations may be written for other solids of revolution
like semi-spheroids.
Daily radiation interception of isolated trees may be calculated using intercep-
tion for zenith angle 1 rad. For spheroids and semi-spheroids, of height h and
horizontal radius r, the relative intercepted radiation, i.e. the ratio of radiation
intercepted by the tree and incoming radiation on a horizontal surface may be
calculated as:

0 0 0:0036 Rsn N s
RRi ¼ 0:95 ð1 tc1 Þ PEA1 1 α þ α ð3:19Þ
0:75 RA
where Rsn (W m2) is average solar radiation normal to the sun beams for clear sky
conditions, which can be calculated as:
Rsn ¼ 0013 x2 þ 2:19 x 79:7 ð3:20Þ
where x ¼ 0:75 106 RA =ð3600 Ns Þ. The previous equations can be simplified to:

0 0 0:75 RA
RRi ¼ c1 PEA1 1 α þ α 1:84 ð3:21Þ
3:6 N s
Ns is daylength (hour) while RA is daily extraterrestrial radiation (MJ m2 day1).

The factor α0 reflects sky conditions, being 0 for completely overcast skies and 1 for
clear sky, and may be calculated using measured radiation:
Rs
α0 ¼ 2 0:5 ð3:22Þ
RA
The factors related to the trees are, in the first place, the projected envelope area
for 1 rad:

PEA1 ¼ π r 2 ap þ bp h=r ð3:23Þ
The mean interception of the tree envelope for 1 rad is:
1 tc1 ¼ 1 exp cp A þ dp A2 ð3:24Þ
where
μ π r3 h
A¼ ð3:25Þ
2 PEA1
The coefficients depend on tree shape:

For spheroids: ap ¼ 0.3, bp ¼ 0.35, cp ¼ 0.64, dp ¼ 0.026
For semi-spheroids: ap ¼ 0.36, bp ¼ 0.4, cp ¼ 0.646, dp ¼ 0.047
The equations presented here require knowing the value of leaf area density, the
ratio of leaf area and crown volume. We may take values as low as 0.5 m2m3 for
very sparse crowns up to 2–3 m2m3 for dense crowns.
Example 3.5 We have an olive orchard in Cordoba, Spain. Plant spacing is

7 3.5 m. The trees have horizontal radius 0.5 m, height 1.5 m and Leaf Area
Density 2 m2 m3. Let’s calculate radiation interception on 21 March, under
clear sky conditions. For that day and location: RA ¼ 29.8 MJ m2 day1 and
N ¼ 12 h.
Using Eqs. 3.23, 3.24, and 3.21:
Assuming a spheroid:
PEA1 ¼ 1.06 m2 and c1 ¼ 0.49, so RRi ¼ 0.69 m2 and intercepted PAR is 7 MJ
day1.
Assuming a semi-spheroid:
PEA1 ¼ 1.23 m2 and c1 ¼ 0.44, so RRi ¼ 0.71 m2 and intercepted PAR is
7.2 MJ day1.
If we want to express radiation interception at the orchard level to get the
fraction of intercepted radiation we simply divide the value of RRi by the area
per tree (7 3.5 ¼ 24.5 m2):
f PI ¼ 0:69=24:5 ¼ 0:03
The same orchard would have RRi ¼ 0.79 m2 on January 1 and 0.65 m2 on
June 21. This illustrates the fact that isolated trees will intercept a higher
fraction of radiation in winter than in summer.
As in herbaceous crops, the degree of radiation interception by a tree may be

estimated by determining the ground cover as the horizontal projection on the
ground of the tree shade at solar noon divided by the tree spacing. Given the
wide diversity of tree architecture, this is only a first approximation of the
intercepted radiation as calculated above.
Bibliography
Allen, R. G., Pereira, L. S., Raes, D., & Smith, M. (1998). Crop evapotranspiration. Guidelines for
computing crop water requirements (Irrigation and drainage paper 56). Rome: FAO.
Campbell, G. S., & Norman, J. M. (1998). An introduction to environmental biophysics.
New York: Springer.
Monteith, J., & Unsworth, M. (2013). Principles of environmental physics: Plants, animals, and
the atmosphere. Oxford: Academic.
Rosenberg, N. J., Blad, B. L., & Verma, S. B. (1983). Microclimate: The biological environment.
New York: Wiley.
Chapter 4
Wind and Turbulent Transport
Francisco J. Villalobos, Elias Fereres, and Luca Testi
Abstract The flow of wind over crop canopies causes a transfer of momentum
from the air to the canopy that generates turbulence which enhances the exchange
of matter and energy between the atmosphere and crops. Turbulence increases with
wind velocity and aerodynamic roughness which is proportional to crop height.
Wind speed varies logarithmically with height. This profile can be described
mathematically by two parameters that are related to crop height. Turbulence can
also be expressed as an inverse function of aerodynamic resistance, which is
indicative of the difficulty for turbulent transport and is therefore very high when
U is low and for smooth (short) crops. Inside the canopy layer the wind speed
acquires profiles more dependent on the architecture of the canopy than to the wind
vector over it. Wind speed changes considerably over space and time, being
generally low at night and maximum after noon; over the long term, higher average
wind speeds are often registered at high latitudes.
4.1 Transport of Mass and Energy from Crops
The disposition of net radiation over a crop surface takes place in several forms;
part is spent in convection, thereby increasing the temperature of the air within the
crop and the atmosphere above (sensible heat). The rest is spent in conduction
leading to an increase in the soil (and the crop) temperature, or as latent heat as a
result of evaporation. A small fraction of the energy is also spent on reduction of
CO2 (photosynthesis).

L. Testi

Agriculture, DOI 10.1007/978-3-319-46116-8_4
The transfer of mass (e.g. water vapor) or energy (e.g. sensible heat) is usually
expressed by the analogy of Ohm’s Law:
Cs Ca
Flow ¼
r
where Cs and Ca are the concentrations of material or energy levels at the canopy
top and in the atmosphere, respectively, and r is the resistance to the exchange. This
resistance may refer to different processes (e.g. conduction, convection, etc.). In the
case of heat transport by convection which is very similar to the transfer of
chemicals between the crop and the atmosphere, fluxes are enhanced by turbulence,
which in turn depends on wind speed.
4.2 Wind and Exchange of Matter, Energy and Momentum
Crops are subject to the mechanical action of wind, which moves and bends their
leaves, stems and branches. But wind has another essential effect on crops, as it
enhances the turbulent transport of water vapor, CO2 and heat. This flow is
characterized by turbulent air currents or eddies of many different sizes and
variable direction and is very effective as a transport mechanism. If the heat and
the gases were transported by a pure diffusion mechanism, the surface conditions on
earth would not be suitable for plant life due to the high temperatures which would
be reached and the limitation on the rate of downward flow of CO2 required for
photosynthesis.
Wind speed is determined by the transport of turbulence in the surface boundary
layer, i.e. the layer of atmosphere closer to the crops and the soil. If air flows
parallel to a flat surface, the profile of wind speed would be logarithmic (exponen-
tial increase with height) and velocity would tend to zero as we approach the
surface. This tendency is due to a frictional force between the surface and the air,
which is transmitted to upper air layers through the intermediate layers. The friction
force per unit area is called the shear stress (τs) and is proportional to the gradient of
wind speed:
dU
τ s ¼ μa ð4:1Þ
dz
where μa is the dynamic viscosity of air, U is wind speed and z is the height. The
dimensions of τs are the same as those of momentum per unit area and unit time
(momentum flux). This variable allows an analogy between heat (or mass) transport
and vertical transfer of momentum. The magnitude of the momentum flux is
indicative of the amount of eddies that are formed and, therefore, of the effective-
ness of the turbulent exchange of water vapor, heat, CO2 and other particles,
between the crop and the atmosphere.
4 Wind and Turbulent Transport 45
4.3 Profiles of Wind Speed Above Crop Canopies
Perfectly flat surfaces are very rare in nature, especially on land. When obstacles are
present (for example stones, soil aggregates, crop canopies) the wind profile is
affected. The obstacles hinder the movement of the air in comparison to an ideal flat
surface. First, the wind velocity is not zero at the surface as shown in Fig. 4.1, but at
an intermediate level between the surface and the obstacles’ height. This height is
called the zero plane displacement (d) and indicates the level above which momen-
tum is absorbed, namely the virtual level where friction forces are exerted by the
crop.
Another feature that influences the aerodynamics of a surface is the roughness
length (z0). This is a parameter used to quantify the distortion of the real wind
profile over a rough surface from the ideal logarithmic profile if the surface were
smooth; it is thus a measure of the aerodynamic surface roughness. Like the zero
plane displacement, it has also the dimensions of a length and its value ranges from
106 m for smooth ice to 0.3 m for orchards to 1 m for forests. Both d and zo are
related to the form of the crop canopy. Obviously, zo depends on the roughness of
the crop (uniformity of height among plants, distance between plants or between
rows, amount of ground cover, etc.). The parameter d also depends on the height
and flexibility of the plants, on foliage density, etc. A simple approximation for the
values of the two parameters for different crops is to calculate d as 0.65 h and zo as
0.13 h, where h is the crop height.
Typical profiles of mean horizontal wind speed above crops are shown in
Fig. 4.1. These profiles may be calculated for neutral conditions (see Chap. 5) as:
12
TALL
10 SHORT
8
HEIGHT (m)
0
0 1 2 3 4 5 6 7
WIND SPEED (m s–1)
Fig. 4.1 Wind profiles over a short (height 0.1 m) and a very tall (height 4.0 m) canopy

u* z dz
U ðzÞ ¼ ln ð4:2Þ
kk z0
where U(z) is mean wind speed at height z, kk is the von Kármán’s constant (about
0.4), u* is friction velocity, zo is roughness length and dz is the zero plane
displacement.
The friction velocity (so named for having the dimension of m s1), is related to
momentum flux by:
τs ¼ ρa u2* ð4:3Þ
where ρa is air density (kg m3), that depends on air temperature (T in K) and
atmospheric pressure (Pat in kPa) (the effect of air humidity is neglected):
Pat
ρa ¼ 3:484 ð4:4Þ
T
Applying the above equations, once zo and d are known, we may generate the whole
profile of wind speed as a function of wind speed measured at a reference height
(zm):
½lnðzm dÞ lnz0
U ðzÞ ¼ U ðzm Þ ð4:5Þ
½lnðz d Þ lnz0
Example 4.1
Figure 4.1 shows the wind profiles of two crops with height 0.10 m
(e.g. prairie) and 4.0 m (e.g. grain maize), respectively. To construct this
curve we assumed that the wind speed at 100 m height is 10 m/s in both cases.
The wind speed above maize is significantly lower than above grass, namely
maize slows down the wind more, or, in other words, it takes a greater amount
of momentum away from the wind. Friction velocities are 0.45 m/s and
0.67 m/s, for grass and corn, respectively, which correspond to values of
momentum transfer of 0.26 and 0.59 N/m2.
Often the only available information on wind speed comes from a nearby
weather station where the wind is measured at a standard height (z ¼ 2 m in
agrometeorological stations). However we often need to know U at a height z
over a canopy of height h. We can use Eq. 4.5 to calculate first U100, the speed at a
height of 100 m, which we may assume that does not vary spatially, i.e. it is the
same above the weather station and above the crop. If the station is located over
grass (h ¼ 0.12 m) then:
½lnð100 0:078Þ lnð0:0156Þ

U 100 ¼ U 2g ¼ 1:82 U 2g ð4:6Þ
½lnð2 0:078Þ lnð0:0156Þ
where U2g is wind speed measured at 2 m over grass.

And then we can calculate U(z) above our crop of height h:
½lnðz 0:65 hÞ lnð0:13 hÞ

U ðzÞ ¼ 1:82 U 2g ð4:7Þ
½lnð100 0:65 hÞ lnð0:13 hÞ
Example 4.2
Wind speed at z ¼ 2 m above grass (U2g) is 2.5 m/s. We will calculate U at
z ¼ 5 m over a 4-m maize crop.
½lnð5 0:65 4Þ lnð0:13 4Þ

UðzÞ ¼ 1:82 2:5 ¼ 1:33 m=s
½lnð100 0:65 4Þ lnð0:13 4Þ
4.4 Aerodynamic Resistance
The equation describing the flow of momentum in terms of the gradient of hori-
zontal wind speed (Eq. 4.1), can be written using the Ohm’s Law analogy, by
introducing an aerodynamic resistance to the transfer of momentum between
heights z1 and z2. Therefore, if
U2 U1
τ s ¼ ρa ð4:8Þ
r aM
and using Eqs. 4.2 and 4.3, then the aerodynamic resistance between a height z
where wind speed is U(z) and height d þ zo (where the extrapolated wind speed is
zero), will be:
h i2
ln zd
z0
z
U ðzÞ
r aM ¼ ¼ ð4:9Þ
k k 2 U ð zÞ u* 2
This equation indicates that the aerodynamic resistance to the flux of momentum
will be greater for short than for tall crops. In fact short and smooth surfaces are less
effective in slowing the wind that flows above them, thus less energy is transferred
from the wind to the surface for a given wind flow. The resistance decreases as wind
speed increases. In theory the resistance tends to infinity as U tends to 0. However in
the atmosphere that does not occur during the daylight hours because of buoyancy,
which will be treated later in Chap. 5. Suffice it to note here that as U decreases, the
lack of turbulence reduces the exchange of heat between the crop and the atmo-
sphere thereby increasing the temperature of the canopy and of the air in contact
therewith. This heated air tends to rise because of its lower density which causes
turbulence. This type of turbulence is called “thermal” as opposed to “mechanical”
turbulence which is due to friction of the wind on the crop. Both types of turbulence
coexist although mechanical turbulence prevails when U is high and thermal
turbulence is enhanced during the daytime when U is low.
Example 4.3
Let us calculate the aerodynamic resistance for z ¼ 5 m in the two cases
mentioned in Example 4.1 (maize with h ¼ 4 m and grass with h ¼ 0.12 m) if
the wind speed at 2 m height over grass (U2g) is 2.5 m/s.
We need to know first U(z) for z ¼ 5 m in both cases. We saw in Example
4.2 that U(z ¼ 5) ¼ 1.33 m/s for maize.
For grass, using Eq. 4.5:
½lnð5 0:078Þ lnð0:0156Þ

Uðz ¼ 5Þ ¼ U2g ¼ 1:195 U2g ¼ 2:99 m=s
½lnð2 0:078Þ lnð0:0156Þ
then applying Eq. 4.9 for the two surfaces:

Maize:
2
½lnð50:654
0:134 Þ
r aM ¼ ¼ 11 s=m
0:42 1:33
Grass:

5 0:65 0:12 2
ln
0:13 0:12
raM ¼ ¼ 69:2 s=m
0:42 2:99
4.5 Wind Speed and Turbulence at Canopy Height
The mathematical analysis of wind profiles above crops presented above should
allow the calculation of wind speed at canopy height as a function of crop height.
However, the existence of the so called Roughness Sub-layer which extends up to
2–2.5 times canopy height has not been taken into account. In that layer the profiles
of temperature and wind are distorted due to the proximity of vegetation. Including
this effect, the following equation allows calculating wind speed at canopy height
as a function of wind speed measured at 2 m height over grass:
2:6 U2g
Uh ¼ ð4:10Þ
6:6 lnðhÞ
This equation indicates that for most agricultural crops, with heights between 0.2
and to 3 m, wind speed at canopy height ranges between 0.32 and 0.46 times the
wind speed at 2 m over grass.
In the previous sections we have defined turbulence as an ensemble of eddies of
different size and properties (temperature, humidity, CO2 concentration) moving up
or down and following, on the average, the direction of wind speed. This vertical
exchange of eddies is responsible for the fluxes between the crop and the atmo-
sphere. For instance the convective transport of heat is due to warm eddies moving
up and cooler eddies moving down. We can characterize this turbulent exchange
using the average velocity of eddies going up which is equal to the average velocity
of those going down (as the mean vertical velocity is always close to zero). This
upward velocity, accompanied by a downward velocity of the same magnitude may
be seen as a mean renovation rate for the air located below, and we name it wr. The
renovation rate is proportional to horizontal wind speed (U) at canopy height. For
neutral conditions we can write:
wr ¼ 0:14 U ð4:11Þ
Note that wr has dimensions of velocity. By combining Eqs. 4.10 and 4.11 we can
calculate the renovation rate of air in contact with a canopy of height h:
2:6 U 2g
wr ¼ 0:14 ð4:12Þ
6:6 lnðhÞ
And the relative renovation rate with dimensions T1 would be obtained by
dividing wr by h.
Example 4.4
When wind speed over grass is 2 m/s the renovation rate of a maize crop with
height 2 m is:
2:6 2
wr ¼ 0:14 ¼ 0:123 m=s
6:6 lnð2Þ
And the relative renovation rate is 0.123/2 ¼ 0.0615 s1

This means that the air in contact with the canopy would be
completely renovated in 16.3 s (inverse of the relative renovation rate).
A similar concept of renovation rate is applied to greenhouses as a
measure of ventilation, and usually expressed in number of renovations
per hour. In the example above the relative renovation rate is equivalent
to 0.0615 s1 3600 s/h ¼ 221.4 h1.
4.5
4.0
3.5
3.0
HEIGHT (m)
2.5
2.0
1.5
1.0
0.5
0.0
0.00 0.25 0.50 0.75 1.00 1.25 1.50
U/U CANOPY HEIGHT
Fig. 4.2 Profile of wind speed (relative to that at canopy height 3.25 m) measured in an olive
orchard close to Cordoba (Spain) in summer 2012. Horizontal segments represent twice the
standard deviation
4.6 Wind Speed Inside Canopies
If the wind profiles are complicated above the canopy, they are even more compli-
cated inside it. In simple terms, the canopy height is divided into two or three zones.
The top layer (above dz), absorbs most of the momentum. In this layer the wind
speed decreases logarithmically as we enter the canopy and has the same direction
as the average wind over the canopy. This is clearly seen in Fig. 4.2 for a hedgerow
olive orchard down to 2.5 m height. Below that height, wind speed is almost
constant and rather small (35–40 % of that at the top of the canopy in the example
of Fig. 4.2).
4.7 Daily and Seasonal Variation of Wind Speed

and Direction
Both the daily and the seasonal time courses of wind speed are highly variable. The
predominant winds have traditionally been characterized by the wind rose, which is
a representation of the frequencies of occurrence of each wind direction. The
seasonal time course, besides being highly variable from year to year, is often
site-specific. As an example, Fig. 4.3 shows the wind speed in Fuente Palmera
(Spain) during the year. In this case the highest values of U occur in spring and
summer and lower values occur in autumn and winter. The large differences
WIND SPEED (m s–1) 6
0
0 30 60 90 120 150 180 210 240 270 300 330 360
DOY
Fig. 4.3 Mean daily wind speed at 2 m height at Fuente Palmera (Spain) 2007–2013
7
AVERAGE WIND SPEED (m s–1)
5
AMARILLO
4 CORDOBA, SPAIN
BANGUI
3
0
1 2 3 4 5 6 7 8 9 10 11 12
MONTH NUMBER
Fig. 4.4 Time course of monthly mean wind speed in Bangui (Central African Republic),
Amarillo (Texas, USA) and Cordoba (Spain)
between locations in wind patterns are evident in Fig. 4.4 which shows the monthly
average values of U in Bangui (Central African Republic), Cordoba (Spain) and
Amarillo (Texas). The large variation among locations in wind speed is illustrated
in Fig. 4.5 that shows mean annual wind speed for many locations as a function of
latitude. Both mean wind speed and its variability increase as we move away from
the Equator.
12
WIND SPEED (m s–1) 10
0
–80 –60 –40 –20 0 20 40 60 80
LATITUDE (degree)
Fig. 4.5 Mean annual wind speed for different locations as a function of latitude (Source:
Climwat-FAO)
6
WIND SPEED (m s–1)
0
0 2 4 6 8 10 12 14 16 18 20 22 24
TIME GMT
Fig. 4.6 Wind speed at Cordoba (Spain) in June 2003 plotted as a function of time. Each point is
the 10-min average wind speed
In the diurnal time course we usually observe that calm winds predominate
during the night while the maximum wind speed occurs during the day. Figure 4.6
shows the diurnal time course of wind over grass at Cordoba during June: we
observe that U is low at night, especially at dawn, and that U increases during the
day.
Fig. 4.7 Wind rose for wind speed at Espiel (Spain) 2003–2012
A wind rose is a graphical representation of the distribution of wind speed and

wind direction at a given site over a specific period of time. Presented in a circular
format, the wind rose shows the frequency of winds blowing from each direction,
typically classified in 16 cardinal directions, such as north (N), NNE, NE, etc.,
where North corresponds to 0 , East to 90 , South to 180 and West to 270 . The
distance from the center represents the frequency of that particular direction, which
may also be given for different wind speed classes (see example in Fig. 4.7).
Bibliography

New York: Springer.
the atmosphere. Oxford: Academic Press.
New York: Wiley.
Chapter 5
Air Temperature and Humidity
Francisco J. Villalobos, Luciano Mateos, Luca Testi, and Elias Fereres
Abstract Air temperature shows unstable profiles during the day and stable pro-
files during the night. Therefore, canopy temperature is generally higher than that of
the air during the day and lower during the night. Heat transfer between the crop and
the atmosphere is sustained by turbulence and will be more effective the higher the
wind speed, i.e. when aerodynamic resistance is low. In situations of unstable
atmosphere, turbulence is enhanced (by added thermal turbulence) while in a stable
condition turbulence is reduced. The water vapor content of the atmosphere can be
expressed by different variables (vapor pressure, relative humidity, vapor pressure
deficit, mixing ratio, vapor density). The flow of water vapor (equivalent to energy
spent as latent heat) between the crop and the atmosphere is directly proportional to
the vapor pressure difference and inversely proportional to the sum of the canopy
and aerodynamic resistances.
5.1 Introduction
Air temperature controls the functioning of terrestrial ecosystems. Crop tempera-

ture affects photosynthesis, growth and development rates, transpiration, etc. Crops
are heated by radiation absorption. Part of the absorbed energy is used to heat the air
(sensible heat) which, in turn, determines the air temperature above the crop.
Air humidity is important in crop production for several reasons. First, water
and/or humidity are essential for living organisms in the agricultural ecosystem to
grow and complete their life cycles. Secondly, moisture plays an important role in
energy exchange. Changing water from liquid to vapor state, which occurs in the

L. Mateos • L. Testi

Agriculture, DOI 10.1007/978-3-319-46116-8_5
process of evaporation, requires 2.45 MJ/kg. When 1 kg of water vapor condenses

and then freezes, 2.8 MJ are released. The first process relates radiation absorption
to water use by crops and the second process is the basis of some methods used for
frost protection in horticulture.
5.2 Thermal Stability
An air parcel that rises adiabatically (see Box 5.1) from one level to a higher one is
always at the same temperature (and thus, the same density) as the surrounding air.
The corresponding adiabatic temperature profile is drawn as the dashed line in
Fig. 5.1. Consider now an actual temperature profile such as that shown in Fig. 5.1
(right). In this case, by raising a parcel of air adiabatically from point C to D, that
parcel of air will have a higher temperature than the surrounding air (D0 ), will be
less dense and therefore will tend to continue rising. This atmospheric condition is
called unstable. This adiabatic rise happens when a hot eddy jumps suddenly from
the crop by a wind gust. If the movement is fast enough, the heat exchange between
the eddy and the air it finds will be very small allowing the assumption of adiabatic
conditions.
Suppose now that the actual temperature profile is of the type shown in Fig. 5.1
(left). In this case, by raising an air parcel adiabatically, its temperature will be
lower than the surrounding air and its density will be higher, so it will tend to return
to its original height. This condition of the atmosphere is termed stable and the
temperature profile is called inverted profile.
If the actual temperature profile follows the adiabatic profile, the corresponding
atmospheric condition is called neutral (dashed line in Fig. 5.1).
Box 5.1 Adiabatic Processes

A process (change of state) that occurs without gain or loss of heat is called
adiabatic process. If the pressure of an air volume changed adiabatically from
P1 to P2, the change of temperature (from T1 to T2) is given by:
ðT2 =T1 Þ ¼ ðP2 =P1 Þ1Cv=Cp ð5:1Þ
where Cp and Cv are the specific heats of air at constant pressure and constant
volume, respectively.
The above equation means that a mass of air at pressure 100 kPa and
temperature 30 C will cool down to 20.8 C if it is risen adiabatically to a
height where the pressure is 90 kPa. This temperature drop as height increases
is called adiabatic lapse rate. Its value for dry air is about 0.01 K/m while for
moist air it shows lower values (approx. 0.008 K/m).
5 Air Temperature and Humidity 57
Fig. 5.1 Typical

Night Stable Day Unstable
temperature profiles during
the day and during the night
B• • B’ D’ • •D
HEIGHT
•A •C
Adiabatic Adiabatic
TEMPERATURE
5.3 Basic Principles on Air Humidity
According to Dalton’s Law, the pressure of air in the atmosphere is the sum of the
partial pressure of water vapor (ea) and the partial pressure of dry air (Pd). Both
partial pressures may be expressed as the product of their respective molar fractions
by the total pressure (P).
From a water surface some molecules are escaping in the evaporation process
while some others return to the liquid state. When the number of molecules
escaping equals that of those returning, the system has reached a steady state and
the atmosphere is said to be saturated. In this state, the vapor pressure has reached
its saturation value (es). For a given temperature there is a single value of es, and the
relation between the two variables is an exponential function:

17:27T
es ¼ 0:61078exp ð5:2Þ
237:3 þ T
where es is expressed in kPa and T is temperature ( C). The atmosphere is usually

not saturated, thus, ea is lower than es for that temperature.
Relative humidity (%) indicates the degree of saturation as:
ea
RH ¼ 100 ð5:3Þ
es
However, as it depends on the temperature, it is not a good indicator of the amount

of water vapor in the air.
Another interesting feature of air humidity, widely used in crop ecology, is the
vapor pressure deficit (VPD) which is the difference es-ea. Its value gives an idea of
the drying power of the atmosphere and is therefore a key factor determining the
rate of evaporation and transpiration.
Air humidity may also be quantified by the dew point temperature (Td) which is
the temperature required for a portion of air at constant pressure and constant water
vapor content, to reach saturation. If air temperature falls below Td, condensation
will start. This is the phenomenon that can be seen on cold mornings when a film of
water covers the soil or plants.
The mixing ratio is the ratio of mass of water vapor per unit mass of dry air and
can be calculated (in g/kg) as:
ea
Xv ¼ 622 ð5:4Þ
Pat ea
where Pat is atmospheric pressure (kPa) and ea is vapor pressure (kPa). Finally, air
humidity may be expressed as vapor density (ρv), also known as absolute humidity,
which is the mass of water vapor per unit volume of air, and can be calculated
(g water vapor m3) as:
1000 ea
ρv ¼ ð5:5Þ
0:4615 T
where T is air temperature (K).
Example 5.1 Air temperature is 20 C and relative humidity is 80 %. We

will calculate vapor pressure, VPD, mixing ratio, vapor density and dew point
temperature. We will assume standard atmospheric pressure (101.3 kPa).
First we calculate saturation vapor pressure :

es ¼ 0:61078 exp½17:27 20=ð20 þ 237:3Þ ¼ 2:338 kPa
Therefore: ea ¼ ðHR=100Þes ¼ 80=100 2:338 ¼ 1:87 kPa
VPD ¼ es ea ¼ 2:338 1:87 ¼ 0:467 kPa
Mixing ratio: Xv ¼ 622 1:87=ð101:3 1:87Þ ¼ 11:7 g=kg
Vapor density: ρv ¼ 1000 1:87=ð0:4615 293Þ ¼ 13:83 g m3
Dew point temperature may be deduced by equating Eq. 5.2 to 1.87 kPa, which
is the actual vapor pressure, and then solving for temperature as:
ea

237:3ln 0:6108
Td ¼ ¼ 16:44∘ C
17:27 ln 0:6108
ea
If this air is cooled down to 16.4 C we would reach saturation. This may be
checked by putting T ¼ 16.44 C into Eq. 5.2, yielding 1.87 kPa.
5.4 Temperature Profiles Within and Above Crop

Canopies
The soil or crop surface undergoes cooling during the night as solar radiation is
zero, while maintaining its emission of long-wave radiation (Rn negative). The air
in contact with the surface transfers heat to the surface, and thus also cools and its
sunrise sunset noon afternoon

HEIGHT
Canopy top
AIR TEMPERATURE
Fig. 5.2 Typical temperature profiles above and within a crop canopy
density increases. The consequence is that a temperature inversion develops and the
atmospheric condition is stable (Fig. 5.2).
During the day the opposite occurs. The surface absorbs radiation that, in part,
serves to heat the lower layers of the atmosphere. The temperature now decreases
with height and the atmospheric condition is unstable (this can be clearly seen in the
afternoon in Fig. 5.2).
The shape of the temperature profiles has important implications for the tempo-
ral and spatial distribution of temperature on crops. As we approach the crop
surface, thermal oscillation (the difference between maximum and minimum tem-
perature) increases as higher maximum and lower minimum temperatures will be
observed.
Also during the day, inverted profiles can be observed by cooling as a result of
crop evapotranspiration. This situation is typical of summer when hot dry air blows
over irrigated crops. The phenomenon is called sensible heat advection and is
explored in more detail in Chap. 7.
Both the day and night profiles, are often highly variable and are affected by
other factors, particularly wind speed. When wind speed is very high, the temper-
ature profile above the crop is very uniform with height, so that the air and the crop
temperature are similar. In contrast, during low wind the temperature profile is very
sharp and the crop is much hotter than the air (during the day) or much colder
(overnight). The temperature profiles (stable or unstable) have an important effect
on turbulence (Box 5.2).
The temperature profiles within the crop canopy are quite different from those
observed above the canopy. During the day, in general, temperature reaches a
maximum at the level where leaf density is highest. A high leaf area density allows
greater radiation absorption and therefore higher temperature. Above this level the
daytime temperature profile is generally unstable and below that height it is usually
slightly inverted. During the night, the profile within the canopy is usually isother-
mal as the crop traps and re-emits radiation emitted from the soil (Fig. 5.3).
5.0
4.5
4.0
3.5
HEIGHT (m)
3.0
2.5
2.0
1.5 PREDAWN
AFTERNOON
1.0
0.5
0.0
–0.5 0.0 0.5 1.0 1.5 2.0 2.5
TEMPERATURE DEVIATION (K)
Fig. 5.3 Air temperature profiles above and within an olive canopy of 3.25 m height on a summer
day (July 30, 2011) near Cordoba (Spain). The values are shown as departures from the temper-
ature measured at 3.0 m height
Box 5.2. Effects of Atmospheric Stability on Turbulence

The mathematical treatment of stability effects lies outside the scope of this
book. However, some remarks should be made on the importance of stability.
During the daytime, turbulent transport is insured as wind speed is higher. If
wind speed is low, buoyancy increases and turbulence will be sustained.
Therefore only in rare cases the lack of turbulence in the daytime may restrict
the vertical transport of energy or matter. The opposite will occur at night as
wind speed is usually lower and the temperature profile is inverted. That
creates a layer above the crop where turbulence is almost absent so H and LE
are small and canopy cooling follows the loss of long wave radiation.
5.5 Air Humidity Profiles Above Crop Canopies
During the day, because of transpiration, the vapor pressure is high near the crop
surface and decreases with height. Water vapor is removed more effectively with
increasing wind and turbulent transport. Therefore, the more pronounced profiles
occur at noon when evapotranspiration is high and the water vapor is easily
removed. The vapor pressure profile at night is much more uniform, and may
even increase with height in nights when dew deposition occurs
5.6 Time Course of Air Temperature and Humidity
The time course of air temperature along a clear day follows a sine function with the
minimum around sunrise and a maximum that occurs 2–3 h after the peak of
radiation (Fig. 5.4). The delay of air temperature relative to radiation is due to the
balance between the energy reaching the surface and the energy being used. Part of
the radiation in the morning is spent in heating the soil and the crop. Once these
surfaces have been heated, there will be sensible heat transfer to the air that will
then be heated. Furthermore, other factors (such as advection) can contribute to
raising air temperature in the afternoon.
Vapor pressure varies relatively little in comparison with other variables related
to humidity. Therefore variations in VPD along the day are mainly due to variations
in air temperature. In any case, vapor pressure close to the canopy is proportional to
evaporation rate, so it will be higher during midday (Fig. 5.5).
In contrast with the vapor pressure, the maximum relative humidity occurs
during the night because, although the water vapor content of the atmosphere is
somewhat lower, the temperature is much lower. The minimum relative humidity
occurs sometime after noon because although the vapor pressure can be high then,
the temperature will also be high.
To calculate the average daily vapor pressure as a function of maximum and
minimum RH we can use the following equation:

RH n RH x
eavg ¼ 0:5 esx þ esn ð5:6Þ
100 100
45
40 19-Jul
11-Dec
35
AIR TEMPERATURE (°C)
30
25
20
15
10
5
0
–5
0 2 4 6 8 10 12 14 16 18 20 22 24
TIME GMT
Fig. 5.4 Time course of air temperature throughout the day on a summer (19 July 2012) and a late
fall day (11 December 2012) in Espiel (Spain)
3.5
ea
3.0
VPD
VAPOR PRESSURE (kPa)
2.5
2.0
1.5
1.0
0.5
0.0
0 2 4 6 8 10 12 14 16 18 20 22 24
TIME GMT
Fig. 5.5 Time course of vapor pressure and vapor pressure deficit throughout the day on a spring
day (5 May 2013) in Cordoba (Spain)
where RHn and RHx are minimum and maximum RH, respectively and esx and esn
are saturated vapor pressure for the maximum and minimum temperature, respec-
tively. Therefore the average VPD will be:

RH n RH x
VPDavg ¼ 0:5 esx 1 þ esn 1 ð5:7Þ
100 100
The annual time courses of the maximum and minimum temperatures of the air
follow a pattern very similar to the daily curve (Fig. 5.6). Something similar occurs
for vapor pressure.
Figure 5.7 shows the mean monthly vapor pressure in two locations with the
same latitude (31.8 S) but with low (Klawer, South Africa) and high rainfall
(Pelotas, Brazil). In both locations vapor pressure is maximum in summer and
minimum in winter, but the two curves reflect clearly the difference in rainfall.
On average, the minimum and the maximum temperature occur with a certain
delay with respect to the minimum and maximum radiation, respectively. This is
seen, for example, in Fig. 5.8, that shows the annual curves of average radiation and
temperature in Cordoba, normalized with respect to their extreme values. The
reasons for the delay between the temperature and radiation waves are similar to
those described for the daily curve. During the spring part of the radiation is used to
heat the soil. As the soil temperature rises above the air temperature, more energy
will be available to be converted into sensible heat. A similar reasoning can be
applied to explain when the minimum winter temperature occurs.
40
Tmax
35
Tavg
30 Tmin
AIR TEMPERATURE (°C)
25
20
15
10
0
0 30 60 90 120 150 180 210 240 270 300 330 360
DAY OF YEAR
Fig. 5.6 Time course of average maximum, minimum and mean air temperature. Cordoba
(Spain). 1964–2002
2.2
KLAVER
2.0 PELOTAS
1.8
1.6
1.4
1.2
1.0
0.8
1 2 3 4 5 6 7 8 9 10 11 12
MONTH NUMBER
Fig. 5.7 Time course of mean monthly vapor pressure for two locations, Pelotas (Brazil) with
mean annual rainfall of 1395 mm and Klawer (South Africa) with mean annual rainfall 174 mm
The amplitudes of the waves of air temperature, both for daily and annual cycles,
depend largely on the partitioning of energy between evaporation and heating of the
soil and air above as we will see in following sections. So in arid areas the
amplitude will be higher than in wet areas (or near the sea).
30
25
RADIATION (MJ m–2 day–1)
TEMPERATURE (°C)
20
or
15
10
5 TEMPERATURE
RADIATION
0
0 30 60 90 120 150 180 210 240 270 300 330 360
DAY OF YEAR
Fig. 5.8 Time course of mean air temperature and solar radiation. Cordoba (Spain). 1964–2002
50
45
40
MAX TEMPERATURE (°C)
35
30
25
20
15
10
0
–80 –60 –40 –20 0 20 40 60 80
LATITUDE (degrees)
Fig. 5.9 Mean maximum temperature in July (Northern locations) or January (Southern loca-
tions) as a function of latitude. Only locations with altitude lower than 1000 m have been included
The maximum temperatures expected in summer are typically between 30 and

45 C with larger values occurring at mid latitudes (Fig. 5.9). Minimum winter
values are around 20 C for zero latitude and decrease as we move North or South
(Fig. 5.10). The variation of mean annual vapor pressure parallels that of minimum
temperature in winter (Fig. 5.11).
40
30
20
MIN TEMPERATURE (°C)
10
–10
–20
–30
–40
–50
–60
–80 –60 –40 –20 0 20 40 60 80
LATITUDE (degrees)
Fig. 5.10 Mean minimum temperature in February (Northern locations) or August (Southern
locations) as a function of latitude. Only locations with altitude lower than 1000 m have been
included
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
–80 –60 –40 –20 0 20 40 60 80
LATITUDE (degrees)
Fig. 5.11 Mean annual vapor pressure for different locations as a function of latitude
5.7 Sensible Heat Flux
One consequence of the temperature gradients that occur over canopies will be a
sensible heat flux (convective heat transfer) from the layers of higher temperature to
those of lower temperature. Normally, the flux is directed towards the crop at night
and from the crop to the atmosphere during the day. The term “sensible” means that
this flux has an effect that can be detected as it involves changes in air temperature.
The sensible heat transfer is mainly controlled by turbulence. The sensible heat
flux (H) between the canopy and the atmosphere at height z where temperature is
Ta, may be calculated as:
Tc Ta
H ¼ ρ Cp ð5:8Þ
r aH
where ρ is air density, Cp is specific heat of air at constant pressure, Tc is canopy

temperature and raH is aerodynamic resistance for heat transport (s m1).
The product ρ Cp (units J K1 m3) can be calculated as:

Xv 29 103 Pat 0:622 ea
ρCp ¼ ρa 1:01 þ 1:88 ¼ 1:01 þ 1:88 ð5:9Þ
1000 8:31T Pat ea
where ρa is the density of dry air (g m3), Xv is the mixing ratio (g kg1), ea is vapor
pressure (kPa) and Pat (in kPa) is atmospheric pressure which can be calculated as a
function of altitude (AL, m) as:

AL 5:2568
Pat ¼ 101:3 1 ð5:10Þ
44308
Calculation of raH is not straightforward as it depends on atmospheric stability.

For neutral conditions, the following equation may be applied:

zdz
ln z0 ln zd
z0H
z
raH ¼ ð5:11Þ
k2k U ðzÞ
where zoH is roughness length for heat exchange which may be estimated as 0.2 zo.
5.8 Latent Heat Flux
Water vapor flows normally from the surface of the soil or the crop to the air above
the crop. This involves a latent heat flux required for the evaporation of water. A
downward water vapor flow and therefore a negative latent heat flux can be
observed in cases of deposition of dew. Similar to the fluxes of momentum or
sensible heat, the latent heat flux (LE) is expressed as:
ρ Cp esc ea
LE ¼ ð5:12Þ
γ r aW þ r c
where γ is the psychrometric constant (around 0.067 kPa/K), esc is saturation vapor
pressure at canopy temperature, rc is canopy resistance and raW is aerodynamic
resistance to water vapor flow, which should be equal to aerodynamic resistance for
heat transport (Eq. 5.7). Note than in writing this equation we are considering the
gradient of vapor pressure between the inside of the leaves, namely in the
substomatal cavities (where air is saturated and at the same temperature as the
crop) and the air above. Therefore water vapor transfer will be subjected to two
resistances in series, one due to the stomata and the second due to aerodynamic
conditions. The canopy resistance is then a parameter reflecting the degree of
stomatal closure and is discussed in more detail in Chap. 9. Typical values of rc
are 40–80 s/m for well watered annual crops and 100–200 s/m for forests and some
fruit crops (e.g. olives) with good water supply. In water deficit situations, rc can
reach much higher values than those indicated above.
Bibliography

New York: Springer.
Rosenberg, N. J., Blad, B. L., & Verma, S. B. (1983). Microclimate: The biological environment
(2nd ed.). New York: Wiley.
Chapter 6
Soil Temperature and Soil Heat Flux
Francisco J. Villalobos, Luca Testi, Luciano Mateos, and Elias Fereres
Abstract The soil temperature regime depends on its thermal properties (specific
heat, thermal conductivity, thermal diffusivity and thermal admittance). The main
factors affecting the thermal regime are water content, soil texture and compaction.
The rate of heating (or cooling) of the soil is proportional to its diffusivity which is
higher in sandy soils. The amount of energy stored in the soil is proportional to its
thermal admittance. Soil heating occurs as a wave train with the amplitude decreas-
ing with depth and a phase shift that also increases with depth.
6.1 Introduction
The heat flux in the soil is an important component of the energy balance of the
crop. The soil acts as a large energy accumulator that stores heat during the day and
releases it at night. Something similar happens in annual terms. The balance of
these exchanges determines the time course of soil temperature. Soil temperature is
important in many crop growth and development processes such as seed germina-
tion, root growth and distribution in the soil, nutrient uptake, root respiration,
microbial activity, etc.

L. Testi • L. Mateos

Agriculture, DOI 10.1007/978-3-319-46116-8_6
6.2 Soil Thermal Properties
The specific heat of the soil is the heat required to raise 1 K the temperature of the unit
mass (specific heat per unit mass, CM) or unit volume (specific heat per unit volume,
CV) of soil. CM and CV are related through the actual soil density (CV ¼ ρ CM). Using
the definitions of bulk density (ρb ¼ mass of solids/volume) and gravimetric moisture
content (θg ¼ mass of water/mass of solids), the relationship between CV and CM can
also be expressed as:

CV ¼ ρb 1 þ θg CM ð6:1Þ
CV can be computed as the sum of specific heats of the soil components (air,
water, solid fraction), weighted by the volumetric mass densities of each of these
components:
Msolid Mwater Mair

CV ¼ CM solid þ CM water þ CM air ð6:2Þ
V V V
where M and CM refer to mass and specific heat per unit mass, respectively, for each
component. Neglecting the third term due to the small specific heat of air, assuming
that CM solid ¼ 0.85 J/g/K and reorganizing:

CV ¼ ρb CM solid þ θg CM water ¼ ρb 0:85 þ 4:18 θg
¼ 0:85 ρb þ 4:18 θv ð6:3Þ
This equation shows that soil specific heat per unit volume increases linearly
with volumetric water content (θv) and with soil bulk density.
6.3 Soil Heat Flux
Applying Fourier’s law of heat conduction, the flux of heat in the soil (J ) can be
expressed as:
dT
J ¼ k ð6:4Þ
dz
where dT/dz is the gradient of temperature and k the thermal conductivity of the
soil. When referring to J at the soil surface (z ¼ 0), it is denoted G.
The thermal conductivity has units of W/m/K and depends on the porosity of the
soil, on water content and organic matter content. The dependence of k on soil water
content is complex. The thermal conductivity of a very dry soil increases twofold
when a rather small amount of water is added. This is because relatively large
amounts of energy can be transferred by evaporation and condensation of water in
6 Soil Temperature and Soil Heat Flux 71
the pores of the soil. For example, for a sandy soil, k can increase from 0.53 to
1.1 W/m/K when its water content increases from the permanent wilting point
(PWP) to the field capacity. A further increase of the soil water content to saturation
implies a smaller gain in k, since the water vapor diffusion is restricted as pores are
filled with water (Table 6.1). Therefore, the thermal conductivity of wet soils is
affected very little by water content variations in that range.
The change with time of the stored heat in a soil layer of thickness Δz has to be
equal to the difference between the heat flow going in, J(z), and the heat flow going
out, J(zþΔz) so, using Eq. 6.4 we may write:

∂T ∂J ∂ ∂T
ρCM Δz ¼ Δz ¼ Δz k ð6:5Þ
∂t ∂z ∂z ∂z
and assuming that k does not vary with depth, we obtain:

2
∂T ∂ T
¼D ð6:6Þ
∂t ∂z2
where D is thermal diffusivity (m2/s), defined as k/Cv. According to Eq. 6.6 thermal
diffusivity characterizes how fast a soil warms or cools.
By adding water to a very dry soil, k initially increases faster than CV so that
D also increases with the water content. However, in a wet soil, k increases more
slowly than CV, so that D becomes constant or even may decrease. As shown in
Table 6.1 the diffusivity of a sandy soil is higher than that of a clay soil especially
when wet. In addition it should be noted that the diffusivity varies little with
changes in water content in the clay soil and more in the sandy soil.
6.4 Spatial and Temporal Variations of Soil Temperature
When a soil is exposed to solar radiation, part of it is reflected and the rest absorbed,
increasing its temperature during part of the daytime. The temperature will then
decrease during the night, when no energy input is present and the heat is trans-
ferred to the rest of the soil and/or is irradiated into the atmosphere. Soils are
subjected to cycling intensity of input energy at daily or annual frequency.
To analyze the cyclic behavior of soil temperature we will use T(z,t) to denote
temperature at depth z and time t. Then we will assume that the temperature of the
soil surface follows a sine function:
T ð0; tÞ ¼ T m þ Að0Þ sin ðω tÞ ð6:7Þ
where Tm is the time-averaged temperature of the surface, A(0) is the amplitude of

the temperature wave at the surface, ω ¼ 2π/Po and Po is the period of the
72
Table 6.1 Thermal properties of soils and soil components (based on Monteith, 1973). We have assumed bulk densities of 1.5, 1.4 and 1.3 t/m3 for sandy,
loam and clay soil, respectively. For each soil type the table shows the thermal properties at soil water content of wilting point, field capacity and saturation
Component Density Specific heat Thermal Thermal Thermal
Unit mass Unit volume Conductivity Difussivity Admittance
Components t/m3 MJ/t/K MJ/m3/K W/m/K 106m2/s J/K/m2/s0.5
Air at 20 C 0.0012 1.01 0.001212 0.025 20.63 5.5
Water 1 4.18 4.18 0.57 0.14 1544
Soil minerals 2.65 0.87 2.31 2.5 1.08 2403
Quartz 2.66 0.8 2.13 8.8 4.14 4327
Clay 2.65 0.9 2.39 2.92 1.22 2639
Granite 2.64 0.82 2.16 3 1.39 2545
Organic matter 1.3 1.92 2.5 0.25 0.1 790
Soils
Sandy PWP 1.53 0.93 1.43 0.53 0.37 871
Sandy FC 1.59 1.06 1.68 1.1 0.65 1359
Sandy SAT 1.87 1.52 2.85 1.8 0.63 2265
Loam PWP 1.52 1.13 1.72 0.65 0.38 1057
Loam FC 1.67 1.41 2.35 0.8 0.34 1371
Loam SAT 1.84 1.66 3.06 1.1 0.36 1835
Clay PWP 1.57 1.44 2.26 0.7 0.31 1258
Clay FC 1.7 1.65 2.8 0.8 0.29 1497
Clay SAT 1.77 1.8 3.18 0.9 0.28 1692
F.J. Villalobos et al.
temperature oscillations (86,400 s for daily cycles and 31,536,000 for annual
cycles). With this boundary condition, we can integrate the differential Eq. 6.6
for T:
Tðz, tÞ ¼ T m þ Að0Þ ez=M sin ðωt z=MÞ ð6:8Þ
where M ¼ (2D/ω)0.5 (m) is called the damping depth which determines how much
the amplitude of the wave is attenuated with depth and how much the phase is
shifted in time.
The above equation indicates that the amplitude of the temperature wave decays
exponentially with depth in the soil. Furthermore, the wave is shifted in proportion
to depth, i.e. the maximum temperature occurs later as the depth increases. Both
deductions are reflected in Fig. 6.1 that represents the daily time course of the
measured soil temperature at various depths. The annual trend shows a similar
pattern (Fig. 6.2).
The depth at which most of the heat exchange occurs in the soil is called
effective depth and is equal to √2 M. The heat flux density at the soil surface may
be deduced by differentiating Eq. 6.8, applying Eq. 6.4 and setting z ¼ 0, which
leads to:
pffiffiffi k
G¼ 2 Að0Þ sin ðωt þ π=4Þ ð6:9Þ
M
This equation indicates that the temperature wave on the surface (Eq. 6.7) and the
heat flow are offset by π/4, i.e. the maximum temperature occurs 3 h after the
50
45
SOIL TEMPERATURE (°C)
40
35
30
0.025 m
25 0.050 m
0.100 m
20
0 2 4 6 8 10 12 14 16 18 20 22 24
TIME GMT
Fig. 6.1 Daily time course of soil temperature at various depths (0.025, 0.050 and 0.10 m) in an
olive orchard. Cordoba (Spain). July 19, 2004
40
35 0.1 m
0.2 m
30
0.8 m
25
20
15
10
0
0 30 60 90 120 150 180 210 240 270 300 330 360
DAY OF YEAR
Fig. 6.2 Annual time course of soil temperature at various depths (0.10, 0.20 and 0.80 m) in an
olive orchard. Cordoba (Spain). 2003
maximum G (daily cycle) or 46 days (annual cycle). If the maximum of G occurs at

the time of maximum radiation, (1200 h, solar time) then the surface temperature
should reach its maximum at 1500 h (solar time) for daily cycles. For annual cycles
the maximum temperature should occur on day 216 (August 5) in the Northern
Hemisphere and on day 36 (February 5) in the Southern Hemisphere.
Soil heat flux at the surface may be integrated over a half-cycle to determine the
total heat input into the soil:
Z rffiffiffiffiffiffiffiffi
pffiffiffi 2P0 pffiffiffiffiffiffiffiffiffi
G ¼ 2 M Að0Þ Cv ¼ k Cv Að0Þ ð6:10Þ
π
This total flux that enters the soil in one half-cycle will be equal to that going out in the
other half-cycle since we start from a sinusoidal model of temperature, i.e. the
temperature at the end of the period is equal to the initial temperature. According to
Eq. 6.10 the amount of heat stored in the soil (and released by the soil) will
be proportional to (k CV)0.5, which is called the thermal admittance and has the units
of J K1 m2 s0.5. For sandy soils thermal admittance increases dramatically with
water content from values of less than 1000 at PWP to more than 2000 J K1 m2 s0.5
at saturation (Table 6.1). In clay soils admittance changes little with water content.
The amplitude of the temperature wave decreases with soil texture in the order
sandy-loam-clay. This order is due to differences in thermal admittance. In Med-
iterranean climate conditions, in late winter, soils are often wet. The sandy soils
have a higher diffusivity than the clay soils, so they will warm up more rapidly in
0.0
–0.1
–0.2
SOIL DEPTH (m)
–0.3
–0.4
–0.5
SUNRISE
–0.6
AFTERNOON
–0.7 SUNSET
–0.8
20 25 30 35 40 45 50 55
Fig. 6.3 Profiles of soil temperature in an exposed area of an olive orchard. Cordoba (Spain)
3 August 2002
spring. If the water content is high in autumn, sandy soils will also cool faster,
which has agronomic implications in the decision about planting date. Conversely if
the soil is very dry the diffusivity of sandy and clay soils is similar, so we should not
expect significant differences in their thermal regime.
The temperature profile in the soil changes significantly over the day (Fig. 6.3).
In the early morning, the soil surface is the coldest zone and in the afternoon it
becomes the hottest. The profile along most of the day indicates downward heat flux
while at night the flow is towards the surface.
6.5 Effects of Evaporation and Wind on Soil Temperature
The prediction of soil temperature at a given time and depth is important at the time
of making some agronomic decisions. The practical importance of understanding
the soil heat flux will become more evident when we address the energy balance at
the earth surface in Chap. 7.
For a dry soil surface with no evaporation the ratio of soil heating and atmo-
spheric heating may be written as:
pffiffiffiffiffiffiffiffiffi
G k Cv
¼ ð6:11Þ
H μatm
where μatm is the atmospheric admittance which increases with wind speed and may
go from 2000 for a calm atmosphere to 10,000 J K1 m2 s0.5 for windy
conditions. This implies that a very dry soil will show ratios of G/H between 0.5 and
0.1 as wind speed increases.
The thermal regime in a soil may be evaluated using Eq. 6.8 which requires
knowledge of the amplitude at the soil surface. Using Eq. 6.10 we may write:
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Z
u
u
u G dt
t π
Að0Þ ¼ pffiffiffiffiffiffiffiffiffi ð6:12Þ
2 P0 k Cv
The integral of G during the daytime depends on wind speed (Eq. 6.11) and on the
amount of energy spent in soil evaporation. For a wet soil, evaporation may take a
large fraction of net radiation (say 70–80 %) while for a very dry soil evaporation
may be negligible. Table 6.2 shows an example of calculated values of surface
temperature amplitude for wet or dry clay and sandy soils in clear winter and
summer days in Cordoba (Spain). The amplitude is large when the soil is dry and
for calm conditions. The sandy soil presents always a larger oscillation than the clay
soil due to the smaller admittance. Equation 6.12 may be used to calculate the
expected minimum and maximum soil temperatures.
Table 6.2 Calculated temperature amplitude at the soil surface for winter and summer days in
sandy and clay soils
Conditions Soil surface Soil type G integral Admittance T amplitude
MJ/m2 J/K/m2/s0.5 K
Winter Calm Dry Sandy 2 870 9.8
Windy Dry Sandy 0.55 870 2.7
Calm Wet Sandy 0.4 1300 1.3
Windy Wet Sandy 0.1 1300 0.3
Winter Calm Dry Clay 2 1200 7.1
Windy Dry Clay 0.55 1200 2.0
Calm Wet Clay 0.4 1500 1.1
Windy Wet Clay 0.1 1500 0.3
Summer Calm Dry Sandy 4.7 870 23.0
Windy Dry Sandy 1.3 870 6.4
Calm Wet Sandy 0.9 1300 3.0
Windy Wet Sandy 0.3 1300 1.0
Summer Calm Dry Clay 4.7 1200 16.7
Windy Dry Clay 1.3 1200 4.6
Calm Wet Clay 0.9 1500 2.6
Windy Wet Clay 0.3 1500 0.9
Bibliography

New York: Springer.
New York: Wiley.
Chapter 7
The Energy Balance
Francisco J. Villalobos, Luca Testi, Luciano Mateos, and Elias Fereres
Abstract The main components of the energy balance are net radiation, latent heat
flux (LE), sensible heat flux (H) and soil heat flux (G). These can be manipulated
through changes in net radiation, LE, H or G. The relative importance of the
components depends mainly on the availability of water for evaporation. The
extreme cases will be the humid environment (LE approaches Rn) and the desert
environment (Rn is partitioned between H and G). The energy balance of farming
(energy produced per unit energy consumed in all farming operations) may be also
analyzed in terms of inputs and outputs which can be estimated by assessing the
energy embodied in the amount of materials employed (fertilizers, water, seeds)
and in the operations performed.
7.1 Introduction
The exchange of matter and energy between the crop and the atmosphere deter-
mines the variations of air temperature and humidity and the soil temperature as
seen in the previous chapters. Turbulence facilitates the flux of heat, water vapor
and carbon dioxide which are all key factors in crop production. It also affects the
transport of contaminants, pesticides, spores or pollen. Understanding the
partitioning of available energy among the different processes is required for
manipulating the aerial or the soil environment of the crop, which not only affect
the plant community but the whole ecosystem (weeds, insects, pathogens, and other
soil microorganisms).

L. Testi • L. Mateos

Agriculture, DOI 10.1007/978-3-319-46116-8_7
7.2 The Energy Balance Equation
According to the principle of conservation of energy, the equation of the energy

balance of a crop may be written as:
Rn ¼ H þ LE þ G þ P þ S ð7:1Þ
where all terms have been defined in previous chapters expect for P, that represents
the flux of energy consumed in photosynthesis, and S, which is the flux of heat
stored in the crop biomass and the surrounding air. As will be seen later, P is
negligible compared to the others. S may be significant only in very tall plant
communities (e.g. forests) but can be neglected in the case of crops (Box 7.1). Note
that in Eq. 7.1 fluxes are considered positive when they involve a loss of energy
from the crop, i.e. when moving away from the canopy.
Changes in net radiation may be accomplished by variations in the short wave or
long wave radiation balances. The former may be due to changing the incoming flux
using shades or due to changes in the soil slope and aspect and also by manipulating
the albedo of the soil (mulches) or the plants (e.g. whitewash or kaolinite). The long
wave radiation balance is usually manipulated by blocking losses with glass or
plastic covers or non solid barriers (e.g. smoke).
Soil heat flux may be changed by artificial soil heating or altering the thermal
admittance by applying irrigation or compacting the soil.
LE is mainly determined by water availability, so rainfall or irrigation manage-
ment will affect greatly LE fluxes. If the soil is wet, important reductions in LE may
be achieved using plastic impermeable mulches (Box 7.2).
The only alternative for manipulating sensible heat flux in the field is using
barriers (windbreaks) to reduce wind speed. In controlled environments (green-
houses) we may add or remove heat or increase turbulence with fans.
Box 7.1 Calculation of Energy Spent in Heating the Canopy

The maximum standing biomass of annual crops usually does not exceed 20 t/
ha (dry matter).
Assuming water content of 70 g water/100 g fresh mass (e.g. maize for
silage) we may calculate the energy term S for an increase in canopy
temperature of 20 K in 9 h. This would be indicative of typical conditions
of summer in mid latitudes.
The water mass can be calculated based on dry biomass (B) as:
t dm 0:7 tt water t water g water

mass water ¼ 20 fresh
¼ 46:67 ¼ 4667
ha ð1 0:7Þ t tfresh
dm ha m2
The specific heat of water (CMw) and organic matter (CMo) are 4.18 and
1.92 J/g/K, respectively.
(continued)
7 The Energy Balance 81
Box 7.1 (continued)

The amount of heat stored for a temperature rise of 20 K is:
ΔQ ¼ ΔTðB CMo þ mw CMw Þ ¼ 20ð2000 1:92 þ 4667 4:18Þ

¼ 466961 J m2 ¼ 0:47 MJ m2
The average heat flux for 9 h will be obtained by dividing 0.47 106 J m2 by
the duration (9 3600 s) which yields 14.4 W m2, which is very small
(about 2 %) as compared to typical values of solar irradiance under those
conditions (600–800 W m2).
Box 7.2 Analyzing the Effect of a Black Plastic on the Energy Balance
of Bare Soil
The plastic sheet creates a barrier that suppresses evaporation (LE ¼ 0) so all
the energy is spent in G and H. A small variation in Rn may be expected as the
albedo is reduced and long wave loss may increase when the plastic gets hot.
The partitioning between H and G will depend on the contact between the
plastic sheet and the soil as any air layer between them will reduce flux into
the soil. On the other hand, aerodynamic resistance is reduced as the temper-
ature profile over the hot plastic is very unstable. Therefore, sensible heat flux
is enhanced.
In summary, covering the soil with black plastic suppresses LE and
reduces the G/H ratio, which promotes a higher temperature of the aerial
environment.
7.3 Relative Importance of the Components of the Energy

Balance
As water availability is the main factor determining LE, two extreme conditions can
be distinguished:
(a) Humid/well watered areas where water availability does not limit evaporation.
Most of the net radiation is used in LE. It is the case of large water bodies (seas,
lakes), wetlands, large irrigation schemes with abundant water or any area after
widespread rainfall.
(b) Dry/arid environments where no water is available: LE is negligible and
therefore net radiation is partitioned only into heating the air and the soil.
700
Rn
600
G
500 LE
H
400
FLUX (W/m2)
300
200
100
–100
0 2 4 6 8 10 12 14 16 18 20 22 24
TIME GMT
Fig. 7.1 Energy balance components over an irrigated cotton crop. Cordoba (Spain). June
27, 2003. For the 24-h period the total fluxes were Rn ¼ 15.9, G ¼ 0.2, LE ¼ 15.7 and
H ¼ 0.6 MJ m2 day1
A special case is that of the oasis, a small well watered area surrounded by arid
lands. Here horizontal transport of sensible heat from the arid surrounding area
enhances LE in the oasis, so it may exceed Rn. This process of horizontal transport
of energy from dry to wet areas is called advection.
The energy balance of crops is usually between the two extremes (wet-desert)
depending on the availability of water and the presence of vegetation (this issue will
be further explored in Chap. 9). Four different situations are analyzed below.
Figure 7.1 represents the daily time course of Rn, LE, H and G on an irrigated
cotton field measured on June 2003 in Cordoba (Spain). This would be a typical
well watered environment. The Rn presents typical values of clear days at this time
of the year in Cordoba. Most of the energy is spent in LE which is less than Rn
during the morning and higher during the afternoon (advection). The sensible heat
flux is small and reaches its maximum during the morning. From 14:00 the H flux is
reversed, being then directed towards the surface. The soil is heated for most of the
daytime, although G is very small as the crop covers the soil almost completely.
The balance for 24 h indicates that net radiation was invested mostly in evaporation.
In this case the heating of the air and the soil in the daytime period is offset by their
cooling during the night.
A case of water stressed crops is represented in Fig. 7.2 for a wheat crop in spring
(around flowering) after a very dry winter in Cordoba (Spain). The soil was very dry
and consequently, crop transpiration was very low. As a result, latent heat flux is
very small during most of the day while H and G use most of the energy during the
600
Rn
500
G
LE
400
H
FLUX (W m–2)
300
200
100
–100
0 2 4 6 8 10 12 14 16 18 20 22 24
TIME GMT
Fig. 7.2 Energy balance components on a severely stressed wheat crop. Cordoba (Spain). April
19, 1999. For the 24-h period the total fluxes were Rn ¼ 11.6, G ¼ 1.2, LE ¼ 1.4 and H ¼ 9 MJ m2
day1
700
Rn
600
G
500 H
FLUX (W m–2)
400
300
200
100
–100
0 2 4 6 8 10 12 14 16 18 20 22 24
TIME GMT
Fig. 7.3 Energy balance components on wheat stubble. Cordoba (Spain). June 17, 1995. For the
24-h period the total fluxes were Rn ¼ 13.7, G ¼ 1.3 and H ¼ 12.4 MJ m2 day1
daytime. For the period of 24 h, the percentages of energy invested in LE, H and G
are 12 %, 78 % and 10 %, respectively.
Contrasting with the well watered crop is the case of wheat stubble shown in
Fig. 7.3. Around the measurement date (June 17), extraterrestrial solar radiation
600
Rn
500
G
LE
400
H
FLUX (W m–2)
300
200
100
–100
0 2 4 6 8 10 12 14 16 18 20 22 24
TIME GMT
Fig. 7.4 Energy balance components in an olive grove. Agricultural Research Center of Cordoba
(Spain). September 8, 1997. For the 24-h period the total fluxes were Rn ¼ 13.1, G ¼ 0.1,
LE ¼ 10.3 and H ¼ 2.9 MJm2 day1
peaks in the northern hemisphere. However, the maximum Rn is slightly lower than
in the case of Fig. 7.1, which is explained by the high albedo of stubble and straw
covering the soil. The availability of water in this case is zero (the crop had
extracted all soil water) which explains the absence of LE. The sensible heat flux
parallels Rn throughout the day, peaking at around noon. The soil heat flux is now
larger than for cotton and decreases during the afternoon. Considering the 24-h
period 90 % of the energy is spent in heating the air and 10 % in heating the soil.
Note that the straw and stubble covering the soil serve as a thermal insulation.
Finally, Fig. 7.4 presents an intermediate case between the wet and dry cases. It
is a drip irrigated olive orchard at Cordoba where the trees cover only a fraction of
the ground. At the time of the measurements the tree canopy represented 40 %
ground cover and the soil surface was dry. In the daytime the three fluxes (LE, H
and G) are important but LE predominates (Fig. 7.4). The H flux peaks at noon and
is small at night. G presents a pattern similar to the above cases: the maximum
occurs at noon, because at that time the percentage of soil exposed to direct
radiation is maximum. For the period of 24 h the percentages of energy invested
in LE and H are 79 % and 21 %, respectively, while G gets no share.
We have seen that the relative importance of the components of the energy
balance varies throughout the day and that when summed for 24-h periods, G is
usually small compared to H and LE, whose relative magnitudes depend on the
availability of water (or presence of vegetation) on the soil surface. The impact of
the fluxes on the oscillations of air and soil temperature depends on the absolute
value of H and G. The difference between the maximum and the minimum
temperature will be greater when positive fluxes are higher and negative fluxes
are smaller.
Table 7.1 Annual energy (MJ/m2) available at different levels and losses (partitioning of energy)
at each stage for an irrigated olive orchard with 40 % ground cover in Cordoba (Spain)
Available energy Losses
MJ/m2 MJ/m2
Reflection Emission
Solar radiation 5975 1100 2065
LE H
Net radiation 2810 1570 1124
Respiration
Energy photosynthesis 40 20
Vegetative
Energy fixed 20 8
Residues
Energy harvested 12 5
Energy in olive oil 7
The partitioning of energy for annual periods is also governed by the availability
of water. Table 7.1 shows the values of energy available at different levels and the
losses (partitioning of energy) at each stage for an irrigated olive orchard with 40 %
ground cover in Cordoba (Spain). Starting from an incoming solar radiation of around
6000 MJ m2, the energy produced by the crop as oil is equivalent to 7 MJ m2 as oil
(2000 kg oil ha1). About 50 % of incoming radiation is lost by reflection and
emission of long wave radiation. Then net radiation is allocated to evaporation
(56 %), heating of the atmosphere (40 %) and photosynthesis (only 4 %). Note that
the soil heating component is zero for annual periods as long as the mean soil
temperature does not change in the long run. Around 50 % of energy converted by
photosynthesis is lost as respiration, leaving another 50 % fixed in tree biomass
(shoots, roots and fruits). Only 60 % of fixed energy is harvested. After extraction
of the oil, the energy captured is just 7 MJ m2. This low apparent efficiency does not
take into account other energy inputs required for the production of olives (machin-
ery, fertilizers, pumping of irrigation water). In the same environment, an intensive,
irrigated wheat-maize rotation in about the same period (as a double crop, maize
planted after wheat is harvested) could yield up to 19,000 kg ha1of grain total, which
would be equivalent to about 26 MJ m2, still a very small fraction of the incoming
solar radiation of 6000 MJ m2.
7.4 Energy Requirements in Farming
The energy balance of a field may be also analyzed in terms of the energy
requirements associated to the different agricultural practices and the final energy
stored in crop products. Energy requirements may be direct (fuel used, energy spent
in pumping water) or indirect (energy spent in production of inputs, machinery or
Table 7.2 Classification of types of energy requirements for agricultural operations

Direct Indirect
Operation Fixed Variable Fixed Variable
Sowing Fuel Machine, Seed, pesticide,
tractor fertilizer
Tillage Fuel Tools, tractor
Irrigation Fuel or electric energy for Irrigation Desalination,
pumping system transport
Pesticide Fuel Machine, Pesticide
application tractor
Harvest Fuel Machine
Processing Fuel or electric energy Machine
equipment). Each category may be also divided in a variable component (being

proportional to input rate) and a fixed component (not dependent on input rate).
Therefore:
Ereq ¼ Edir fix þ Edir var þ Eind fix þ Eind var ð7:2Þ
Table 7.2 summarizes the types of energy inputs associated to different farm
operations. For instance, the application of a pesticide requires a fixed direct
input of energy to run the tractor, a fixed indirect input associated to tractor and
machine manufacturing and maintenance and a variable indirect input, that of
manufacturing the pesticide, that depends on the dose applied.
Table 7.3 shows some energy coefficients which are useful for calculating the
energy requirements of the farm. In Table 7.4 representative values are given for
different operations including both direct and indirect energy requirements. The
actual values for specific operations will be discussed in more detail in Chap. 17.
The energy requirement (MJ ha1 year1) for irrigation may be calculated as:
I
Eirrig ¼ ρwater g 106 ðHlif t þ 1:2 H op Þ þ Eind f ix þ Eind var ð7:3Þ
μp μm
where:
ρwater: density of water (103 kg m3)
g: acceleration of gravity (9.81 m s2)
106 is used to convert from J to MJ
I: seasonal applied irrigation (m3 ha1)
μp: pump efficiency (typically between 0.6 and 0.8)
μm: motor efficiency, which we may assume 0.4 for diesel and 0.9 for electric
engines
Hlift: the energy required to lift water from the water source (m). It is roughly equal
to water table depth for ground water and negligible for surface water sources.
Table 7.3 Energy coefficients for different inputs used in crop production
Energy
Input Observations coefficient Units
Human labor Embodied energy 100–1000 MJ/day/
person
Gasoline Energy content 38 MJ/L
Diesel Energy content 39 MJ/L
Ethanol Energy content 22 MJ/L
Coal Energy content 17–30 MJ/kg
Wood Energy content 18–23 MJ/kg
Tractors Manufacture and transport 87 MJ/kg
Implements Manufacture and transport 70 MJ/kg
Pesticides Manufacture and transport 358 MJ/kg
N fertilizer Manufacture and transport 77 MJ/kg N
P fertilizer Manufacture and transport 37 MJ/kg P
K fertilizer Manufacture and transport 17 MJ/kg K
Anhydrous ammonia Manufacture and transport 60 MJ/kg N
Ammonium nitrate Manufacture and transport 85 MJ/kg N
Urea (solid) Manufacture and transport 80 MJ/kg N
Drying grain 6.4–10 MJ/kg water
Transport By truck 6.3 MJ/t/km
Water Desalinization sea water 9.0–13.0 MJ/m3
Water Desalinization brackish 3.6–10 MJ/m3
water
Primary tillage Total 1200 MJ/ha
Secondary tillage Total 300 MJ/ha
Spray pesticide Total (excluding pesticide) 90 MJ/ha
Spread fertilizer Total (excluding fertilizer) 90 MJ/ha
Sowing Total (excluding seeds) 340 MJ/ha
Harvest cereals and Total 1200 MJ/ha
legumes
Harvest tubers and roots Total 2200 MJ/ha
Table 7.4 Energy requirements of different agricultural operations

Total energy (excluding labor)
MJ/ha
Primary tillage 1200
Secondary tillage 300
Spray pesticide 90
Spread fertilizer 90
Sowing 340
Harvest cereals and legumes 1200
Harvest tubers and roots 2200
These values include both direct and indirect energy requirements and are only indicative of the
order of magnitude
Hop: operating pressure of the irrigation method (m). We may use typical values of
10–25 m for drip and 30–40 m for sprinklers. The coefficient 1.2 is based on the
assumption that 20 % additional energy is required to keep enough pressure in
the whole network.
Eind fix: Energy spent in the manufacturing and installation of the irrigation system
divided by its life span. We may use values of 7000–9000 MJ ha1 year1 for
sprinkler and 13,000 MJ ha1 year1 for drip irrigation. For surface irrigation we
should add here the energy required for land levelling and for shaping the ridges.
Eind var: Energy spent in desalination (if needed) and in delivering the water to the
farm, which is proportional to the amount of irrigation applied:
Eind var ¼ cw I ð7:4Þ
where cw is the energy spent per volume unit of water. Therefore, we can calculate
the energy requirement for irrigation as the sum of two terms, one variable and one
fixed:
!
Hlift þ 1:2 H op
Eirrig ¼ þ cw I þ Eind fix ð7:5Þ
102μp μm
The energy requirement of fertilizer application (MJ ha1 year1) is calculated

as:
Efert ¼ nf A þ N a cN þ Pa cP þ K a cK þ Eind ð7:6Þ
where
nf: is the number of applications of fertilizer
A: is the energy consumption as fuel in each application (MJ ha1)
Na, Pa, Ka: amounts of N, P and K (kg ha1 year1) applied
cN, cP, cK: energy required for producing and transporting the fertilizers (MJ/kg N,
MJ/kg P, MJ/kg K)
Eind: Energy spent in the manufacturing, repair and maintenance of the fertilizer
spreader or injector divided by its life span.
The energy spent in human labor is not easy to calculate. An average human
being needs between 10 and 15 MJ day1 of energy as food, but we need to add the
energy required for humans to live which depends strongly on the standards of
living so we may finally arrive to a wide range of values of the so-called embodied
energy (100–1000 MJ day1person1). For calculation purposes, we may take the
minimum value when almost all the operations are performed with human labor and
the maximum when machines are used for all operations.
Example 7.1 The average needs of human labor for manual harvesting of
wheat (yield of 6 t/ha) is around 1000 h/ha, while using just machinery it is
reduced to 2 h/ha. Assuming working days of 8 h, the energy requirements
of manual harvest are 12,500 MJ/ha. If we use a combine, the energy require-
ments would be the sum of a human component (2 h/8 h/day 1000 MJ day1
person1) and a machine component (1200 MJ/ha, Table 7.3), which gives
1450 MJ/ha. Not surprisingly, the energy requirement of manual harvesting is
much higher as humans are not as efficient as machines for performing
physical work.
7.5 Farm Energy Outputs and Efficiency
The energy outputs of a farm may be computed according to different criteria. We

usually consider only the energy contained in all materials exported from the farm,
so the energy content of crop residues is ignored (unless they are sold for feed
outside the farm) and so is the energy stored in soil organic matter. On the other
hand, the energy content of a given material varies according to the use of the
energy. We use values of gross energy which are close to the combustion energy of
the material. Table 7.5 shows typical values of gross energy in crop products and
crop residues. Apart from the composition of dry matter,an important factor in
evaluating the energy content of biomass is its water content. The Heating Value
(HV, MJ/kg fresh biomass) is calculated as:
HV ¼ ð1 wÞHCB 2:45 w ð7:7Þ
where w is the water content of biomass (g water/g fresh biomass).
Table 7.5 Energy content in agricultural products and residues

Energy content (HCB,MJ/kg DM)
Crop Use Harvested product Residues
Cereals Grain 18–19 15.5–18.5
Forage 18–19
Legumes Seed 19–20 18–19
Forage 18–19
Soybean Seed 23.6 19
Cotton Fiber+seed 23.8 19
Oil crops Seed 26–29 18.3–18.8
Sugar beet Root without crown 17 16.7
Tuber and root crops Tubers, roots 17 18–20
Crop species have been grouped whenever possible: Grain cereals (wheat, barley, rye, maize,
millets, sorghum, rice, triticale), forage cereals (maize, sorghum), seed legumes (bean, cowpea,
faba bean, lentil, pea, peanut, chickpea), forage legumes (clover, alfalfa), oil crops (safflower,
rapeseed, sunflower, flax), tuber and root crops (potato, cassava, yam, sweet potato). The total
energy output will be the product of yield and the energy content of the harvested product
Example 7.2 Rainfed wheat is grown as monoculture in a farm producing

(dry matter) 2500 kg grain/ha and 3750 kg straw/ha. The total energy
requirements are 14,779 MJ/ha (3400 for operations, 9613 for fertilizer,
1050 for seed and 716 for pesticides). If both grain and straw are exported
the output is 110,000 MJ/ha and the ratio output/input is 7.4. If only the grain
is exported the output/input ratio is 3.1. In both cases, the energy balance of
farming is positive (more energy produced than consumed).
Bibliography
New York: Wiley.
Chapter 8
The Water Budget
Francisco J. Villalobos, Luciano Mateos, Francisco Orgaz,

and Elias Fereres
Abstract The components of the water balance (infiltration, deep percolation,

evaporation from the soil surface, etc.) determine the amount of water available
to the crop. Water flow in the soil occurs following the gradient of water potential
and can be analyzed by the Richards equation, but there are simpler alternative
methodologies to quantify the water balance components. Deep percolation can be
estimated based on soil properties and water content above Field Capacity,
depending also on soil evaporation and transpiration. Runoff is calculated based
on the Curve Number and the amount of precipitation. For monthly values, effec-
tive rainfall can be calculated by the methods of FAO and SCS.
8.1 Introduction
The functioning of terrestrial ecosystems depends largely on the inputs and outputs
of water, which determine the quantity and quality of water available for life on
earth. The availability of water is considered the main limiting factor of the
productivity of agricultural systems. Of all uses of water diverted by man, agricul-
ture is the main consumer of water, on the average consuming two thirds of total
globally, and in most countries of the arid regions irrigation consumes more than
80 % of the developed water. Therefore, it is essential to understand and quantify
the dynamics of the flows in and out of the agricultural system, that is, to calculate
the water balance components that determine the availability of water for crops and,
where appropriate, to quantify irrigation needs.

L. Mateos • F. Orgaz

Agriculture, DOI 10.1007/978-3-319-46116-8_8
As described in Chap. 2, soils are comprised of solids, liquid, and gas, with
typical fractions of 45–50 % of mineral material, 0–5 % of organic matter, and 50 %
of pore space which allows the flow of water and gases. The porosity (volume
fraction of pore space) is determined by the arrangement of the soil particles, being
low when soil particles are very close together (e.g. compacted soil) and higher
when soils have high organic matter. Typical values of porosity are in the range
0.35–0.50 in sandy soils and 0.40–0.60 in medium to fine-textured soils. Porosity
usually decreases with soil depth because the subsoil tends to be more compacted
than the topsoil. Soil bulk density (ρb) is a measure of the mass of soil per unit
volume (solids + pore space) and is usually reported on an oven-dry basis (Chap. 2).
8.2 The Status of Water in the Soil: Water Content

and Water Potential
The water content of the soil can be expressed in terms of volume (θv ¼ volume of
water/volume of soil) or mass (gravimetric, θg ¼ mass of water/mass of dry soil).
Both measurements are related through the soil bulk density so that θ ¼ ρb θg. The
amount of water (expressed in mm) in a soil depth Z (mm), i.e. the total soil water
for that depth (TSW, mm) will be:
TSW ¼ ð1 FVC ÞZ θ ð8:1Þ
where FVC is the volume fraction of coarse fragments (soil particles exceeding
2 mm in diameter) which can be estimated as a function of the mass fraction of
coarse fragments (FMC) as:
FMC ρb
FVC ¼ ð8:2Þ
2:65 FMC ð2:65 ρb Þ
where ρb has units of t m3 and is calculated excluding coarse fragments. Note that
2.65 t m3 is the average density of coarse soil fragments.
Example 8.1 A soil of 1 m depth has 30 % of coarse fragments (mass basis)

and bulk density 1.4 t m3. If the water content is 0.2 m3 m3.
FMC ρb 0:3 1:4

FVC ¼ ¼
2:65 FMC ð2:65 ρb Þ 2:65 0:3 ð2:65 1:4Þ
¼ 0:18 m3 m3
TSW ¼ ð1 FVC ÞZ θ ¼ ð1 0:18Þ1000 0:20 ¼ 164 mm
8 The Water Budget 93
Water is held in the matrix of soil particles by adsorption and moves by

capillarity in the pores. Water is always moving albeit very slowly when the soil
is dry and thus it can be considered retained by the soil for many practical purposes.
There are three water content values that characterize the soil water retention
capacity:
– Permanent wilting point (PWP, also called Lower limit) (θPWP): the soil water
content below which plant roots are unable to extract water.
– Field capacity (FC, also called drained upper Limit) (θFC): It is the value at
which soil water content will stabilize after drainage (drainage continues but at a
rate that may be considered negligible).
– Saturation (θSAT): This is the maximum soil water content that can be observed
in the soil and is on average 85 % of the porosity (η). As the average density of
the soil solid fraction is 2.65 t m3, we can calculate the porosity as a function of
bulk density as η ¼ 1ρb/2.65.
The state of water in the soil can be characterized as a function of its potential
(Ψ), which is the potential energy per unit mass or volume and has units of pressure
(1 J/kg ¼ 1 kPa 0.1 m water column). The soil water potential is the sum of four
components:
– Pressure potential (Ψp) which is the pressure exerted by a column of water above
the point considered. In an unsaturated soil Ψp ¼ 0.
– Gravitational potential (Ψg) is the potential energy of water due to its position in
a gravitational field. It is calculated as g h where g is the gravitational constant
(9.81 m/s2) and h is the height above the arbitrary reference plane.
– Matric potential (Ψm) is caused by the attraction of the soil matrix and the water
molecules. The relationship between matric potential and soil water content, Ψm
¼ f (θv), is called the soil-water characteristic curve (Fig. 8.1). The matric
potential is zero in saturated soil and becomes more negative as the soil dries.
Fig. 8.1 Soil-water

characteristic curves for a
fine and a coarse texture soil
SOIL SUCTION
FINE TEXTURED SOIL
COARSE
TEXTURED
SOIL
WATER CONTENT
Campbell has proposed the following equation for computing the soil-water
characteristic curve:
b
θv
Ψm ¼ Ψe ð8:3Þ
θSAT
where Ψe is called the air entry water potential, b is an empirical parameter and
θSAT is the saturation water content. The values of Ψe and b for different soil
textures are presented in Table 8.1.
– Osmotic potential (Ψo) is due to the presence of salts in the soil solution. This
potential is zero for pure water and becomes more negative as the concentration
of salts increases. An approximate relationship between Ψo (kPa) and the salt
concentration (Cs, g m3) is:
Ψo ¼ 0:05625 Cs ð8:4Þ
Example 8.2 The soil characteristic curve of a Sandy loam soil is given by
Ψm ¼ 0:152 θ3:1 ðkPaÞ
We will calculate water potential for this soil at 1 m depth if the water
content is 0.2 m3 m3 and the salt concentration in the soil solution is 64 g/m3.
We fix the reference level on the soil surface. Therefore:
Ψ ¼ Ψm þ Ψg þ Ψo ¼ 0:152 0:23:1 9:81 1 0:05625 64

¼ 22:3 9:8 3:6 ¼ 35:7 kPa
which is equivalent to 0.036 MPa or 0.36 bar.
Example 8.3 A water table is located at a depth of 1 m in the soil of example

8.2. What would be the water content at the soil surface under conditions of
hydraulic equilibrium if the osmotic potential does not change with depth?
If we take the soil surface as reference level, the water potential at the
water table will be:
Ψ ¼ Ψm þ Ψg þ Ψo ¼ 0 1 9:81 þ Ψo ¼ 9:81 þ Ψo ðkPaÞ
The water potential at the soil surface will have the same value. If osmotic
potential is the same at the reference level and at the water table, the matric
potential at the surface will be:
(continued)
Example 8.3 (continued)

Ψm surface ¼ 9:81 þ Ψo water table Ψo surface ¼ 9:81 kPa
According to the characteristic curve:
Ψm ¼ 0:152 θ3:1
!1=3:1 !1=3:1
0:152 0:152
θ¼ ¼ ¼ 0:26m3 m3
Ψm 9:81
8.3 Water Flow in the Soil
In the simplest case we consider one-dimensional (vertical) water flow in the soil
that obeys Darcy’s law:
dΨ
J w ¼ K ðΨ m Þ ð8:5Þ
dz
where K(Ψm) is the hydraulic conductivity which is a function of matric potential

and Jw is water flow (kg/m2/s). At saturation (zero matric potential) K reaches its
maximum value called the saturated hydraulic conductivity (Ks). The components
of the water potential to be considered are the matric and gravitational potentials.
Then, the equation above is equivalent to:

dΨ m dΨ g dΨ m
J w ¼ K ðΨ m Þ K ðΨ m Þ ¼ K ðΨ m Þ þg ð8:6Þ
dz dz dz
The hydraulic conductivity may be computed as:

2bþ3
θ
K ðθ Þ ¼ K s ð8:7Þ
θs
or
2þ3=b
Ψe
K ðΨ m Þ ¼ K s ð8:8Þ
Ψm
where the parameters have already been defined and given in Table 8.1.
96
Table 8.1 Physical properties of soils with different texture

Silt Clay Ψe b Ks Bulk dens Porosity θPWP θFC θSAT
Texture Fraction J/kg 106 kg s m3 t/m3 Fraction m3/m3 m3/m3 m3/m3
Sand 0.05 0.03 0.7 1.7 5800 1.68 0.37 0.03 0.09 0.23
Loamy sand 0.12 0.07 0.9 2.1 1700 1.64 0.38 0.06 0.13 0.26
Sandy loam 0.25 0.1 1.5 3.1 720 1.56 0.41 0.1 0.21 0.31
Loam 0.4 0.18 1.1 4.5 370 1.43 0.46 0.12 0.27 0.37
Silt loam 0.65 0.15 2.1 4.7 190 1.41 0.47 0.13 0.33 0.40
Sandy clay loam 0.13 0.27 2.8 4 120 1.4 0.47 0.15 0.26 0.37
Clay loam 0.34 0.34 2.6 5.2 64 1.31 0.51 0.2 0.32 0.41
Silty clay loam 0.58 0.33 3.3 6.6 42 1.27 0.52 0.20 0.37 0.45
Sandy clay 0.07 0.4 2.9 6 33 1.33 0.50 0.21 0.34 0.42
Silty clay 0.45 0.45 3.4 7.9 25 1.23 0.54 0.21 0.39 0.46
Clay 0.2 0.6 3.7 7.6 17 1.21 0.54 0.21 0.4 0.47
8.4 The Water Budget
Figure 8.2 shows a schematic diagram of the water balance of a field. We may write
the mass conservation equation for water inputs and outputs from that field during a
period to calculate the increment in total soil water content in the crop root zone
(TSWC):
Δ TSWC ¼ P þ I Es Ep SR DP þ WTC ð8:9Þ
where P is precipitation, I is applied irrigation, Es is evaporation from the soil

surface, Ep is transpiration, SR is surface runoff, i.e. water not infiltrated, DP is deep
percolation and WTC is upward water flow from the water table. For Irrigation
Scheduling it is better to express the amount of soil water as a deficit (Soil Water
Deficit, SWD), which is the amount of water required to bring the soil to FC or the
upper limit.
The water balance may be computed for different time periods (hour, day,
decade, months). We will focus first on several methods to calculate it for daily
values.
Fig. 8.2 Diagram of the

water balance components
of a field
Precipitation Evaporation &
T Transpiration
E
Irrigation Runoff
Root zone
ΔSW
Deep
Percolation
Capillary
rise
8.5 Infiltration
The rate at which water enters the soil through its surface (infiltration rate) is
reduced with time, until a relatively constant value is reached, which depends on
the soil (Fig. 8.3), being higher for sandy soils and lower for clay soils. The initial
infiltration rate is inversely proportional to the initial water content. This is because
the water potential gradient between the water and the soil is greater if the latter is
drier (since the matric potential is more negative). The water that does not infiltrate,
stays on the surface and moves to the lower parts running off the field. In Sect. 8.7
we will revise a methodology for evaluating the amount of runoff.
8.6 Deep Percolation
There are different approaches to compute the amount of water that moves out
below the crop root zone. The method presented here requires knowing the values
of soil depth (Z, mm), and the soil water content values at field capacity and at
saturation. Additionally it requires a dimensionless parameter (SWCON) which is
the fraction of water lost by percolation in 1 day in relation to the amount of soil
water which exceeds field capacity. The amount of water that exceeds the saturation
water content is considered instantly lost by deep percolation.
Suppose that on a given day an amount of water PI (mm) has infiltrated into the
soil. The amount of water that can be stored in the short term (SWCC) may be
calculated as:
INFILTRATION RATE, DEPTH/TIME
FINE SAND
STEADY STATE OR
BASIC INFILTRATION RATE
SILT LOAM
ELAPSED TIME OR OPPORTUNITY TIME
Fig. 8.3 Infiltration rate for two soils differing in saturated hydraulic conductivity
SWCC ¼ ðθSAT θÞZ ð8:10Þ
where θ is the average water content in the soil. If PI is higher than SWCC the
excess is lost by deep percolation on the same day. If the water content, after adding
infiltrated water, does not exceed Field Capacity then it is assumed that there is no
percolation. If the water content is between field capacity and saturation then
percolation is calculated as:
DP ¼ SWCON Z ðθ θFC Þ ð8:11Þ
Some models of this type have assumed that SWCON ¼ 1, that is, all water that
exceeds the upper limit is instantly lost. Such simplification may be valid in very
permeable soils or fallow situations or early crop stages. However, if SWCON is
less than 1 and the crop is extracting water from the soil, plant roots can extract
some of the water that exceeds the upper limit, which will not be lost by percolation.
When there is no crop or it has just been planted, deep percolation estimates are less
sensitive to the SWCON value. Ritchie has suggested SWCON values for different
types of soil:
Clay soil (very slow to moderately slow drainage): 0.01–0.25
Medium textured soils (moderate to moderately rapid drainage): 0.40–0.65
Sandy soils (fast to very rapid drainage): 0.75–0.85
Example 8.4 We will calculate deep percolation for a maize crop growing
on a loam soil 1000 mm depth with the following parameters:
θPWP ¼ 0:10 m3 m3 , θFC ¼ 0:25 m3 m3 , θSAT ¼ 0:35 m3 m3 ,

SWCON ¼ 0:4
The soil starts with a soil water content θ ¼ 0.30 m3 m3. A rainfall of
45 mm has fallen and 5 mm have not infiltrated.
Total water infiltrated will be:
PI ¼ 45 5 ¼ 40 mm
Short term storage is given by:
SWCC ¼ Zðθs θÞ ¼ 1000ð0:35 0:30Þ ¼ 50 mm
which is greater than total infiltration. Therefore 40 mm will be stored in the

short term. Soil water content increases to:
(continued)

θ ¼ θ þ 40=1000 ¼ 0:30 þ 0:04 ¼ 0:34 m3 m3
which is higher than θFC. Therefore some deep percolation will occur:
DP ¼ SWCON Zðθ θFC Þ ¼ 0:4 1000ð0:34 0:25Þ ¼ 36 mm
In the previous example we applied Eq. 8.11 to a single day. But we can extend
this analysis to the days after the rainfall event to calculate the total DP. We
consider a soil of depth Z (mm) with an initial water content θi, that loses water
by evaporation from the soil and crop transpiration at a rate equal to ET (mm/day)
(see Chap. 9 for more detail). After rainfall, an amount PI (mm) has infiltrated, so
that the water content is now: θi + PI/Z, which is greater than θFC (otherwise there
would be no percolation). We start from the differential equation describing the
variation of water content:
dθ ET
¼ SWCON ðθ θFC Þ ð8:12Þ
dt Z
And, after integrating:

SWCON t
SWCON θi þ PZI θFC þ ET
Z e ET=Z
θ ¼ θFC þ ð8:13Þ
SWCON
Equation 8.13 may be used to calculate the soil water content at time t (days after
rainfall) as a function of SWCON and ET. This equation may also be used to
calculate the time it takes to reach Field Capacity:

SWCON ðθi þ ZI θFC Þ
P
ln 1 þ ET=Z
tFC ¼ ð8:14Þ
SWCON
During that time, a total of tFC∙ET will be lost by evaporation from the soil and
the plants, so we may deduce the total percolation from rainfall until time tFC, when
the soil water content returns to Field Capacity:
X
tFC
PI
DP ¼ Z θi þ θFC tFC ET ð8:15Þ
1
Z
This equation shows that evaporation losses partly counterbalance percolation

losses, as shown in the example below.
Example 8.5 Let’s consider a loam soil 1 m (1000 mm) deep with a water
content of 0.23 m/m. A total of 50 mm infiltrates into the soil. We will
calculate percolation if: (a) ET ¼ 1 mm/day, (b) ET ¼ 8 mm/day. These
values are typical of soil covered by vegetation in winter and summer,
respectively, in the South of Spain.
Soil data given: θFC ¼ 0.25, SWCON ¼ 0.6
(a) ET ¼ 1 mm/day

0:6 ð0:23þ1000
50
0:25Þ
ln 1 þ 1=1000
tFC ¼ ¼ 4:91 days
0:6
X
tFC
PI
DP ¼ Z θi þ θFC tFC ET
1
Z

50
¼ 1000 0:23 þ 0:25 4:91 1 ¼ 25:1 mm
1000
(b) ET ¼ 8 mm/day

0:6 ð0:23þ1000
50
0:25Þ
ln 1 þ 8=1000
tFC ¼ ¼ 1:96 days
0:6
X
tFC
50
DP ¼ 1000 0:23 þ 0:25 1:96 8 ¼ 14:3 mm
1
1000
8.7 Surface Runoff
The main factors that determine surface runoff are rainfall intensity, soil type,
vegetation type, topography and surface roughness. In the method of the Soil
Conservation Service (US-SCS) all these factors are combined into a single factor,
called “runoff curve number” (CN) which is proportional to runoff potential.
To calculate the CN, soils are classified into four hydrologic groups from low to
high runoff potential:
(A) Low runoff potential. These are soils with high infiltration rate when wet. It is
generally the case of sandy or gravely soils, deep and well drained.
(B) Soils with moderate infiltration rate when wet, average depth and medium
texture.
(C) Soils with low infiltration rates when wetted. These soils are of fine texture or
have a horizon that hinders the drainage.
(D) High runoff potential. Includes soils with very low infiltration rates when wet,
such as expansive clay soils, soils with high water table, soils with a clay layer
near the surface and shallow soils over impervious materials.
In addition to soil characteristics, in the calculation of CN, the hydrological
condition of the field is considered, which can be good or bad depending on slope
and cultural practices. Table 8.2 shows the CN values based on hydrologic condi-
tion and soil group, for different types of crops and conservation practices. The CN
value shown in Table 8.2 implies average conditions of soil moisture when precip-
itation occurs (Antecedent Moisture Condition AMC 2) and is called CN2. CN
values that correspond to low AMC (AMC 1; CN1) or High (AMC 3; CN3) are
calculated by the following equations:
Table 8.2 Runoff curve number (CN) for different soils and cover types
Soil hydrological
Hydrologic group
Cover type Treatment condition A B C D
Fallow Bare soil crop resi- – 77 86 91 94
due (CR) Poor 76 85 90 93
Good 74 83 88 90
Row crops Straight row (SR) Poor 72 81 88 91
Good 67 78 85 89
SR and CR Poor 71 80 87 90
Good 64 75 82 85
Contoured (C) Poor 70 79 84 88
Good 65 75 82 86
C and CR Poor 69 78 83 87
Good 64 74 81 85
Contoured and ter- Poor 66 74 80 82
raced (C&T) Good 62 71 78 81
C&T + CR Poor 65 73 79 81
Good 61 70 77 80
Small grain SR. Poor 65 76 84 88
Good 63 75 83 87
SR+ CR Poor 64 75 83 86
Good 60 72 80 84
C Poor 63 74 82 85
Good 61 73 81 84
C&CR Poor 62 73 81 84
Good 60 72 80 83
C&T Poor 61 72 79 82
Good 59 70 78 81
C&T + CR Poor 60 71 78 81
Good 58 69 77 80
(continued)
Table 8.2 (continued)

Soil hydrological
Hydrologic group
Cover type Treatment condition A B C D
Close-seeded or broadcast legumes SR Poor 66 77 85 89
or rotation meadow Good 58 72 81 85
C Poor 64 75 83 85
Good 55 69 78 83
C&T Poor 63 73 80 83
Good 51 67 76 80
Pasture, grassland, or range- Poor 68 79 86 89
continuous grazing Fair 49 69 79 84
Good 39 61 74 80
Meadow – continuous grass, 30 58 71 78
protected mowed for hay
Brush – weed-grass mixture Poor 48 67 77 83
Fair 35 56 70 77
Good 30 48 65 73
Woods-grass combination (orchard) Poor 57 73 82 86
Fair 43 65 76 82
Good 32 58 72 79
Woods Poor 45 66 77 83
Fair 36 60 73 79
Good 30 55 70 77
100 CN 2
CN 1 ¼ CN 2 20 ð8:16Þ
100 CN 2 þ CN 2 e2:5330:0636 ð100 CN2 Þ
CN 3 ¼ CN 2 e0:00673 ð100CN2 Þ ð8:17Þ
The values of CN1 or CN3 cannot exceed 100. For values of soil water content
other than dry (CN1) or saturated soil (CN3), the curve number can be calculated as
a function of soil water content in the upper soil layer:
(a) If soil water content is higher than Field Capacity:
CN ¼ CN2 þ ðCN3 CN2Þðθ θFC Þ=ðθSAT θPWP Þ ð8:18Þ
(b) If soil water content is lower than Field Capacity:
CN ¼ CN1 þ ðCN2 CN1Þðθ θPWP Þ=ðθFC θPWP Þ ð8:19Þ
Once CN is known we calculate the maximum water depth (SMX, mm) that may
be infiltrated or stored above the soil surface:

100
SMX ¼ 254 1 ð8:20Þ
CN
If daily rainfall (P) is lower than 0.2 SMX, runoff is zero. Otherwise, runoff (SR,
mm) is calculated as:
ðP 0:2 SMXÞ2
SR ¼ ð8:21Þ
P þ 0:8 SMX
Example 8.6 We have a sunflower crop growing on a deep medium texture

soil with almost zero slope. A rain event of 40 mm occurs when the soil is
almost saturated
(a) The soil can be included in the B type, and the hydrologic condition is
good due to the absence of slope. In Table 8.2 we choose a value of
CN2 ¼ 78.
(b) As the soil is wet we calculate CN3 ¼ 90 using Eq. 8.15 and assign
CN ¼ 90.
(c) We calculate SMX:

100
SMX ¼ 254 1 ¼ 28 mm
90
(d) We compare 20 % of SMX (5.6 mm) with rainfall (40 mm). As P > 0.2
SMX then runoff will occur:
ð40 0:2 28Þ2

SR ¼ ¼ 19 mm
40 þ 0:8 28
In this case almost 50 % of rainfall would have infiltrated.
8.8 Effective Rainfall
Effective rainfall (Pe) is the fraction of total precipitation during a specific time
period that is not lost by runoff or percolation and thus is stored in the crop root
zone. It is a broad concept, sometimes used to characterize the seasonal or monthly
water balance or to assess the disposition of a rainfall event. A number of methods
for calculating effective rainfall for a monthly period have been proposed. These
methods have been developed for monthly periods and should not be used for
shorter time intervals. In any case they can only provide rough estimates as they
ignore key factors like soil properties or the actual rainfall distribution within the
month.
8.8.1 FAO Method
This method has been the result of a study conducted by FAO in arid and sub-humid
areas. The equation was developed to estimate the monthly effective rainfall (Pe)
that is exceeded in 80 % of the years, and is used for irrigation system design.
Effective rainfall is estimated by the following equations:
Pe ¼ 0:6 P 10 if P < 70 mm ð8:22aÞ

Pe ¼ 0:8P 24 if P > 70 mm ð8:22bÞ
8.8.2 USDA-Soil Conservation Service Method
In this method, in addition to rainfall, crop evapotranspiration (ET) and soil water
deficit before irrigation are taken into account according to:

Pe ¼ f ðSWDÞ 1:25 P0:824 2:93 100:001 ET ð8:23Þ
f ðSWDÞ ¼ 0:53 þ 0:0116 SWD 8:94 105 SWD2 þ 2:32 107 SWD3 ð8:24Þ
where SWD (mm) is the soil water deficit just before irrigation and ET is given in
mm/month.
Bibliography

New York: Springer.
Marshall, T. J., Holmes, J. W., & Rose, C. W. (1996). Soil physics. New York: Cambridge
University Press.
Williams, J. R. (1991). Runoff and water erosion. In J. Hanks & J. T. Ritchie (Eds.), Modeling
plant and soil systems (pp. 439–456). Madison: ASA-CSSA-SSSA.
Chapter 9
The Components of Evapotranspiration
Abstract Evapotranspiration is the sum of evaporation from the soil surface and
the plant surfaces, and transpiration. The evaporation from the soil in agronomy
follows a two-stage process depending if the soil surface is wet after a rain or
irrigation or has already dried up. When the soil surface is wet the rate of evapo-
ration is potentially very high; that’s why the rainfall frequency is the main driver of
the soil evaporation, especially at low ground cover. The core model to quantify the
process of evaporation is the combination equation, later applied to crop canopies
and for computing plant transpiration known as the Penman-Monteith equation.
This equation has two resistance variables (the aerodynamic and canopy resistance)
which are hard to quantify as they are constantly changing with the physical
environment and the plant physiological state. The Penman-Monteith equation is
the established method to analyze the evaporation processes in plants and stands
and it has been thoroughly verified. The transpiration of trees is heavily dependent
on canopy conductance and scales up well with the ground cover or the fraction of
intercepted radiation.
9.1 Introduction
Evaporation from vegetated surfaces (or evapotranspiration, as it combines evap-

oration from soils and transpiration from plants) is the main component of water
loss from terrestrial ecosystems so its quantification is of great importance in
hydrology, agronomy and related sciences. Moreover in agronomy, evaporation is
usually directly proportional to crop productivity, as discussed in Chap. 14.

L. Testi

Agriculture, DOI 10.1007/978-3-319-46116-8_9
Fig. 9.1 Effect of soil

water content on 1
transpiration or
TRANSPIRATION
evapotranspiration
0
qPWP SOIL WATER CONTENT qFC
Evapotranspiration (ET) is the sum of direct evaporation from the soil surface
(Es), plant transpiration (Ep) and direct evaporation from plant surfaces (Eps):
ET ¼ Es þ Ep þ Eps ð9:1Þ
Strictly transpiration is the water vapor flow through the stomata of plants. For this
flow to occur, evaporation must take place in the substomatal cavities. If the canopy
surface is dry, Eps ¼ 0, so that:
ET ¼ Es þ Ep ð9:2Þ
Maximum transpiration (and ET) occurs when soil water is not limiting root water
uptake, which usually happens for soil water content above one third of available
soil water (Fig. 9.1).
9.2 Measurement of Evapotranspiration
Crop ET can be measured directly by determining the mass of water lost from a
vegetated surface or estimated indirectly. Direct ET measurement is done in
weighing lysimeters which are large containers open at the top to enclose a volume
of soil whose mass can be measured accurately and where plants are grown.
Lysimeters are placed in the middle of large fields to ensure that the microclimate
experienced by the plants inside them is the same as that of the surrounding plants.
They are large in size and are deep enough so that root systems are not limited by
the lysimeter walls. For example, at the Agricultural Research Center of Cordoba
(Spain) two weighing lysimeters were installed in 1987 with an area of 6 m2 each
and a depth of 1.5 m. The measurement systems are accurate enough to register
water losses equivalent to the ET of short sub-hourly periods (5–10 min).
Estimates of ET and fluxes of other scalars (e.g. canopy photosynthesis) may be
obtained using micrometeorological methods (Box 9.1).
9 The Components of Evapotranspiration 109
The simplest method for estimating ET is the water balance, which requires the
estimation of the water balance components, so that ET is obtained by difference.
From Eq. 8.3 in Chap. 8:
ET ¼ P þ I SR DP þ WTC ΔTSWC ð9:3Þ
Example 9.1 Soil water content measurements were taken of a 1-m deep soil
under a soybean crop on two dates (August 11 and August 19). The average
soil water content was 0.22 (August 11) and 0.175 cm3 cm3 (August 19). In
that period there has been a rain episode of 20 mm. Assuming no runoff, no
deep percolation and that the water table is too deep to contribute water to the
root zone through capillary rise, we can calculate the ET for the period as the
difference between rainfall and the increase in total soil water content:
ET ¼ P ΔTSWC
The soil water content in the first date will be:
TSWCð11AugÞ ¼ 0:22 1000 ¼ 220 mm
Analogously we calculate the water content in the second date, resulting in

TSWC (19 Aug) ¼ 175 mm.
The increase will be:
ΔTSWC ¼ TSWC (19 Aug) – TSWC (11 Aug) ¼ 175–220 ¼ 45 mm
And so the ET will be:
ET ¼ 20 þ 45 ¼ 65 mm
Box 9.1 Micrometeorological Methods

Measurements of meteorological variables close to the canopy allow the
estimation of the components of the energy balance. For instance by measur-
ing net radiation, soil heat flux and the gradients of temperature and vapor
pressure above the canopy it is possible to estimate latent heat flux, using the
so called Bowen Ratio method.
If instead of measuring the gradients we use an infrared thermometer to get
canopy temperature we can calculate sensible heat flux and thus deduce latent
heat flux.
(continued)
Box 9.1 (continued)

A more sophisticated approach is the eddy covariance method. It is based
on high frequency measurements of scalars (temperature, absolute humidity,
CO2 concentration) and vertical wind velocity. The covariance of any scalar
and vertical wind speed (w) provides a measure of the flux of the scalar. For
instance the covariance of temperature and w is directly proportional to
sensible heat flux, while that of water vapor concentration and w leads to
latent heat flux. Note that this technique may be applied to measure the flux of
any chemical (e.g. ammonia) provided that we have the proper instrument for
measuring its concentration.
9.3 Evaporation from the Soil Surface
Philip described the evaporation from a bare soil surface (Es) after wetting, as a
three-stage process. In the first stage (energy limited) the soil surface is wet and
the hydraulic conductivity is high, so that the evaporation rate is only limited by
the amount of energy available for evaporation at the surface. In this case the
evaporation is approximately equal to the evaporation from a short grass field
which is defined in Chap. 10 as the reference evapotranspiration ET0. This stage
continues until a certain amount of water has evaporated (Ue) that depends on
soil type, ranging from 5–6 mm (well drained soils) to 12–14 mm (heavy clay
soils).
When the second (soil limited) stage begins, the soil hydraulic conductivity has
been reduced to values which limit the water flow to the soil surface from the deeper
soil layers. During this phase Es is decreasing as a function of the square root of
time since the start of the second phase (t):
pffi pffiffiffiffiffiffiffiffiffiffi
Es ¼ ce t t1 ð9:4Þ
where ce is a constant that depends on the soil type, although its value is usually
close to 3.5 mm day0.5.
The third stage described by Philip corresponds to extremely dry soil in which
water is transported to the soil surface as water vapor and the evaporation is
extremely low. For practical purposes in agronomy we can calculate Es considering
only the first two phases.
Therefore, the evaporation of a soil depends primarily on the availability of
energy at its surface and on the water content of the upper soil layers (down to about
30 cm). Thus, when the soil is thoroughly wet, Es is similar to the evaporation from
a full cover crop, and can be assumed equal to ET0. By contrast, when the soil
surface is dry, Es is reduced to very low values (Fig. 9.2).
5.0
SOIL EVAPOTRATION (mm day–1)

4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
1 2 3 4 5 6 7 8 9 10
TIME AFTER WETTING (days)
Fig. 9.2 Soil evaporation rate after soil wetting. Reference ET is 4.5 mm/day. The parameter Ue is
9 mm
If the soil is partly covered by a crop canopy or by crop residues, the amount of
energy reaching the soil surface is reduced, and so will be Es for the first stage (Es1):
Es1 ¼ ET 0 ð1 f PI Þ ð9:5Þ
where fPI is the fraction of radiation intercepted that does not reach the soil surface.
Example 9.2 Rain has fallen and wetted a bare soil thoroughly. The soil
parameter for evaporation during the first stage (energy limited) is Ue ¼ 9 mm
and ET0 is 4.5 mm/day.
In this case the first stage (Es equal to ET0) will last 2 days as it is the time
required to evaporate 9 mm:
Ue =ET0 ¼ ð9 mmÞ=ð4:5 mm=dayÞ ¼ 2 days
Therefore Es(1) ¼ Es(2) ¼ 4.5 mm/day

The second stage of Es starts on the third day thus, taking ce as 3.5 mm/
day0.5:
Es ð3Þ ¼ 3:5ð10:5 ð1 0Þ0:5 Þ ¼ 3:50 mm=day

Es ð4Þ ¼ 3:5 20:5 ð2 1Þ0:5 ¼ 1:45 mm=day, and so on:
(continued)

If the same situation occurred in an olive orchard intercepting 40 % of
radiation (fPI ¼ 0.4) then evaporation in stage 1 would be:
Es1 ¼ ET 0 ð1 f PI Þ ¼ 4:5 ð1 0:4Þ ¼ 2:7 mm=day
In this case the first stage will last 3.3 days (it may be rounded to 3). On the
fourth day the second stage will start so:

Es ð4Þ ¼ 3:5 10:5 ð1 0Þ0:5 ¼ 3:50 mm=day

Es ð5Þ ¼ 3:5 20:5 ð2 1Þ0:5 ¼ 1:45 mm=day, and so on:
According to the two stage model of evaporation, the average Es for a given
period (month, year) will be proportional to the frequency of wetting. If wetting
events have an average duration of WD, and the average interval between two
consecutive events is WI, we may calculate the average daily Es using the equations
for evaporation in stage 1 for WD-0.5 days and the two equations (stage 1 and stage
2) for WI – WD + 0.5 days, assuming that on day WD the rainfall stops in the
middle of the day.
The average soil evaporation for the period will be:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ðWD 0:5Þ ET 0 þ Ue þ ce WI ðWD 0:5Þ ET Ue
t0
Esm ¼ ð9:6Þ
0
WI
where t0 ¼ (ce/ET0)2 if ce>ET0. Otherwise t0 ¼ 0. Please note that the square root
function in Eq. 9.6 is valid only for positive values. A negative value would indicate
that the soil stays in first stage evaporation so its average evaporation rate is
equivalent to ET0.
Example 9.3 A farmer has sown a summer crop during June in southern Italy
(ET0 ¼ 6 mm/day). To ensure crop emergence, irrigations are applied every
5 days (WI ¼ 5 days, WD ¼ 1 day). The soil has a parameter for first stage
Ue ¼ 9 mm and ce ¼ 3.5 mm day0.5. Calculate the average evaporation for
the period.
Applying Eq. 9.6:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
0:5 6 þ 9 þ 3:5 5 0:5 9=6
Esm ¼ ¼ 3:6 mm=d
5
Example 9.4 The average number of rainy days in Cordoba (Spain) during
March (ET0¼ 3 mm/day) is 9.3. In principle, this would imply that on average
a rainfall event would occur every 3.3 days. However, rainy days tend to
cluster, to occur in consecutive days. According to Villalobos and Fereres the
average interval between two consecutive rainy spells may be estimated as:
1
WI ¼ ð9:7Þ
0:75 f w ð1 f w Þ
where fw is the mean frequency of rainy days. In the case of Cordoba

fw ¼ 9.3/31 ¼ 0.3, thus WI ¼ 6.3 days.
And each period of 6.3 days would be composed of 1.9 rainy days (30 % of
6.3 days) and 4.4 dry days. Now taking WI ¼ 6.3 days and WD ¼ 1.9 days we
may apply Eq. 9.6 (we will assume a soil with the parameter for first stage
Ue ¼ 6 mm and ce ¼ 3.5 mm0.5) to calculate the average soil evaporation
during March:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ð1:9 0:5Þ 3 þ 6 þ 3:5 6:3 ð1:9 0:5Þ 6=3
Esm ¼ ¼ 2:6 mm=d
6:3
9.4 Analysis of Evapotranspiration with the Penman-

Monteith Equation
The first formulation of a combination equation to calculate evaporation is due to

Penman in 1948, who combined the energy balance equation with those of latent
heat and sensible heat fluxes. Similar solutions were proposed by Ferguson in
Australia and Budyko in Russia. The most widespread formulation of the combi-
nation equation is due to Monteith that started from the following equations:
the energy balance (Chap. 7):
Rn G ¼ LE þ H ð9:8Þ
Latent heat flux (Chap. 5):
ρCp ðesc ea Þ
LE ¼ ð9:9Þ
γ ðr c þ r a Þ
Sensible heat flux (Chap. 5):

ðT c T a Þ
H ¼ ρCp ð9:10Þ
ra
In this method, the slope of the saturation vapor pressure function versus
temperature (Δ, kPa K1) is approximated as:
esc es
Δ¼ ð9:11Þ
Tc Ta
where esc is the saturation vapor pressure at canopy temperature (Tc) and es is the
saturation vapor pressure at air temperature (Ta). Therefore:
esc ¼ es þ ΔðT c T a Þ ð9:12Þ
Now adding and subtracting es in part of Eq. 9.9 and using Eq. 9.12:
esc ea ¼ esc ea þ es es ¼ ΔðT c T a Þ þ es ea

¼ ΔðT c T a Þ þ VPD ð9:13Þ
So Eq. 9.9 can be written as:
ρCp ΔðT c T a Þ þ VPD

LE ¼ ð9:14Þ
γ ðr c þ r a Þ
From Eqs. 9.10 and 9.8 we can write the term Δ (Tc – Ta) as:
Δ r a H Δ r a ðRn G LEÞ
ΔðT c T a Þ ¼ ¼ ð9:15Þ
ρCp ρ Cp
Which placed in Eq. 9.14, allows solving for LE, leading to the Penman-Monteith
equation:
ρ C
ΔðRn GÞ þ ra p VPD
LE ¼ ð9:16Þ
Δ þ γ 1 þ rrac
The Penman-Monteith equation indicates that crop evaporation depends on mete-

orological (radiation, temperature, humidity, wind speed) and crop factors (rc). In
the case of ra apart from canopy characteristics (height, leaf area) there is a
dependence on meteorological conditions (wind speed).
This equation has the drawback of requiring information on canopy resistance
(rc, see Chap. 5) which in turn depends on different environmental factors such as
temperature, radiation and Vapor Pressure Deficit (VPD). It has been found that in a
large number of crop species, rc is proportional to VPD. Thus, on one hand, a high
VPD increases LE, but if rc increases as well due to the high VPD, it contributes to
reducing crop transpiration when evaporative demand is high.
The Penman-Monteith equation is very useful from a conceptual standpoint. For

example, when the VPD tends to zero (extremely humid conditions) and the canopy
resistance is small compared with the aerodynamic resistance (e.g. flat, smooth
surfaces such as a short grass crop) evaporation tends to:
ΔðRn GÞ
LE ¼ ð9:17Þ
Δþγ
This value has been termed “equilibrium evaporation” and corresponds also to
the evaporation from a canopy that has a very high aerodynamic resistance. In cases
like this evaporation is dependent only on radiation and temperature (as Δ depends
on temperature). These conditions are found in short, smooth crop canopies where
the humidity is high and the aerodynamic resistance is very high due to the absence
of wind or in crops growing in greenhouses in still air.,.
The opposite case is that of very rough canopies and windy conditions (low ra)
with rather high canopy resistance (e.g. forests with variable tree heights and
isolated trees with small leaves such as olives). In this case it is easy to demonstrate
that evaporation (called “imposed evaporation”) depends only on VPD and canopy
resistance:
ρCp 1
LE ¼ VPD ð9:18Þ
γ rc
These two extremes of equilibrium and imposed evaporation are said to correspond
to uncoupled and coupled canopies in relation to the atmosphere. In a perfectly
coupled canopy the absence of aerodynamic resistance makes transpiration highly
responsive to changes in VPD. On the contrary, the uncoupled canopy (e.g. grass in
the absence of wind) is somehow isolated from changes in atmospheric conditions
above it. If stomata close then the reduced transpiration leads to higher canopy
temperature which leads to increased transpiration (see Eq. 5.12 in Chap. 5). In the
field, crop canopies are not perfectly coupled or uncoupled but somewhere in
between. Those better coupled to the atmosphere (for example, tree crops) can
exert more control of transpiration via stomata closure than those that are largely
uncoupled such as smooth field crops.
9.5 Transpiration
The calculation of actual transpiration is difficult as it depends on meteorological

(e.g. radiation) and plant factors, primarily through the response of stomata to the
aerial environment but also including root system responses. For well-watered
conditions, a very simple approach may be followed. For a given condition the
transpiration coefficient, that is the ratio transpiration/ET0, is proportional to the
fraction of intercepted radiation, so transpiration (Ep, mm day1) can be calculated

as:
Ep ¼ f PI K tf ET 0 ð9:19Þ
where fPI is the fraction of radiation intercepted, Ktf is the transpiration ratio for full
interception, which is close to 1 for most herbaceous and evergreen tree crops and
between 1.2 and 1.8 for deciduous tree crops (Table 9.1).
For isolated trees, Eq. 9.19 becomes:
Ep tree ¼ RRi K tf ET 0 ð9:20Þ
where Ep tree is tree transpiration (L day1 tree1) and RRi is the relative radiation
interception (the ratio between total intercepted radiation and incoming radiation;
see Chap. 3).
Example 9.5 In Example 3.5, we calculated the relative interception of a

small olive tree with radius 0.5 m in Cordoba, Spain on 21 March as
RRi ¼ 0.69 m2.
If ET0 is 3 mm day1, calculate tree transpiration.
Ep tree ¼ RRi K tf ET 0 ¼ 0:69 1 3 ¼ 2:07 L day1
Villalobos et al. (2013) proposed a more detailed model for calculating the
transpiration of orchard canopies This model considers a daily “bulk” canopy
conductance (Gc), i.e. the inverse of the canopy resistance for the whole stand, a
parameter that is related to VPD and to the radiation intercepted by the canopies as:
Q Rs
Gc ¼ α ð9:21Þ
a þ b VPD
where Q is the fraction of radiation intercepted by the canopy (see Chap. 3), Rs is
the daily solar radiation (MJ m2 day1), VPD is the vapor pressure deficit (kPa), α
is a generic coefficient and a and b are empirical coefficients which vary with the
tree species (see Table 9.1). As rc ¼ 1/Gc, we can use the Penman-Monteith
equation (Eq. 9.16) to calculate the transpiration of the stand. As orchard canopies
are generally well coupled to the atmosphere,, similar results may be obtained using
the “imposed” evaporation equation (Eq. 9.18), which is much more practical as it
does not require knowledge of the aerodynamic resistance, ra. The transpiration
(in mm day1) can thus be calculated as:
Q Rs VPD
Ep ¼ 37:08 103 ð9:22Þ
a þ b VPD Pat
Table 9.1 Empirical a b Ktf

coefficients to calculate bulk
Species (μE mol1) (μE mol1 kPa1)
canopy conductance of some
cultivated tree species Orange 1002 1666 0.8
Walnut 1287 673 1.4
Apple 442 911 1.8
Olive 1211 1447 1.1
Apricot 452 2050
Peach 333 633 1.8
Pistachio 359 624 1.9
Almond 1.25
where Pat is the atmospheric pressure (kPa). The coefficient 37.08 103 is used to
convert the units to mm day1. This model has been developed and tested in semi-
arid climates, i.e. when the transpiration is primarily regulated by the evaporative
demand of the atmosphere rather than by solar radiation only.
Example 9.6 An intensive olive orchard at an altitude of 100 m a.s.l. intercepts

52 % of the incident daily radiation. Let’s calculate its transpiration on a sunny
day with daily total solar radiation of 27.6 MJ m2 day1 and average VPD of
2.8 kPa. Assume an atmospheric pressure of 101 kPa.
Using Eq. 9.24 with the coefficients of Table 9.1 for olive (a ¼ 1211 μE
mol1 and b ¼ 1447 μE mol1kPa1),
0:52 27:6 2:8

Ep ¼ 37:08 103 ¼ 2:83 mm day1
ð1211 þ 1447 2:8Þ 100:1
9.6 Evaporation from Wetted Canopies
When the plant is wet (for example, immediately after a rainfall event) the water
film and droplets covering the foliage will eventually evaporate directly into the
atmosphere. Note that this evaporation flux is neither transpiration nor evaporation
from the soil, although the water involved is still coming from the rain or irrigation:
it should then be evaluated and considered as part of the evapotranspiration flux
(see Eq. 9.1) to correctly assess the evaporation component of the water budget.
Although the direct evaporation flux from wetted canopies is often overlooked, it
may be appreciable when frequent rains wet dense canopies with high LAI, which
can intercept a significant amount of rain. The maximum capacity of rainfall
interception by agricultural species with full ground cover is around 0.25 mm per
unit of LAI, i.e. a wheat canopy with LAI ¼ 4 can intercepted 1 mm of rain. Another
issue is when mechanically moved sprinkler irrigation systems (center pivots)
irrigate very frequently and a significant part of the application is intercepted by a
full canopy before it infiltrates in the soil. In all cases, the direct evaporation is the
40
30
Transpiration (L/hour)
20
10
0
8 9 10 11 12 13 14 15 16 17 18 19 20
Time (GMT)
Fig. 9.3 Time course of transpiration of a walnut tree in Cordoba (Spain) on June 5, 2009. Rain
started at 16:20 (vertical arrow) and lasted 1 h
main process to be evaluated in order to assess the canopy wetness duration after a
rainfall. This knowledge is important because the time interval that a canopy stays
wet is strongly related with the chances of infection by many fungal diseases;
furthermore, during that interval the transpiration rate is nil (see Fig. 9.3), so no
water is actually extracted from the soil by the roots with the associated reduction in
soil water and nutrient uptake. To calculate the evaporation of wet canopies, the
Penman-Monteith equation may be used assuming zero canopy resistance.
Bibliography
Cuenca, R. H. (1989). Irrigation system design – An engineering approach. Englewood Cliffs:

Prentice Hall.
Doorenbos, J., Pruitt, W. O. (1977). Guidelines for predicting crop water requirements. FAO
irrigation and drainage paper nº 24. FAO, Rome
Monteith, J. L. (1964). Evaporation and environment. Symposia of the Society for Experimental
Biology, 19, 205–234.
Ritchie, J. T. (1972). Model for predicting evaporation from a row crop with incomplete cover.
Water Resources Research, 8, 1204–1213.
Villalobos, F. J., Testi, L., Orgaz, F., Garcı́a-Tejera, O., Lopez-Bernal, A., Gonzalez-Dugo, M. V.,
Ballester-Lurbe, C., Castel, J. R., Alarcon-Caballero, J. J., Nicolas-Nicolas, E., Girona, J.,
Marsal, J., & Fereres, E. (2013). Modelling canopy conductance and transpiration of fruit trees
in Mediterranean areas: A simplified approach. Agricultural and Forest Meteorology,
171–172, 93–103.
Chapter 10
Calculation of Evapotranspiration and Crop
Water Requirements
Abstract Reference ET (ET0) is defined as the ET of short grass with full soil
cover, and an unlimited supply of water and nutrients. In the absence of water
deficit, the ET of any crop may be calculated as the product Kc x ET0, where Kc is
the crop coefficient, which depends on crop related factors (leaf area, roughness)
and ET0, the reference ET (grass), which is a function of climatic variables
(radiation, temperature, humidity and wind speed). The main equation for calcu-
lating ET0 is the Penman-Monteith-FAO, although the Hargreaves equation can be
used when only air temperature data are available. Kc is calculated by the method
proposed by FAO which uses linear functions between the initial, maximum and
harvest dates of the cycle. The initial Kc depends on the frequency of soil wetting
and ET0. The maximum values of Kc of annual crops and deciduous fruit trees are
typically between 1.0 and 1.3 (median 1.2). The crop irrigation water requirement is
the difference between ET and the effective rainfall, although the role of stored soil
water as a contributor to meeting the ET demand may be important in some
situations to reduce the dimensions (and investment) of the irrigation network.
10.1 Introduction
The ET in the absence of water stress is usually calculated as the product:
ET ¼ K c ET 0 ð10:1Þ

L. Testi

Agriculture, DOI 10.1007/978-3-319-46116-8_10
where Kc is a crop coefficient, which depends on factors related to the crop (leaf
area, roughness, crop management) and ET0 is the reference ET, by definition, the
ET of a well-irrigated short grass surface, which is ideally only a function of
climatic variables (radiation, temperature, humidity and wind speed). This expres-
sion is always valid unless water stress reduces ET, which generally occurs when
65–80 % of the extractable soil water is depleted. Below this value the Kc decreases
linearly to 0 when it reaches the Permanent Wilting Point (Fig. 9.1).
The usual method of calculating crop ET is to calculate ET0 based on meteoro-
logical data and apply a variable Kc that changes with crop development stage.
Numerous methods have been proposed for estimating ET0 (e.g. Penman-FAO) or
the Kc (e.g. FAO) which will be discussed below.
10.2 Reference ET
Reference ET (ET0) is defined as the ET of short (8–15 cm height) grass with full
soil cover, and a good supply of water and nutrients. The concept of ET0 came to
replace the term “potential ET” used widely in the past but lacking a precise
definition.
In some areas, networks of automatic weather stations provide daily information
via Internet which usually would allow applying combination formulas based on
the Penman-Monteith equation. However, in many world areas there is a dearth of
agrometeorological data which limits the use of the most precise methods.
If only maximum and minimum temperature data are available the equation of
Hargreaves (see 10.2.1) provides a good approximation in many areas. In some
cases evaporation pans are available on the farm and provide another estimate of
ET0, as shown in Appendix 10.1.
The annual time course of ET0 follows a pattern similar to that of solar radiation.
As an example, Fig. 10.1 shows the daily ET0 calculated by the Penman-Monteith
equation in Santaella (southern Spain, semi-arid Mediterranean climate), typical of
mid-latitudes. The mean values range from 1 mm/day during the winter to 7 mm/
day during the summer. The average annual total ET0 is 1278 mm.
10.2.1 Method of Hargreaves
In 1985 Hargreaves and Samani proposed a simple equation for estimating ET0
(in mm/day):
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ET 0 ¼ 5:52103 K RS RA ðT avg þ 17:8Þ T max T min ð10:2Þ
10 Calculation of Evapotranspiration and Crop Water Requirements 121
5
ETo
0
0 30 60 90 120 150 180 210 240 270 300 330 360
DOY
Fig. 10.1 Annual time course of reference ET calculated using the method of Penman-Monteith-
FAO for Santaella (Spain) from 2000 to 2013
where RA is extraterrestrial radiation (MJ m2 day1), and Tavg, Tmax and Tmin are
the average, maximum and minimum air temperatures ( C), respectively, while
KRS is another coefficient already defined in Chap. 3 on solar radiation (Eq. 3.7).
Usually, the values of KRS vary between 0.16 and 0.19 for interior and coastal
locations, respectively. This equation has shown good performance for different
areas despite being based only on measured air temperature. This is because it
includes a term associated with the potential radiation of the location, by consid-
ering the extraterrestrial radiation and a variable related to the degree of cloudiness
(the amplitude of air temperature). Thus, in very cloudy days there is little heating
during the day (low solar radiation) and little cooling during the night (clouds
reduce long wave radiation loss). Therefore the maximum and minimum temper-
atures will not differ much. By contrast, in clear days the greater warming during
the day and the increased cooling at night lead to a greater difference between the
maximum and minimum temperatures. The Hargreaves method may be less reliable
when applied in areas with little daily temperature oscillation or where the temper-
ature amplitude is influenced by factors not related to solar radiation, e.g. the
presence of massive water bodies (coastal regions).
10.2.2 Penman-Monteith-FAO Method
This equation has become the standard for ET0 calculation as proposed by FAO.
Applying the Penman-Monteith equation (Eq. 9.16) to a hypothetical grass canopy
of height 0.12 m and canopy resistance 69 s m1 we can deduce the ET0 (mm day1)
for 24-h periods as:
ΔRn þ 0:5 VPD U2

ET 0 ¼ ð10:3Þ
2:45½Δ þ 0:067ð1 þ 0:33U 2 Þ
where Δ (kPa K1) is the slope of the saturation vapor pressure function versus
temperature (Eq. 9.13), Rn is the net radiation (MJ m2 day1), VPD is vapor
pressure deficit (kPa) and U2 is wind speed at 2-m height (m s1).
The value of Δ (kPa K1) can be calculated as:
4098es
Δ¼ ð10:4Þ
½237:3 þ T2
where T is air temperature ( C) and es is the saturation vapor pressure (kPa) which
is a function of temperature (Eq. 5.2).
10.3 Crop Coefficients
The crop coefficient is a parameter that reflects the specific features of the crop as
they affect ET, such as leaf area, height, fraction of ground cover, etc. Its value is
determined experimentally as:
ET
Kc ¼ ð10:5Þ
ET 0
In irrigated crops Kc depends primarily on the fraction of ground cover, and, if the
latter is low, it depends on the water content of the soil surface as it determines the
rate of soil evaporation (see 9.3). Thus, when the crop has not emerged yet, the Kc
of a bare dry soil may be as low as 0.1, but if the soil surface is wet, the Kc increases
to values close to 1. When the crop completely covers the ground, the Kc becomes
almost independent of the water content of the soil surface, and usually exceeds
1 (1.05–1.30), with a typical value of 1.20.
In some cases the latent heat flux exceeds the net radiation, that is, an amount of
sensible heat is used to evaporate water. This phenomenon typically occurs due to
the movement of masses of hot, dry air from dry areas surrounding a wet area where
water is available for evaporation (oasis effect). We may distinguish the clothesline
effect when advection occurs at the field level, and is characterized by sensible heat
input decreasing from the edge of the plot inwards. At a smaller scale (micro-
advection) there is a sensible heat flux transported from dry soil to the surrounding
plants. In the event that an isolated irrigated field is surrounded by dry land (fallow,
stubble, dry crops) the clothesline effect provides additional energy for ET so that
the crop coefficient may be much greater than the values indicated previously. This
Fig. 10.2 Crop coefficient kc max

curve for an annual crop
according to the FAO kc
approach. The duration of
the four phases is 40, 30,
30 and 30 days, respectively
kc fin
kc i
A B C D
TIME
enhancement will be greater for small plots with tall plants, but there is no reliable
model to quantify exactly the Kc in these situations. Some authors suggest that the
extreme value of Kc for isolated irrigated plots may be as high as 2.5, but this value
should be considered a hypothetical limit that is reached only under infrequent
extreme conditions and for a limited time.
The Kc is not constant during the season but changes with the ground cover, the
plant height, the soil surface wetting and plant aging. The most widespread method
for estimating the value of the crop coefficient at any time of the growing season is
the one proposed by FAO (Doorenbos and Pruitt 1977). This method represents the
Kc curve as a set of straight lines. To define the curve it is necessary to know in
advance the length of phases A, B, C and D, and the value of Kc at three points (Kc1,
Kc2 and Kc3). The initial phase (A) ends when the crop reaches 20 % of ground
cover, while the rapid growth phase (B) ends when ground cover is 70–80 %, which
usually corresponds to values of Leaf Area Index around 2.5–3.0. Figure 10.2
shows an example of a curve of Kc for annual crops in which the phase durations
are 40, 30, 30 and 30 days, and the crop coefficients which define the curve are 0.3,
1.2 and 0.5. Table 10.1 shows the values of Kc2 (maximum) and Kc3 (final) for a
number of crops and Appendix 10.3 presents a more complete list.
Although the methodology proposed by FAO allows fitting the crop coefficient
to specific climatic conditions, in Table 10.1 we show the intervals of Kc that we
believe may hold for temperate areas. Moreover, Table 10.1 also shows indicative
values of the phase durations for different species. These durations should be taken
merely as examples, since the actual duration depends on many factors (climate
zone, cultivar, date of sowing or bud burst in tree crops, climatic conditions of the
year, etc.). The main driving factor for changes in the duration of crop stages is
temperature, which may change from year to year or if sowing date is changed. For
actual irrigation scheduling, the Kc curve should always be obtained using
on-season information on the phases (beginning and duration) obtained from
empirical observations in the field.
The Kc in the initial phase (Kc1) is a function of the frequency of rain and
irrigation and ET0 during that period because most of the ET of a crop during this
124
Table 10.1 Durations of phases of growth cycle (FAO method) and crop coefficients in phase C (mid) and at harvest (end) for different crop species
Stage duration (days)
Crop species Date of start of crop cycle (N hemisphere) A B C D Kc mid Kc end
Alfalfa (hay) Mar–Octa 5 10–20 10 5–10 0.9–1.15b 0.90
Barley Nov–Mar 20–40 25–30 40–65 20–40 1.15 0.2–0.4
Bean (phaseolus) (dry seed) May–Jun 15–25 25–30 30–40 20 1.15 0.30
Citrus (70 % CC)* Jan 60 90 120 95 0.65 0.75
Coffee 0.95 0.95
Cotton Mar–May; Sept 30–45 50–90 45–60 45–55 1.15–1.25 0.4–0.7
Grapes (table or raisins) Mar–May 20 40–50 75–120 20–60 0.85 0.45
Grapes (wine) April 30 60 40 80 0.70 0.45
Lettuce Nov–April 20–35 30–50 15–45 10 1.00–1.05 0.95
Maize (grain) Mar–Jun 20–30 35–50 40–60 30–50 1.20 0.35–0.6
Millet Apr–Jun 15–20 25–30 40–55 25–35 1.00 0.30
Olives (60 % CC)* Mar 30 90 60 95 0.70 0.70
Palm trees 1.00 1.00
Peas (dry harv.) Mar–May 15–35 25–30 30–35 30 1.15 0.30
Pome fruits, cherries (60–70 % CC)* Mar 20–30 50–70 90–130 30–60 1.10 0.75
Potato Nov–Jan; Apr–May 25–45 30–35 40–70 20–30 1.15–1.25 0.7–0.8
Rapeseed, Canola Nov–Dec 25–40 35–40 60–70 30–35 1.10 0.35
Rice May (Medit.); May–Dec (tropics) 30 30 60–80 30–40 1.20 0.9–0.6
Sorghum (grain) Apr–Jun 20 35 40–45 30 1.00–1.10 0.55
Soybean May–Jun; Dec (tropics) 15–20 15–35 40–75 15–30 1.15 0.50
10
Stone fruits (60–70 % CC)* Mar 21–30 51–70 91–130 31–60 1.10 0.65
Sugar beet Nov–June 25–50 30–75 50–100 10–65 1.20 0.70
Sugar cane (ratoon) 25–35 70–105 135–210 50–70 1.25 0.75
Sugar cane (virgin) 35–75 60–105 190–330 120–210 1.25 0.75
Sunflower Feb–May 25–45 35–40 45–60 25 1.20 0.35–0.5
Tea 1.00 1.00
Tomato Jan–May 30–35 40–45 40–70 25–30 1.15–1.25c 0.7–0.9
Wheat (winter) Oct–Dec 20–30 60–140 40–70 30 1.15 0.25–0.4
Adapted from Doorenbos and Pruitt (1977) and Allen et al. (1998). For some crops the final Kc shows a wide interval, as its value depends on crop use (fresh or
dry). This Table should be used with caution as the actual duration of phases may change for different regions, cultivars and weather conditions of each year
*When cover crop or weeds are present add 0.2 to the crop coefficient. If canopy cover (CC) is lower than the value indicated in the table (CC0 ) then Kc ¼ 0.15
+ CC (Kc0 0.15)/CC0 where Kc0 is the value shown
a
For the first cut cycle use durations twice the values shown
b
Lower value is the seasonal average; higher value is at full cover-before cutting
c
When cultivated on stalks, the Kc mid should be increased by 0.05–0.1
Calculation of Evapotranspiration and Crop Water Requirements
125
9
8
7
6
ET (mm day–1)
5
4
3
2
1
0
125 130 135 140 145 150 155 160 165 170 175 180
DAY OF YEAR
Fig. 10.3 Evapotranspiration of a cotton crop in Cordoba (Spain) for stages A and B. Inter-row
tillage was performed on DOY 156 which caused an increase in soil evaporation
phase is direct evaporation from the soil surface. Bare soil evaporation is approx-
imately equal to the ET0 while the soil surface is wet (energy-limited or first stage
evaporation, see Sect. 9.3). As the soil dries the soil hydraulic conductivity
decreases (soil limited or second stage). The importance of Es in determining the
initial Kc is manifested in significant variations of ET in the early stages of the crop
cycle associated with the occurrence of rainfall or irrigation (Fig. 10.3). Doorenbos
and Pruitt (1977) proposed a method of calculating the Kc in the initial development
stage (until the onset of rapid crop growth, Kc1) which was summarized in the
following equations by Allen et al. (1998):
For WI <4 days:
K c1 ¼ ð1:286 0:27 lnWIÞexp½ð0:01 0:042lnWIÞET 01 ð10:6aÞ
For WI >4 days:
2
K c1 ¼ exp½ð0:02 0:04lnWI ÞET 01 ð10:6bÞ
ðWI Þ0:49
where WI is the interval between irrigations or rainfall events during the initial
stage and ET01 is the average ET0 during the initial stage (mm/day). If we consider
the effect of rainfall major errors may arise if we assume that rainy days are evenly
distributed over the period (Villalobos and Fereres 1989. Transactions of ASAE,
32 (1):181–188.), thus a correction should be applied to calculate the interval
between rainfall events (see Example 9.4). It is important to calculate the initial
Kc accurately because errors in the initial Kc translate into errors in Kc during the
rapid growth period (phase B) as this is determined by interpolation between the

values of Kc1 and Kc2.
If real time information on ground cover fraction (fGC) is available we can assign
values of Kc by considering the proportionality between them. In arid and semiarid
areas:
K c2 K c1
K c ¼ K c1 þ f GC ð10:7Þ
f GC2
where fGC2 is the ground cover fraction associated to Kc2 (fGC2 ¼ 1 for full cover
crops).
The method described above for calculating crop coefficients is valid for well-
watered plants, and would thus be valid for irrigated crops or for periods in rainfed
crops where water does not limit ET. For rainfed or deficit irrigated conditions the
crop reduces its transpiration as soil water is depleted below a threshold (water
stress). Consequently, the Kc is reduced. A very simple model for estimating the Kc
as a function of soil water would be:
3ðθ θPWP Þ
K c ¼ K *c ð10:8Þ
ðθFC θPWP Þ
where Kc* is the crop coefficient with no water stress and θ is the average
volumetric soil water content. This equation is valid for θ lower than θPWP +
(θFCθPWP)/3. If θ is higher, then the Kc is not reduced, which implies that crops
can use around two third of extractable water without having a reduction in ET.
10.4 Crop Coefficients of Perennial Species
Forage crops and pastures have a variable Kc depending on management (cutting

frequency, maximum LAI). For instance alfalfa has an average seasonal Kc close to
1, oscillating between 0.6 after cutting then increasing to 1.2 before the next cut.
For fruit trees evaporation from the soil may play an important role throughout
the cycle depending on the fraction of the soil which is exposed to solar radiation.
Where the tree canopies are small, the Kc will be higher during periods with high
rainfall frequency. The contribution of tree transpiration is proportional to ground
cover and will be nil during the winter season in deciduous fruit trees, contrary to
evergreen fruit trees (olive, citrus) that keep their leaf area during the whole year. In
these evergreen species under mild Mediterranean climates, the Kc could be
maximum in winter (high soil evaporation due to high frequency of rain) and
minimum during summer (low soil evaporation, high stomatal resistance in
response to high VPD) (Fig. 10.4).
1.0
0.9
0.8
0.7
0.6
Kc
0.5
0.4
0.3
0.2
0.1
0.0
0 30 60 90 120 150 180 210 240 270 300 330 360
DAY OF YEAR
Fig. 10.4 Average crop coefficient for an olive orchard with 40 % ground cover in Cordoba
(Spain) in 2001 and 2002. The data have been grouped in 4-week periods
10.5 Evapotranspiration in Greenhouses
The evapotranspiration inside greenhouses is usually lower than in the open field
because the shelter cover reduces the incoming radiation, the driving force of ET
and turbulence. Specific methods have been developed to evaluate the ET of
screened or sheltered crops. In sophisticated greenhouses such as those used for
production of high-value crops or for ornamental horticulture, with climatic control
(heating/cooling, supplemental light, etc.) data from sensors are available for using
complex transpiration models in short time steps (e.g. 10-min). Furthermore,
intelligent use of sensors may provide indirect estimates of greenhouse ET. For
instance, measurements of air flow into the greenhouse (Qin, m3 s1) along with
sensors of temperature and air vapor pressure in air going in (Tin, ein) and out (Tout,
eout) allow calculating ET of a greenhouse (mm/h) of surface area Ag (m2) as:

Qin 3600 eout ein
ET ¼ ð10:9Þ
Ag 0:4615 T out T in
where temperatures are given in K and vapor pressures in kPa.

A different case is that of unheated greenhouses with passive ventilation,
typically with plastic covers, such as those in the Mediterranean coast of Almeria
(Spain) and other mild climate areas. The conditions inside the greenhouse are
characterized by reduced turbulence, higher humidity and higher temperature that
the outside. This in theory leads to equilibrium ET (see Eq. 9.17), i.e. ET is
governed by radiation, a situation which has been confirmed empirically leading to

an equation for reference evapotranspiration (ET0, mm day1) of the form:
1 Δ
ET 0 ¼ 0:7 Rsi ð10:10Þ
2:45 Δ þ γ
where Rsi (MJ m2 day1) is the solar radiation inside the greenhouse, which can
also be estimated from the solar radiation measured in the open if the transmissivity
of the cover (τgc) is known (Rsi ¼ τgc Rs). For instance, transmissivity of polyeth-
ylene film of 0.2 mm is around 0.7.
Once the reference ET is known inside the greenhouse, calculation of ET
requires a crop coefficient, which is usually somewhat higher (10–20 %) than that
of crops grown outside. For instance, measurements of Kc values of greenhouse
tomatoes have reached 1.4 while those in open fields seldom exceed 1.2.
10.6 Calculation of Maximum ET for Designing Irrigation

Systems
The ET of a given crop in a location can vary from year to year depending on
weather conditions. To design irrigation systems it would be desirable to have
historical series of ET (and precipitation) to determine the water requirements at
different levels of probability. Often we only have the average values of crop ET,
which obviously will be exceeded in some years. Doorenbos and Pruitt (1977)
proposed an adjustment method for calculating the ET which corresponds to a
probability level of 75 % (ET75, which will be exceeded only 25 % of the years) as a
function of the average ET (ETavg) and the mean irrigation applied (Ia). The method
is shown in Fig. 10.5, where four different climate types are considered. Each line
can be calculated as:
ET 75 pffiffiffiffi
¼ C 0:06 ðC 1Þ I a ð10:11Þ
ET avg
where the mean irrigation applied is given in mm and the coefficient C is 1.21, 1.49,
1.33 and 1.43, for types 1, 2, 3 and 4, respectively.
Note that, ideally, the system should be designed to supply the peak or maximum
ET level corresponding to the period of highest ET of the crop mix of the farm
rotation or of the irrigated area. The decision to determine the size of the irrigation
network is basically economic, as reducing the flow rate below the maximum
requirements of the extreme year and highest demanding crop of the rotation will
require less capital investment but will increase the risks of crop water deficits in
some years. Also, if the system is dimensioned for annual crops of low requirements
1.4
1
MAXIMUM ET / AVERAGE ET 2
1.3 3
4
1.2
1.1
1.0
4 5 6 7 8 9 10 11 12 13 14 15
SQUARE ROOT OF IRRIGATION AMOUNT (mm0.5)
Fig. 10.5 Ratio of 75 % probability ET and average ET as a function of the square root of mean
irrigation applied in each irrigation event. Four climate types are considered. (1) Arid and semi-
arid with clear skies during summer. (2) Continental climates in mid latitudes and sub-humid
climates with variable cloudiness. (3) Mid latitude continental climates with ET up to 5 mm/day.
(4) Mid latitude continental climates with ET up to 10 mm/day
(for example, winter cereals), its conversion to summer crops or perennials is not
possible without leaving some of the land unirrigated.
Example 10.1 We will calculate the ET of maize during August (Kc ¼ 1.2)
in Evora, Portugal (ET0 ¼ 7.0 mm/day) and Paris, France (ET0 ¼ 4.0 mm/
day) for 75 % probability assuming that irrigation doses of 60 mm are applied.
We calculate the average ET for both locations:
Evora. ET ¼ 1.2 7.0 ¼ 8.4 mm day1
Paris. ET ¼ 1.2 4.0 ¼ 4.8 mm day1
Evora has a semiarid climate so it corresponds to type 1 (Fig. 10.5). Paris
has a subhumid climate during summer (type 2). Applying Eq. 10.11 for
Evora (C ¼ 1.21):
ET 75 pffiffiffiffiffi
¼ 1:21 0:06 ð1:21 1Þ 60 ¼ 1:11
ET avg
For Paris C ¼ 1.49 so:
(continued)

ET 75 pffiffiffiffiffi
¼ 1:49 0:06 ð1:49 1Þ 60 ¼ 1:26
ET avg
Therefore the values of ET75 for the two locations will be:
Evora: ET75 ¼ 1.10 8.4 ¼ 9.2 mm/day
Paris: ET75 ¼ 1.26 4.8 ¼ 6.0 mm/day
10.7 Calculation of Crop Water Requirements
Crop irrigation water requirement is the amount of water to be supplied to maintain

a maximum level of ET, and can be calculated invoking water balance (Chap. 8) as
the difference between the ET (the potential value, without restrictions of any kind)
and water supplied by rainfall or extracted from the soil during a given period of
time:
IWRn ¼ ET Pe ðΔTSWCÞ ¼ ET Pe þ ΔTSWC ð10:12Þ
Note that the extraction of water from the soil is expressed as increase in the total
soil water content with a negative sign. In this equation Pe is effective precipitation,
i.e. precipitation not lost by runoff or deep percolation (see Chap. 8).
We can distinguish between net and gross water requirement. In the first case we
refer to the amount of water required assuming no losses during irrigation and
perfect uniformity in the spatial distribution of irrigation water. In almost all cases
these assumptions are not met and we are forced to apply more water than the actual
crop water consumption (Chap. 19). The total amount to apply including excess
water is the gross irrigation water requirement (GIWR ¼ net requirement/applica-
tion efficiency).
The calculation of IWR may be performed for different time intervals (weeks,
months) and for different spatial scales (field, farm, irrigation scheme). The first
step is always computing IWR of each field and then obtaining the weighted
average using the fractions of area as weights.
The term related to soil water storage is frequently omitted in the calculations
which may lead to large overestimations of IWR in some cases. The soil water
stored at sowing depends on the recharge during the fallow period since the harvest
of the previous crop. This can be calculated by adding effective precipitation and
discounting soil evaporation (Eq. 9.6) during fallow. This value should not exceed
the soil water storage capacity (Chap. 8). Stored soil water is seen in irrigated
agriculture as an insurance against irrigation system failures and extremely high ET
periods, thus keeping a moderately high level of soil water stored during much of
the irrigation season reduces risks. However, by the end of the season soil water
content may be nearly depleted, thus it is possible to use a large fraction of stored
soil water (e.g. 80–90 %) which should be discounted from the requirements of the
final period (see Chap. 20 on irrigation scheduling).
After calculating IWR for each crop of the farm we calculate the average farm
IWR as:
X
n
IWRfarm ¼ IWRi si ð10:13Þ
1
where n is the number of crops, and IWRi and si are irrigation water requirement
and fraction of farm area of crop i, respectively.
Appendix 10.1: Class A Evaporation Pan
A relatively simple way to obtain ET0 is to empirically observe the evaporation

from a free water surface in a standardized device and then apply some empirical
relationships to convert the direct evaporation of water to that of a grass surface.
This method became quite popular in the past because it does not require meteo-
rological measurements taken in expensive weather stations, thus is affordable for
farmers even in undeveloped countries. However, the corrections to apply depend
also on the micrometeorology of the place where the device is installed.
The most popular model of these devices is the standard National Weather
Service Class A type evaporation pan which has a diameter of 1.21 m and height
0.254 m.
It is normally installed on a wooden platform set on the ground. The pan is filled
with water to within 6 cm of the top and exposed to represent an open body of water.
The evaporation rate is measured as the difference in water level between consec-
utive measurements.
Then, reference ET may be calculated as:
ET 0 ¼ K p Epan ðA10:1:1Þ
where Epan is the measured pan evaporation (mm/day) and Kp is the pan coefficient
which is in the range 0.35–0.85 with an average value of 0.75. The actual value will
depend on the surroundings of the pan, on relative humidity and on wind speed.
According to FAO manual 24, if the pan is surrounded by crops, the pan coefficient
for the class A may be calculated as:
K p ¼ 0:108 0:0286 U þ 0:0422lnðXÞ þ 0:1434lnðRH Þ

0:000631 ½lnðXÞ2 lnðRH Þ ðA10:1:2Þ
where U is wind speed (m/s) at 2-m height, X is the distance (m) covered with crops
around the pan and RH is mean relative humidity.
If the pan is located on a dry location (bare soil, stubble) the pan coefficient is
calculated as:
6

K p ¼ 0:61 þ RH ð0:00341 0:000162
h U Þ þ U 0:00327ln ðX Þ 9:59 10 X i
þ ½4:459 þ lnðU Þ 0:0106lnðXÞ þ 0:00063 ½lnðXÞ2 0:00289 U
ðA10:1:3Þ
A simpler formula may be applied to both cases:

2:45 Epan
K p ¼ 0:85 exp 0:15 ðA10:1:4Þ
Rn
where Rn is calculated net radiation (MJ m2 day1).
Appendix 10.2: Calculating Crop Coefficients Following

the Model of Ritchie’s
As a crop grows, intercepted radiation increases and so does energy available for
transpiration. At the same time energy available for evaporation at the soil surface
decreases. In 1972 Professor Joe T. Ritchie proposed a model for calculating the ET
of crops by computing separately transpiration and evaporation from the soil
surface (Ritchie 1972). Using this model the Kc may be calculated as a function
of ground cover and soil wetting frequency.
Model results are summarized in Eq. A10.2.1 where Kc is a function of ET0, the
interval between rain events or irrigations (WI) and the fraction of ground covered
by the crop (fGC). The lower values of Kc logically occur when WI is large and the
ET0 is high, if fGC is very small. In contrast when fGC is high Kc varies little with
ET0 and WI.
13:3 5:2 f gc
K c ¼ 0:14 þ 1:08 fgc þ ðA10:2:1Þ
WI ET 0
Example A10.1 A garlic crop in March has a ground cover of 0.3. The last
rain occurred 7 days ago. Since then, the average ET0 has been 3.5 mm/day.
Let’s calculate the Kc during that period.
13:3 5:2 0:3

K c ¼ 0:14 þ 1:08 0:3 þ ¼ 0:14 þ 0:32 þ 0:48
7 3:5
¼ 0:94
Appendix 10.3: Crop Coefficients, Root Depth and Factor

for Allowable Depletion
Crop coefficients in phase C (mid) and at harvest (end) for different crop species.
Adapted from Doorenbos and Pruitt (1977) and Allen et al. (1998). For some crops
the final Kc shows a wide interval, as its value depends on crop use (fresh or dry).
Values of maximum crop height, maximum root depth and a coefficient for
calculating allowable depletion (Chapter 20) are also shown
Crop species
Cereals and Max. crop Max. root
pseudocereals Kc mid Kc end height (m) depth (m) FAD
Barley 1.15 0.2–0.4 1 1.0–1.5 0.09
Maize (grain) (field 1.2 0.35–0.60 2 1.0–1.7 0.09
corn)
Maize, sweet (sweet 1.15 1.0–1.05 1.5 0.8–1.2 0.10
corn)
Millet 1 0.3 1.5 1.0–2.0 0.09
Oats 1.15 0.2–0.4 1 1.0–1.5 0.09
Rice 1.2 0.90–0.60 1 0.5–1.0 0.16
Rye 1.15 0.2–0.4 1 0.9–2.3 0.08
Sorghum (grain) 1.00–1.10 0.55 1.0–2.0 1.0–2.0 0.09
Sorghum (sweet) 1.1–1.2 1.05 2.0–4.0 1.0–2.0 0.10
Wheat (spring) 1.15 0.25–0.4 1 1.0–1.5 0.09
Wheat (winter) 1.15 0.25–0.4 1 1.5–1.8 0.09
Max. crop Max. root
Forages Kc mid Kc end height (m) depth (m) FAD
Alfalfa Hay 0.9–1.15a 0.9 0.7 1.0–2.0 0.09
Bermuda hay 1 0.85 0.35 1.0–1.5 0.09
Bermuda (spring crop 0.9 0.65 0.4 1.0–1.5 0.08
for seed)
Clover hay, berseem 0.9 0.85 0.6 0.6–0.9 0.10
Rye grass hay 1.05 1 0.3 0.6–1.0 0.08
Sudan grass hay 0.9 0.85 1.2 1.0–1.5 0.09
(annual)
(continued)
Crop species
Pasture (rotated 0.85–1.05 0.85 0.15–0.30 0.5–1.5 0.08
grazing)
Pasture (extensive 0.75 0.75 0.1 0.5–1.5 0.08
grazing)
Turf grass (cool season) 0.95 0.95 0.1 0.5–1.0 0.12
Turf grass (warm 0.85 0.85 0.1 0.5–1.0 0.10
season)
Fruit trees, trees and Max. crop Max. root
shrubs Kc mid Kc end height (m) depth (m) FAD
Almonds (70 % CC)* 1.1–1.2 0.65 5 1.0–2.0 0.12
Apple (60–70 % CC)* 1.1 0.75 4 1.0–2.0 0.10
Apricot (60–70 % CC)* 1.1 0.65 3 1.0–2.0 0.10
Avocado (70 % CC)* 0.85 0.75 3 0.5–1.0 0.06
Banana (year 1) 1.1 1 3 0.5–0.9 0.13
Banana (year 2) 1.2 1.1 4 0.5–0.9 0.13
Berries (bushes) 1.05 0.5 1.5 0.6–1.2 0.10
Cacao 1.05 1.05 3 0.7–1.0 0.14
Citrus (70 % CC)* 0.65 0.75 3.0–4.0 1.0–1.5 0.10
Cherry (60–70 % CC)* 1.1 0.75 4 1.0–2.0 0.10
Coffee 0.95 0.95 2.0–3.0 0.9–1.5 0.12
Conifers 0.9–1 0.9–1 10 1.0–1.5 0.06
Grapevine (table or 0.85 0.45 2 1.0–2.0 0.13
raisin)
Grapevine (wine) 0.7 0.45 1.5–2 1.0–2.0 0.11
Kiwi 1.05 1.05 3 0.7–1.3 0.13
Olives (60 % CC)* 0.7 0.7 5–7 1.2–1.7 0.07
Palm (date) 0.95 0.95 8 1.5–2.5 0.10
Palm tres 1 1 8 0.7–1.1 0.07
Peach (60–70 % CC)* 1.1 0.65 3 1.0–2.0 0.10
Pear (60–70 % CC)* 1.1 0.75 4 1.0–2.0 0.10
Pineapple 0.3 0.3 0.6–1.2 0.3–0.6 0.10
Pistachio (60–70 % 1.1 0.45 3–6 1.0–1.5 0.12
CC)*
Plum (60–70 % CC)* 1.1 0.65 3 1.0–2.0 0.10
Rubber trees 1 1 10 1.0–1.5 0.12
Tea (non-shaded) 1 1 1.5 0.9–1.5 0.12
Tea (shaded) 1.15 1.15 2 0.9–1.5 0.11
Walnut (70 % CC)* 1.1 0.65 4–5 1.7–2.4 0.10
Max. crop Max. root
Horticultural crops Kc mid Kc end height (m) depth (m) FAD
Artichokes 1 0.95 0.7 0.6–0.9 0.11
Asparagus 1 0.3 0.2–0.8 1.2–1.8 0.11
Bean (green) 1.1 0.95 1.5 0.5–0.7 0.11
Beet (table) 1.1 0.95 0.2 0.6–1.0 0.10
(continued)
Crop species
Broccoli 1.05 0.95 0.3 0.4–0.6 0.11
Brussel sprouts 1.05 0.95 0.4 0.4–0.6 0.11
Cabbage 1–1.1 0.9–1 0.4 0.5–0.8 0.11
Carrots 1.05 0.95 0.3 0.5–1.0 0.13
Cauliflower 1.05 0.95 0.4 0.4–0.7 0.11
Celery 1.05 1 0.6 0.3–0.5 0.16
Cucumber 1 0.75–0.9 0.3 0.7–1.2 0.10
Egg plant 1.05 0.9 0.8 0.7–1.2 0.11
Fababean (fresh) 1.1 0.9 0.5–0.7 0.11
Lettuce 1–1.05 0.95 0.3 0.3–0.5 0.14
Melon 1.05–1.1 0.7–0.75 0.4 0.8–1.5 0.12
Melon (cantaloupe) 0.85 0.6 0.3 0.9–1.5 0.11
Mint 1.15 1.1 0.6–0.8 0.4–0.8 0.12
Peas (fresh) 1.2 1 0.7 0.6–1.0 0.13
Pepper 1.05–1.15 0.70–0.90 0.7 0.5–1.0 0.14
Pumpkin, winter 1 0.8 0.4 1.0–1.5 0.13
squash
Radish 0.9 0.85 0.3 0.3–0.5 0.14
Spinach 1 0.95 0.3 0.3–0.5 0.16
Squash, zucchini 0.95–1.0 0.75–0.9 0.3 0.6–1.0 0.10
Strawberries 0.85 0.75 0.2 0.2–0.3 0.16
Tomato 1.15–1.25b 0.70–0.90 0.6 0.7–1.5 0.12
Watermelon 1 0.75 0.4 0.8–1.5 0.12
Max. crop Max. root
Legumes Kc mid Kc end height (m) depth (m) FAD
Beans (Phaseolus) 1.1–1.25 0.3–0.9 0.4 0.6–0.9 0.11
Beans (lima) 1.1 0.5 0.5 0.8–1.2 0.11
Chick pea 1 0.35 0.4 0.6–1.0 0.10
Fababean (broad bean) 1.15–1.25 0.3–1.1 0.8 0.5–0.7 0.11
Green gram and 1.05 0.3–0.6 0.4 0.6–1.0 0.11
cowpeas
Groundnut (peanut) 1.15 0.6 0.4 0.5–1.0 0.10
Lentil 1.1–1.2 0.3–0.5 0.5 0.6–0.8 0.10
Peas 1.15 0.3–1.1 0.5 0.6–1.0 0.12
Soybeans 1.15–1.25 0.5 0.5–1.0 0.6–1.3 0.10
Roots, tubers and Max. crop Max. root
bulbs Kc mid Kc end height (m) depth (m) FAD
Cassava (year 1) 0.8 0.3 1 0.5–0.8 0.13
Cassava (year 2) 1.1 0.5 1.5 0.7–1.0 0.12
Garlic 1–1.2 0.7–1.05 0.5 0.3–0.5 0.14
Onions 1–1.1 0.75–1 0.3–0.5 0.5–0.8 0.14
Parsnip 1.05 0.95 0.4 0.5–1.0 0.12
Potato 1.15–1.25 0.70–0.80 0.6 0.4–0.6 0.13
Sugar beet 1.2 0.7 0.5 0.7–1.2 0.09
(continued)
Crop species
Sweet potato 1.15 0.65 0.4 1.0–1.5 0.10
Turnip (and Rutabaga) 1.1 0.95 0.6 0.5–1.0 0.10
Sugar, oil and fiber Max. crop Max. root
crops Kc mid Kc end height (m) depth (m) FAD
Cotton 1.15–1.25 0.4–0.7 1.2–1.5 1.0–1.7 0.07
Castorbean (Ricinus) 1.15 0.55 0.3 1.0–2.0 0.10
Flax 1.1 0.25 1.2 1.0–1.5 0.10
Hops 1.05 0.85 5 1.2 0.10
Rapeseed, canola 1.1 0.35 0.6 1.0–1.5 0.08
Safflower 1.1 0.25 0.8 1.0–2.0 0.08
Sesame 1.1 0.25 1 1.0–1.5 0.08
Sisal 0.4–0.7 0.4–0.7 1.5 0.5–1.0 0.04
Sugar cane 1.25 0.75 3 1.2–2.0 0.07
Sunflower 1.2 0.35–0.5 2 0.8–1.5 0.11
Tobacco 1.15 0.8 1.5–2.0 0.8 0.10
a
Lower value is the seasonal average; higher value is at full cover-before cutting
b
When cultivated on stalks, the Kc mid should be increased by 0.05–0.1
*When cover crop or weeds are present add 0.2 to the crop coefficient. If canopy cover (CC) is
lower than the value indicated in the table (CC0 ) then Kc ¼ 0.15 + CC (Kc0 0.15)/CC0 where Kc0 is
the value shown.
Bibliography
Allen, R. G., Pereira, L. S., Raes, D., Smith, M. (1998). Crop evapotranspiration. Guidelines for
computing crop water requirements. FAO irrigation and drainage paper 56. FAO, Rome.
Cuenca, R. H. (1989). Irrigation system design – An engineering approach. Englewood Cliffs:
Prentice Hall.
Doorenbos, J., & Pruitt, W. O. (1977). Guidelines for predicting crop water requirements (FAO
irrigation and drainage paper n 24). Rome: FAO.
Ritchie, J. T. (1972). Model for predicting evaporation from a row crop with incomplete cover.
Water Resources Research, 8, 1204–1213.
Chapter 11
Crop Development and Growth
Abstract Growth of crops, plants or plant parts is defined as the irreversible

increase in size whereas development is the continuous change in plant form and
function with characteristic transition phases. Growth is primarily associated with
capture and allocation of resources whereas development is mostly related to
non-resource environmental cues such as temperature, photoperiod and light qual-
ity. We separate development and growth conceptually, but both types of processes
are closely linked. Thermal time and variations of thermal time corrected to account
for photoperiod and vernalization are useful to model crop phenological develop-
ment. Crop development, in particular the time of flowering, is one of the most
important traits for crop adaptation. Breeders, agronomists and growers understand
the importance of matching the pattern of phenological development to their
particular environments, and use a combination of genetic and agronomic tools to
manipulate development. Crop growth depends on the capacity of the canopy to
capture CO2 and radiation, the capacity of the root system to capture water and
nutrients from soil, and the efficiency of the crop to transform resources (water,
nutrients, radiation, carbon dioxide) into dry matter. Stresses such as water deficits
or soil compaction reduce growth by reducing the amount of resources captured by
the crop, by reducing the efficiency in the use of resources or both.
V.O. Sadras (*)

South Australian Research and Development Institute, Waite Campus, Urrbrae, Australia
F.J. Villalobos • E. Fereres

Agriculture, DOI 10.1007/978-3-319-46116-8_11
142 V.O. Sadras et al.
11.1 Introduction
Growth of crops, plants or plant parts is defined as the irreversible increase in size
whereas development is the continuous change in plant form and function with
characteristic transition phases. The expansion of a leaf or the accumulation of crop
biomass are typically growth processes whereas the transition from a vegetative
meristem, producing leaves, to a reproductive meristem producing flowers is a
characteristic developmental process. We distinguish morphological development
(e.g. appearance of successive structures in the plant) from phenological or phasic
development which deals with the duration of the different phases of the crop cycle.
The distinction between growth and development is important for two reasons.
First, growth is primarily associated with capture and allocation of resources
whereas development is mostly related to non-resource environmental clues such
as temperature, photoperiod and light quality. Second, the physiological processes
involved are different, as discussed in this chapter. Whereas we separate growth and
development conceptually, organs, plants and crops grow and develop simulta-
neously, and for many agronomically important traits the limits between growth
and development are blurred. For example, a wheat grain grows, i.e. it expands in
volume and gains mass, and also develops, e.g. leaf and root primordia are
differentiated in the embryo. Developmental biology (Box 11.1) and crop growth
analysis are thus distinct perspectives underpinning the investigation of develop-
ment and growth. In this chapter, we outline agronomically important aspects of
these processes.
Box 11.1 Developmental Biology

The fundamental question of developmental biology is this: how do different
cellular phenotypes emerge from cells which share a common set of genes? A
typical flowering plant has 30 different cell types, whereas a typical verte-
brate has about 120 cell types. All this diversity has to be explained in terms
of differential gene expression, as the 30 or so cell types in a plant share the
same genome – genetically, the cells of the wheat root endodermis and the
mesophyll cells in the flag leaf of the same plant are essentially identical.
Likewise, your neurons and liver cells are genetically identical, but their
shape and physiology are obviously different. For readers interested in this
question, we recommend the book of Mary Jane West-Eberhard: Develop-
mental plasticity and evolution (Oxford University Press, 2003).
Some examples demonstrate the practical implications of understanding
the process of cell differentiation. Stem cells, which are undifferentiated cells
with potential to generate any cell type, present animal and human health with
potential opportunities for new therapies. Short time between generations is
one of the cornerstones of successful plant breeding programs. Making use of
advanced knowledge of cell differentiation and tissue culture, breeders can
(continued)
11 Crop Development and Growth 143
Box 11.1 (continued)

currently grow up to six generations of chickpeas in a single year. Likewise,
tissue culture exploiting the principles of cell differentiation is a rapid,
effective and cheap method to generate virus-free seedlings of high value in
horticulture.
11.2 Phenological Development
Scales have been devised to characterise phenological development in annual and

perennial crops. They are based on the concept of pheno-stage; major pheno-stages
in annual crops include
1. Sowing
2. Germination
3. Emergence
4. Juvenile Phase/Initiation of leaves
5. Floral initiation (formation of primordia of reproductive structures)
6. Flowering
7. Physiological maturity
8. Harvest maturity
The case of perennial species is more complex. Trees stay in juvenile phase, thus
not flowering, for several years after seed germination. However, most trees of
agricultural interest are not propagated by seeds but vegetatively from cuttings that
are rooted in nurseries. In this case the cutting may not be juvenile which reduces
the time until flowering. In some tree species, the growth of the main stem beyond a
certain length accelerates the end of the juvenile period. Once the juvenile phase is
over, the tree will follow annual cycles that resemble those of an annual plant,
following two possible strategies:
(a) Deciduous species: Most fruit trees and vines belong to this category. All the
leaves fall in autumn-winter in response to cold temperatures and/or short
photoperiod. Buds stay dormant during winter and usually require low temper-
atures for an extended period (chilling requirement) until they respond to warm
temperature and bud break occurs. After that vegetative and reproductive
growth will occur with a degree of overlap that depends on the species. For
instance, flowering may occur before leaf growth starts (stone fruits, e.g. peach)
or later (pome fruits e.g. apple), while harvest may occur in early to midsummer
(several months before leaf fall, e.g. cherry) or in late summer or start of autumn
(e.g. apple).
(b) Evergreen species: They include citrus spp., olive and tropical fruit trees
(e.g. mango). Leaves stay in the tree for long periods (2–3 years). In some
cases (e.g. olive) the tree stay dormant (no vegetative growth) during winter and
resume growth in the spring.
The period between two phenostages constitutes a phenophase; we can be
interested, for example, in the phase sowing-emergence or emergence-flowering.
Some of these phases are well-defined biologically, for example the phase between
floral initiation and flowering. Other phases are not defined biologically but with
agronomic criteria; for example the phase from physiological maturity, when grain
reaches its maximum dry matter, to harvest maturity, when grains reach a moisture
content suitable for mechanical harvest. Harvest maturity of wine grapes is defined
by oenological criteria, including sugar concentration and acidity, and complemen-
tary traits such as colour and aromas. Sugar:acid ratio is an important trait for the
decision of harvest in most fruit crops.
The duration of the cycle of different crops is shown in Appendix 11.1.
All phenophases are responsive to temperature, which is the main environmental
influence on development (Box 11.2). The phase sowing-emergence can also be
influenced by the content of water and oxygen in the soil. In some species,
photoperiod also affects the duration of some phenophases. Some species and
phases are also responsive to low temperature in a process called vernalisation.
Here we outline the effects of mean temperature, low temperature (vernalisation)
and photoperiod on phenological development.
Box 11.2. Phenology and Global Warming

Phenological shifts are the most conspicuous biological signal of global
warming. Using a systematic phenological network data set of more than
125,000 observational series of 542 plant and 19 animal species in 21 -
European countries between 1971 and 2000, it was found that 78 % of all
leafing, flowering and fruiting records advanced (30 % significantly) and 3 %
were significantly delayed.
The consequences of warming for agriculture are many fold and varied. At
high latitudes, warming is extending the window for cropping, with overall
positive implications for crop production. In China and USA, milder winters
are allowing for earlier crop sowing, which combined with new varieties and
practices is improving crop production. Climate projections and modelling
indicate that Finland’s crop productivity by 2050 would be close to the
current productivity in Denmark. In 2000, the EU has accepted Denmark as
a wine producing country, and the Association of Danish Winegrowers now
counts more than 1400 members. In temperate environments like the Pampas,
warming over the last few decades has shortened the season of wheat crops,
allowing for early sowing and higher yield of soybean in wheat/soybean
double cropping. In temperate and subtropical environments, warming is
shortening the season of crops, with potential for yield reduction in the
absence of adaptive practices. Increasing frequency and incidence of heat
(continued)

waves may reduce both yield and quality of crops. Thus, the outcomes of
warming are complex and varied, particularly when warming interacts with
changes in rainfall, but a good deal of crop responses to warming are related
to phenological changes.
Another interesting consequence of warming is the decoupling of pro-
cesses, from ecosystems to molecules. In the last five decades, the produc-
tivity of Northern Sea fisheries has declined. The main reason is that warming
has “decoupled” the phenology of the components of the trophic web. This
means that for example, predators and preys which were phenologically
synchronized before are now out of phase, with direct consequences for the
structure and function of the whole ecosystem. In red grapevine varieties,
warming is decoupling sugars and anthocyanins. This means that fruit reaches
sugar maturity with less pigmentation; growers have therefore two choices.
They can wait longer to harvest, hence allowing for color to develop; this
leads to undesirably high sugar and alcohol concentration. Or they can
harvest at the right sugar level, and deal with lack of color in the winery.
The process of decoupling is therefore an agronomically important aspect of
warming, which is related to developmental and growth processes.
11.2.1 Effects of Daily Mean Temperature
Figure 11.1 shows the relationship between the daily rate of phenological develop-
ment (R, unit: day1) and daily mean temperature (T, unit: C). The daily rate of
development is the inverse of the duration of the phenophase (D, unit: day); for
example if it takes 10 days to complete the phase sowing-emergence, the daily rate
of development is 0.1 day1. The rate of development increases linearly with
temperature between the base (Tb) and optimum temperature (To), and decreases
between the optimum and maximum temperature (Tm) for development. These
Fig. 11.1 Responses of

development rate to
DEVELOPMENT RATE
temperature. The dashed

line represents the actual
response while the solid line
is a linear approximation
TEMPERATURE
three parameters, Tb, To, and Tm constitute the “cardinal” temperatures for devel-
opment, and depend on the species and phenological phase.
The daily rate of development is assumed to be zero (i.e. the plant does not
develop) if the mean temperature is below Tb or above Tm. The concept of "thermal
time" (also called degree days or heat units) is useful to predict the duration of a
phase for different temperatures. Thermal time (TT, unit: Cd) is defined as the sum
of daily mean temperature (T, C), above the base temperature, from the beginning
to the end of the phase; for example for the phase sowing-emergence:
X emergence
TT ¼ sowing
ðT T b Þ ð11:1Þ
and the daily rate of development when the daily mean temperature is between Tb
and To is
R ¼ 1=D ¼ ðT Tb Þ=TT ð11:2aÞ
The duration of a phase can thus be calculated if we know the daily mean
temperature, the base temperature and the thermal time required to complete the
phase:
D ¼ TT=ðT Tb Þ ð11:2bÞ
Table 16.1 in Chap. 16 shows Tb and TT for the phase sowing-emergence for a
number of crops. With adequate supply of water and oxygen, the thermal time
required to complete the phase is approximately constant. Thus, we can predict that
wheat will take approximately 11 days to emerge at mean temperature of 10 C
½D ¼ 110=ð10 0Þ and 7 days if mean temperature is 15 C ½D ¼ 110=ð15 0Þ.
Base temperatures have physiological and ecological meaning, as they reflect
differences between species and stages, and contribute to the coupling of organisms
in trophic webs. For example, the base temperature for the sowing-emergence
phase is much lower for winter crops than for spring-sown crops (Table 16.1).
Base temperatures normally decline from early to late stages in summer crops,
e.g. for sunflower and soybean, and increase from early to late stages in winter crops
such as wheat. The thermal time model applies not only to plants but also to other
organisms including insects which are – like plants – unable to regulate body
temperature. The base temperature for the emergence of bollworms after
overwintering in the soil is very close to the base temperature for the sowing-
emergence phase of cotton; this coincidence of base temperatures ensures that a
new generation of bollworm emerges in synchrony with a suitable food source.
The thermal time model (Eq. 11.2a, 11.2b) is a simplification of a more complex
response of development to temperature, as it rests on the assumption of a linear
relationship between the rate of development and temperature. However, above the
optimum temperature, the rate of development declines with increasing tempera-
ture, until a maximum temperature (Tm) is reached (Fig. 11.1). For some
phenostages, particularly those related with reproduction, development is also

responsive to vernalisation and photoperiod. In these cases, the thermal time
required to complete the phase is not constant, but influenced by low temperatures
and day length.
11.2.2 Vernalisation and Photoperiod
Annual crops have specific windows of development when grain number, the main
yield component, is most affected by environmental stresses such as frost, heat and
water stress (Fig. 11.2). These windows vary, but are more or less centered on
flowering in most crops. For this reason, mechanisms have evolved that reduce the
probability of coincidence of extreme stress and the most sensitive developmental
stages. These mechanisms are based on two environmental cues: vernalisation and
photoperiod. Flowering time is indeed one of the most important traits for crop
adaptation to particular environments in agricultural systems. Consider for example
wheat in a Mediterranean region. If it flowers too early, a significant frost risk
would reduce seed production in a number of seasons. If it flowers too late, frost risk
is reduced at the expense of increased risk of heat and water stress. Early wheat
varieties introduced to Australia reached flowering at 125 days after sowing, hence
exposing the sensitive reproductive window to high frequency of heat and water
stress. Recognising this problem, breeders selected for shorter season varieties,
which were better adapted to their environments. Where compared under the same
conditions, the time from sowing to anthesis was 119 for cultivars released earlier
than 1950 and 108 for cultivars developed later. Rainfed sunflower in southern
Spain grows on stored soil water that is depleted during the growing season. In these
SOYBEAN
GRAIN NUMBER
WHEAT
MAIZE
RICE
SOWING ANTHESIS PHYSIOLOGICAL

MATURITY
CROP CYCLE
Fig. 11.2 Critical period for grain number determination, the main yield component of annual
crops. Grain number is presented in an arbitrary scale where the vertical line represents 100 %, of the
grain number in unstressed controls. Deviations from this line represent reductions due to stress in
different periods of the crop cycle (Source: Calvi~ no and Monzon (2014) in Sadras and
Calderini and Hsiao TC 1982. In: Drought resistance of crops with emphasis on rice. IRRI, Los
Ba~nos, Manila, Philippines. p 39–52)
environments, hybrids with very long cycle may deplete soil water reserves during
a long vegetative stage thus suffering a stronger water deficit during grain filling
than short-cycle hybrids. To manipulate the timing of key phenological events,
plant breeders make use of fundamental genetic understanding including the
manipulation of vernalisation and photoperiod genes in selecting varieties adapted
to particular environments. To manipulate the timing of key phenological events,
growers combine two practices: cultivar selection and sowing date.
Vernalisation is a response to low temperatures necessary for some plants to
become competent for the transition to the reproductive phase. The plant apex may
sense vernalising temperatures from seed imbibition, throughout the vegetative
phase. Vernalisation requirements are characteristic of temperate crops such as
wheat, barley, Brassicas and field pea. In many of these species, ‘winter’ types
require vernalisation whereas, ‘spring’ types have little or no vernalisation require-
ments. For instance, for winter wheat, temperatures between 0 and 8 C are the most
effective, although vernalisation happens with temperatures up to 15 C.
Vernalisation may be reversed by high temperatures (usually >20 C), in a process
known as ‘devernalization’. In some species, vernalization combines with photo-
period to modulate time of flowering.
Vernalization is also important in horticultural crops. In biennial plants such as
sugar beet and carrot, vernalisation modulates the development of flower buds in
the second year of growth. In proteranthous perennials (i.e. those that flower before
leafing) vernalisation modulates flowering time. In horticulture, the vernalisation
requirement is also known as “chilling hours”, which is the time below a species-
specific base temperature. Understanding vernalisation requirements is important to
determine the geographical limits and risks of growing particular crops. Apple trees
for example, have a high vernalisation requirement, hence they cannot be grown
successfully in warm-winter environments where these requirements are not met.
Almond trees have a relatively low vernalisation requirement, and this implies the
risk of early flowering with subsequent yield losses due to frost. Breeders have
selected horticultural perennials with a broad range of vernalisation requirements to
extend their cropping areas and reduce risks of crop failure.
Virtually in all plant species photoperiod sensitive genotypes can be found, or
rather, genotypes sensitive to the duration of the night. Gardner and Allard classi-
fied annual species into two categories: long-day plants and short-day plants. The
short-day plants accelerate their development (shorten the time to flowering) when
the days are short, while long-day plants develop faster if the days are long.
Small grains (wheat, barley, oats and rye) are long-day species, while maize,
rice, sorghum and soybeans are short-day species. However, within each species
there is often a great variability in sensitivity to photoperiod. In general photoperiod
response is quantitative, i.e. development rate increases or decreases with the
photoperiod but never becomes zero, which would be a qualitative response. By
manipulating photoperiod genes in soybean, varieties have been developed that can
be grown between high latitudes in the northern hemisphere to the tropics; a
classification system of maturity types, with 00 the shortest (90 days) and VIII
the longest season (190 days) shows the wide range of phenological patterns in
soybean.
11.3 Morphological Development
The architecture of the plant is under genetic control, and is modulated by envi-
ronmental factors including temperature, photoperiod, and light-quality. Agronom-
ically, the architecture of the crop is important because it influences traits such as
lodging, harvestability, competition with weeds, responses to herbivory and distri-
bution of light and chemicals in the canopy profile. The introduction of semi-dwarf
genes in rice and wheat has lead to significant improvements in crop production,
and part of the success of these semi-dwarf crops is related to their improved
architecture, which allows for higher nutrient inputs with reduced lodging risk.
The node where the first pod is set is an important trait of grain legumes, as
genotypes with pods too close to the ground cannot be harvested effectively.
Plants have numerous meristems (buds) which can follow one of three fates: they
can remain dormant, they can activate to produce vegetative structures, or they can
become reproductive structures. Different species combine different strategies of
meristem allocation, and these strategies involve trade-offs. For example, the
adaptation of grasses to browsing, and their exploitation in agriculture, is directly
related to their underground, dormant meristems that allow re-growth after grazing.
Plants with profuse branching or tillering are more able to fill gaps originated, for
example, from failures in sowing or damage by pests or diseases. As an example of
trade-off, strong apical dominance, whereby most lateral buds are dormant, favours
growth in height, competition for light, and capacity to recover after herbivory, at
the expense of limited capacity for growth and reproduction, and constrains to
expand into neighbouring gaps.
Interactions between neighbouring plants influence the morphology of individ-
ual plants, and the final architecture of the crop. Part of these interactions are
mediated by the ability of plants to sense changes in the quantity, quality and
direction of light, which in turn trigger developmental responses called photo-
morphogenesis. The main groups of photoreceptors involved in plant photo-
morphogenesis are the red (R)/far-red (FR) light-absorbing phytochromes and the
blue/UV-A light-absorbing cryptochromes and phototropins. As the green tissue of
plants differentially reflects and absorbs light of different wavelengths, plants are
able to detect the presence of neighbours by detecting changes in the spectral
composition of light, and in particular, reductions in the R/FR ratio. Typically, a
shade-avoidant plant responds to neighbours by extending internodes, increasing
stem:leaf ratio, reducing activation of lateral buds, producing more erect shoots and
in some cases advancing the time of flowering. In some weeds, germination can be
triggered by changes in the R/FR ratio associated with soil cultivation. Light signals
interacting with the central circadian oscillator, enable plants to monitor photope-
riod and adjust the timing of the transition from vegetative to reproductive devel-
opment (Sect. 11.2.2).
The successive appearance of plant leaves is an important component of mor-
phological development. In general, the thermal time between the appearance of
two consecutive leaves, known as phyllochron, is constant. For example, wheat has
a phyllochron around 100 C d with a base temperature of 0 C. In sunflower the
phyllochron is around 20 C d with a base temperature of 4 C.
11.4 Growth
Box 11.3 outlines the methods used to quantify crop growth. Crop growth depends
on the capacity of the canopy to capture CO2 and radiation, the capacity of the root
system to capture water and nutrients from soil, and the efficiency of the crop to
transform resources (water, nutrients, radiation, carbon dioxide) into biomass
(Fig. 11.3). The right half in Fig. 11.3 highlight how environmental factors, such
as ambient temperature or soil salinity, modulate the rate of capture of resources
and the efficiency in the transformation of resources in plant biomass. Other
chapters deal in detail with the capture and efficiency in the use of radiation
(Chap. 13), water (Chap. 14) and nutrients (Chaps. 24, 25, and 26).
The capture and efficiency in the use of resources changes with the stage of
phenological development. The growth of a typical annual crop is characterised
with a sigmoid curve (Fig. 11.4) with three phases. First, in a lag-phase, plants grow
slowly, as they mostly depend on seed reserves, whereas small root and canopy
systems constrain their capacity to capture resources. Many practices (e.g. sowing
date, fertiliser) seek to reduce the duration of this lag-phase, also known as “period
lost to growth”. Second, the growth increases rapidly to reach a linear phase when
sufficiently large canopy and root system allow for a high capacity to capture
resources. Third, crop growth slows down as both canopy and root systems age,
entering a senescence phase in parallel to the accumulation of carbon and nitrogen
in reproductive organs. The senescence of leaves and roots is genetically driven in a
Fig. 11.3 Crop growth RESOURCES REGULATORS

depends on the ability of
crops to capture above-
Radiation, CO2
ground and soil resources, - Day length
and on the capacity of crops Temperature
AERIAL -
to transform these resources
- Light quality
into biomass (Adapted from
- Biotic factors
Sadras and McDonald
2012)
Crop biomass
SOIL - Temperature
- Salinity, pH
- Compaction
- Biotic factors
Nutrients, H2O
14000
Leaf
12000
Grain
10000
BIOMASS (kg/ha)
8000
6000
4000
2000
0
0 20 40 60 80 100 120 140
TIME (days after planting)
Fig. 11.4 Time course of biomass of leaves and grain of a maize crop in Florida (Data taken from
DSSAT 4.6, experiment UFGA8201MZ)
Fig. 11.5 Crop biomass

increases with increasing B1
capture of resources,
primarily water, radiation Unstressed
and nitrogen. Maximum
CROP BIOMASS
crop biomass is shown as B2

B1. Stress reduces crop
growth by reducing capture
of resources (horizontal B3
arrow) resulting in B2,
reducing production per
unit resource (vertical
arrow) or both, which leads
to B3 Stressed
CROP CAPTURE OF RESOURCE

(WATER, RADIATION OR NITROGEN)
process known as monocarpic senescence, whereas stresses such as shortage of

water or nutrients may accelerate the process.
Figure 11.5 illustrates the relationship between crop growth and capture of
resources in crops with favorable and stressful conditions. As the season progresses
and roots and canopies expand, the crop captures more soil and above-ground
resources. A straight line represents increasing growth with increasing resource
capture. Stresses such as deficit of nutrients or soil compaction reduce growth
through two processes: reducing the amount of resources captured by the crop
(horizontal arrow in Fig. 11.5, and b) reducing the efficiency in the use of resources.
The vertical arrow in Fig. 11.5 indicates the reduction in growth for the same
amount of resource captured; this means lower efficiency.
In general shortage of resources (water or nutrient deficits) and soil constraints
(compaction, salinity) reduce crop growth by reducing capture of resources, rather
than affecting the efficiency in the use of resources.
Ambient temperature influences growth directly, by changing the rate of pro-
cesses such as cell division, leaf expansion and crop photosynthesis, and indirectly
by affecting phenological development and the duration of key phenophases, as
discussed before. Within agronomically sensible ranges, the main effect of temper-
ature on crop production is related to the modulation of phenological development.
In temperate environments, late sowing shifts the crop cycle to warmer conditions,
development proceeds faster, the period available to capture resources and growth
is reduced, and biomass at maturity is normally lower. Figure 11.6 illustrates the
interplay between development and growth in faba bean crops sown between
October (autumn) and early May (spring) in Lugo, Spain. As the sowing is delayed,
both temperature and photoperiod increase. This shortens the phenophases of the
crop, resulting in a reduction of crop cycle from 209 to 87 days. With shorter cycle
duration, the peak leaf area is reduced, the amount of radiation captured by the crop
is reduced and the final production of biomass is also reduced. Hence, the effects of
temperature and photoperiod on development (cycle length) have a dominant role
in seasonal growth. Of interest, the first sowing date does not conform to this
pattern. For the earliest sowing, the crop has the longest cycle duration and the
highest capture of radiation as expected; therefore it should also have the highest
biomass. However, it has the lowest biomass. The explanation is that the extremely
1800 250
1600
Intercepted PAR, Biomass, RUE
1400 200
1200
Cycle duration
150
1000
800
100
600
400 50
Intercepted PAR (MJ) Biomass (g/m2)
200
RUE (mg/MJ) Cycle duration (days)
0 0
10/3 11/22 1/11 3/2 4/21 6/10
Sowing date
Fig. 11.6 Effect of sowing date on crop duration (time sowing maturity), intercepted PAR, RUE
and biomass production of Vicia faba (Adapted from Confalone et al. (2010) Field Crops Res
115, 140–148)
low temperature in the earliest sowing reduced the photosynthetic efficiency of the
crop. In this case, the physiological response (photosynthesis) was stronger than the
developmental response.
Box 11.3 Quantification of Crop Growth

Depending on the objectives of measurements, we could be interested in
describing growth in terms of crop height, leaf surface area, fruit volume, or
grain mass. As crop growth depends on capture of resources, and this is in
turn related to the size of the canopy and root system, we often use the leaf
area index (LAI, m2 leaf/m2 ground) to measure the size of the canopy
and the rooting depth and density characterized as Root Length Density
(Lv, m root/m3 soil) to quantify the size of the root system. LAI is the ratio
of leaf area (assuming single-sided leaves) and ground area, and Lv is the
length of roots per unit of soil volume. For many agronomic applications,
shoot mass is measured by taking crop samples (e.g. 1 m2 of crop cut to
ground level), which are dried to constant weight to express the dry matter in
g/m2 or kg/ha; this measure of dry matter is also called biomass. Periodic
sampling of biomass combined with periodic measurements of radiation inter-
ception, nutrient uptake and evapotranspiration allows calculating the effi-
ciency in the use of radiation, nutrients and water as illustrated in Fig. 11.5.
Indirect methods are also used for quantifying biomass or LAI. For trees
empirical relationships between biomass and trunk diameter have been
widely used. Transmittance of PAR or reflectance of radiation in different
wavelengths (e.g. red and far red) (see Chap. 3) may be related to LAI and are
the base of most indirect methods for non-destructive measurement of LAI.
Appendix 11.1
Durations of phases of growth cycle (FAO method) for different crop species and
climatic areas. A: sowing to 20 % ground cover, B: 20 % to 80 % ground cover, C:
80 % ground cover to start of leaf senescence, D: start of senescence to harvest
Duration of crop stage (days)

Crop species Climate Sowing date A B C D Total
Horticultural crops
Artichoke (year M April 40 40 250 30 360
1)
Artichoke (year M May 20 25 250 30 325
2)
Asparagus M-warm Feb 50 30 100 50 230
winter
Asparagus M Feb 90 30 200 45 365
Beets (table) M Apr/May 15 25 20 10 70
(continued)

Broccoli A Sept 35 45 40 15 135
Brussel sprouts M Feb/April 20–25 30–35 20–25 10 80–95
Cabbage A, M Sept 40 60 50 15 165
Cabagge M Feb/April 20–25 30–35 20–25 10 80–95
Melon M-warm Jan 30 45 35 10 120
(Cantaloupe) winter
Melon M Aug 10 60 25 25 120
(Cantaloupe)
Carrots A Oct/Jan 20 30 40 20 110
Carrots M Feb/Mar 30 40 60 20 150
Carrots A Oct 30 50 90 30 200
Cauliflower A Sept 35 50 40 15 140
Celery SA Oct/Jan 25–30 40–55 95–105 20 180–210
Celery M April 25 40 45 15 125
Cucumber A May–August 20 30 40 15 105
Cucumber A-warm Nov/Feb 25 35 50 20 130
winter
Egg plant A October 30 40 40 20 130
Egg plant M May/June 30 45 40 25 40
Lettuce M April 20 30 15 10 75
Lettuce M Nov/Jan 30 40 25 10 105
Lettuce A Oct/Nov 25 35 30 10 100
Lettuce M Feb 35 50 45 10 140
Melon M March/May 25–30 30–35 40–50 20–30 120–140
Melon A Aug 15 40 65 15 135
Melon A Dec/Jan 30 45 65 20 160
Onion (dry M April 15 25 70 40 150
harvest)
Onion (dry A Oct/Jan 20 35 110 45 210
harvest)
Onion (green M April/May 25 30 10 5 70
harvest)
Onion (green A October 20 45 20 10 95
harvest)
Pepper T&M April/June 30 35 40 20 125
Pepper A October 30 40 110 30 210
Pumpkin, Winter M Mar, Aug 20 30 30 20 100
squash
Pumpkin, Winter T June 25 35 35 25 120
squash
Radish M, T Mar/Apr 5 10 15 5 35
Radish A Winter 10 10 15 5 40
Spinach M Apr; Sep/Oct 20 20 20 5 65
Spinach A November 20 30 40 10 100
(continued)

Squash, Zucchini M, A Apr; Dec. 25 35 25 15 100
Squash, Zucchini M,T May/June 20 30 25 15 90
Tomato A January 25–30 40 40–70 30 135–155
Tomato M Apr/May 35 40 50 30 155
Tomato A Oct/Nov 35 45 70 30 180
Water melon M April 20 30 30 30 110
Water melon A May/Aug 10 20 20 30 80
Roots, tubers and bulbs
Cassava (year 1) T Rainy season 20 40 90 60 210
Cassava (year 2) T 150 40 110 60 360
Potato SA Jan/Nov 25 30 40 30 125
Potato C April/May 25–30 30–35 45–50 30 130–145
Potato C Apr/May 45 30 70 20 165
Potato A Dec 30 35 50 25 140
Sweet potato M April 20 30 60 40 150
Sweet potato T Rainy seas. 15 30 50 30 125
Sugarbeet M March/April 25–30 35–45 50–80 20–50 160–180
Sugarbeet M Oct/Nov 45 75 80 30 230
Sugarbeet A Sept/Nov 25–35 60–65 70–100 40–65 205–255
Sugarbeet C April 50 40 50 40 180
Legumes
Beans M Feb/Mar 20 30 30 10 90
(Phaseoulus)
(green)
Beans M, A Aug/Sep 15 25 25 10 75
(Phaseoulus)
(green)
Beans C May/June 20–25 25–30 30–40 20 100–110
(Phaseoulus) (dry
seed)
Beans M June 15 25 35 20 95
(Phaseoulus) (dry
seed)
Faba bean, broad C, M Mar/May 15–20 25–30 35 15 90–100
bean (green)
Faba bean, broad C, M Oct 90 45 40 0 175
bean (green)
Faba bean, broad C, M Nov 90 45 40 60 235
bean (dry seed)
Green gram, M March 20 30 30 20 110
cowpeas
Groundnut T Dry season 25 35 45 25 130
(peanut)
Groundnut C, high Spring 35 35 35 35 140
(peanut) latitude
(continued)

Groundnut M May/June 35 45 35 25 140
(peanut)
Lentil C April 20 30 60 40 150
Lentil A Oct/Nov 25 35 70 40 170
Peas C May 15 25 35 15 90
Peas C, M Nov, Mar– 20–35 25–30 30–35 15–20 100–110
Apr
Soybeans T Dec 15 15 40 15 85
Soybeans C, high May 20 25–35 60–75 25–30 140–150
latitude
Sugar, oil and fiber crops
Castor beans SA March 25 40 65 50 180
Castor beans T Nov. 20 40 50 25 135
Cotton M, A Mar–May 30 50 50–65 45–55 180–195
(M), Sept.
(A)
Cotton A Mar 45 90 45 45 225
Flax C April 25 35 50 40 150
Flax A October 30 40 100 50 220
Hops C April 25 40 80 10 155
Safflower M, A, March/April 20–25 35 45–55 25–40 125–145
SA
Safflower A Oct/Nov 35 55 60 40 190
Sesame C June 20 30 40 20 100
Sugarcane, virgin T, low 35 60 190 120 405
latitude
Sugarcane, virgin T 50 70 220 140 480
Sugarcane, virgin Pacific 75 105 330 210 720
Sugarcane, ratoon T, low 25 70 135 50 280
latitude
Sugarcane, ratoon T 30 50 180 60 320
Sugarcane, ratoon Pacific 35 105 210 70 420
Sunflower M April/May 25 35 45 25 130
Sunflower M warm Feb 45 40 60 25 170
winter
Cereals and pseudocereals
Barley, oats, Central November 15 25 50 30 120
wheat (spring India
types)
Barley, oats, Mid March/Apr 20–40 25–30 40–60 20–30 130–135
wheat (spring latitude
types)
Barley, oats, East July 15 30 65 40 150
wheat (spring Africa
types)
(continued)

Barley, oats, C Nov 40 60 60 40 200
wheat (spring
types)
Barley, oats, M Dec 20 50 60 30 160
wheat (spring
types)
Winter wheat M warm December 20 60 70 30 180
winter
Winter wheat M November 30 140 40 30 240
Winter wheat C October 160 75 75 25 335
Grains (small) M March–April 20 30 60 40 150
Grains (small) A Oct/Nov 25 35 65 40 165
Maize (grain) A Dec/Jan 25 40 45 30 140
Maize (grain) T June 20 35 40 30 125
Maize (grain) C (dry, October 20 35 40 30 125
cool)
Maize (grain) M, C March–April 30 40 50 30–50 150–170
Maize (sweet) T March 20 20 30 10 80
Maize (sweet) M May/June 20 25 25 10 80
Maize (sweet) A Oct/Dec 20 30 40 10 100
Maize (sweet) C April 30 30 30 103 110
Maize (sweet) M warm Jan 20 40 70 10 140
winter
Millet A June 15 25 40 25 105
Millet C April 20 30 55 35 140
Sorghum (grain) C, M May/June 20 35 40 30 130
Sorghum (grain) A Mar/April 20 35 45 30 140
Rice T, M Dec; May 30 30 60 30 150
Rice T May 30 30 80 40 180
Forages
Alfalfa* frost free 10 30 – – Variable
period
Alfalfa* M Mar 5 10 10 5 30
Alfalfa* C Jun 5 20 10 10 45
Bermuda (for A March 10 25 35 35 105
seed)
Bermuda (for A – 10 15 75 35 135
hay)
Grass Pasture Frost 10 20 – – Variable
free
period
Sudangrass (first A Apr 25 25 15 10 75
cutting cycle)
Sudangrass (other A June 3 15 12 7 37
cutting cycles)
(continued)

Fruit trees, vines and shrubs
Banana (year 1) M Mar 120 90 120 60 390
Banana (year 2) M Feb 120 60 180 5 365
Citrus M Jan 60 90 120 95 365
Deciduous C high March 20 70 90 30 210
orchard latitude
Deciduous M,C low March 20–30 50–70 120–130 30 240–270
orchard latitude
Grapes Low April 20 40 120 60 240
latitude
Grapes M, C mid Mar/April 20–30 50–60 40–75 60–80 205–210
latitude
Grapes High May 20 50 90 20 180
latitude
Olives M March 30 90 60 95 365
Pineapple T 60 120 600 10 790
Pistachios M Feb 20 60 30 40 150
Walnut C high April 20 10 130 30 190
lat
Adapted from Doorenbos and Pruitt (1977) and Allen et al. (1998). This Table should be used with
caution as the actual duration of phases may change for different regions, cultivars and weather
conditions of each year
*For the first cut cycle use durations twice the values shown
Cimates: M Mediterranean, A arid, SA semi-arid, C continental, T tropical
Bibliography
Hay, R., & Porter, J. (2006). The physiology of crop yield, 2nd ed. Oxford: Wiley-Blackwell.
Sadras, V. O., & Calderini, D. F. (2014). Crop physiology (2nd ed.). San Diego: Academic Press.
Sadras, V. O., & McDonald, G. K. (2012). Water use efficiency of grain crops in Australia:
Principles, benchmarks and management. Canberra: Grains Research and Development
Corporation.
Chapter 12
Plant Density and Competition
Francisco J. Villalobos, Victor O. Sadras, and Elias Fereres
Abstract Crops respond to planting density modifying the characteristics of indi-

vidual plants by changing the number and size of their organs. The density response
can be described mathematically by the “Law of Reciprocal Yield”. Under very
high density mortality of individuals occurs and is often more pronounced when
environmental conditions are more suitable (e.g. high fertility). Yield-density
curves can be asymptotic or parabolic, although the latter may reflect the existence
of an additional limiting factor (e.g. water or nutrients). In general the spatial
variability in planting density leads to yield losses which are higher when the
yield-density response is a parabolic curve.
12.1 Introduction
Plant population dynamics studies the temporal variation in the number of individ-
uals and their attributes. These aspects are important to understand the productivity
of crops, especially to evaluate the effect of planting density on yield. Population
dynamics also helps us understand the weed-crop competition and the process of
plant mortality, which may be critical for the establishment of annual crops or
pasture maintenance.

V.O. Sadras

Agriculture, DOI 10.1007/978-3-319-46116-8_12
12.2 Density and Competition
A crop is a plant population with individuals of the same genotype and of similar
age. The availability of resources changes in time and space and may limit crop
growth and cause competition between neighboring plants. Unlike animals, higher
plants show great plasticity in their growth and in their form, in response to the
stress imposed by competition. Thus, the structure of individual plants is set to
respond to competition stress by varying the rate of formation, growth or mortality
of its organs (leaves, branches, stems, fruits, roots, etc.). The response may involve
changes in the size of the individuals, in their shape or in the number of individuals.
The growth rate of a plant population is proportional to density in the early
stages of development. This proportionality is later reduced as competition for
resources among plants increases, and usually leads to a phase when crop growth
rate is independent of the density. The higher the initial density, the sooner the
competition for resources begins. Variations in initial density are therefore largely
offset by variations in growth rates of individual plants. This has been verified for
many species and has been called “law of constant final yield.” This is true above a
minimum plant population so there is enough opportunity to exploit all the
resources. In other words, in its early stages of growth from seed, the biomass of
a crop depends on the number of plants present, but over time, the supply of
resources starts to control the rate of growth of individuals, until finally it is the
limiting factor of productivity, regardless of the density. The population behaves as
an integrated system in which the behavior of the individual plant is subordinated to
the behavior of the population.
Any factor that reduces the rate of growth of the plants results in a delayed onset
of competition and a reduction in its intensity. The relationship between yield per
plant and density is often expressed by the following equation (called “reciprocal
yield law”):
1
W¼ ð12:1Þ
b1 þ b 2 D p
where W is the dry mass per plant (g), Dp is the planting density (plants m2) and b1
and b2 are empirical coefficients. Crop biomass (B, g m2) is the product of mass
per plant and planting density, and yield (Y) is the product of biomass and Harvest
Index (Chap. 13) so:
HI Dp
Y¼ ð12:2Þ
b1 þ b2 Dp
The coefficient b2 represents the inverse of the crop biomass (B) when the density is
very high. If Dp tends to infinite in Eq. 12.2, B b21
Coefficient b1 represents the inverse of W when competition is absent, i.e. for
very low density. If Dp is zero in Eq. 12.1, W b11
12 Plant Density and Competition 161
Example 12.1 The maximum yield of sunflower for isolated plants of a parti-
cular cultivar is 500 g, and the maximum yield is 500 g m2 when the density is
very high. What is the expected yield if planting density is 5 plants m2?
B 1=b2 when Dp is high ! b2 ¼ 1=B ¼ 1=500 ¼ 0:002 m2 =g

1=W b1 when Dp is very small ! b1 ¼ 1=W ¼ 1=500 ¼ 0:002 plants=g
5
Yield ¼ ¼ 417 g m2
0:002 þ 0:002 5
12.3 Variability Between Plants and Hierarchy
The frequency distribution of weight per plant in a population under density stress
is skewed, i.e. asymmetrical. The bias increases with both time and population
density, as illustrated in Fig. 12.1 for sunflower. In a population in competition we
thus find a large number of small individuals (low biomass per plant) and fewer
large individuals. The place of an individual in the hierarchy of the population is
determined primarily in the early stages of development. It has been shown
experimentally that the amount of biomass produced by an individual in a popula-
tion under competition is very closely related to the relative order of its appearance
(emergence) in the population. The advantage of an early appearance must be
related to an increased use of resources and the corresponding deprivation of
resources for individuals that appear later. This implies that a likely source of
0.5
10 plants / m2
RELATIVE FRECQUENCY
0.4 2 plants / m2
0.3
0.2
0.1
0.0
8 10 12 14 16 18 20
STEM DIAMETER (mm)
Fig. 12.1 Frequency distributions of stem diameter of sunflower plants at plant densities of 10 or
2 plants m2. Cordoba (Spain), 1994
variability of plant mass in the field is variability in time to emergence which in turn
depends on the variation of soil properties (water content, thermal properties,
compaction) and the sowing method.
In addition to heterogeneity of seedling emergence, herbivory, diseases and
other sources of damage (e.g. hail) are agronomically important sources of crop
heterogeneity. For example, insects feeding on the growing meristem of cotton
slow down plant growth, as it takes time for activation of axillary buds that would
re-initiate growth. This hiatus in growth of the damaged plant may favor the
growth of undamaged neighbor plants. The yield per unit ground area of such
heterogeneously damaged crops can be the same (compensation), higher (over-
compensation) or smaller than in undamaged crops; the outcome depends on the
gain in growth of the undamaged plant relative to the reduction of yield of the
damaged neighbor. For instance, researchers in the UK observed how healthy
potato plants next to diseased plants grew much bigger than those in uniformly
healthy crops.
Insects, diseases or other agents that kill seedlings cause “gaps” in the crop. In
the absence of compensatory capacity (when plant loss occurs very late or when it
originates large gaps) yield would be reduced in proportion to the reduction of the
plant stand. The relationship between yield and stand loss, however, demonstrates a
compensatory mechanism which relates to the “relaxation” of competition and
depends on the spatial distribution of plant loss and the time when it occurs
(Fig. 12.2). In conclusion, crop heterogeneity often but not always reduces yield;
the impact of heterogeneity depends on the size of the hierarchy or gap, their spatial
distribution, the ability of plants for compensatory growth (e.g. tillering wheat
100
90
80
% MAXIMUM YIELD
70
60
50
40
30
20 Yield 1
10 Yield 2
0
0 20 40 60 80 100
% STAND REDUCTION
Fig. 12.2 Effects of plant stand loss on yield. The continuous line corresponds to stand loss
occurring very late or when plant loss occurs in large patches so no compensatory growth is
possible. The dashed line represents very early and random plant loss so maximum compensatory
growth occurs
>uniculm sunflower), availability of resources and the time available for compen-
satory growth.
12.4 Density and Mortality
The high density tends to increase the risk of death of the individuals in the
population although there are some examples of the opposed effect. The risk of
mortality that increases with the density has regulatory properties, acting as a
negative feedback on the size of the population. Various studies on “self-elimina-
tion” (density stress induced mortality) have shown that it occurs at high but not at
low density, starts sooner with higher density and depends on environmental
conditions.
In the years 1920–1930 Suskatschew studied the dynamics of self-elimination in
populations of spruce near Saint Petersburg (Russia), finding that final plant density
was higher in poor and shallow soils. In deeper soils he found lower densities of
larger trees. This author then conducted an experiment with an annual plant
(Matricaria inodora) using two levels of fertility and two densities, checking that
mortality was higher with the highest density, and that the risk of death increased
with higher fertility. This corroborated his observations in spruce forests. Fertili-
zation apparently increased the growth rate of individuals thereby increasing the
density stress and thus the mortality rate.
12.5 Mechanisms of Plant Competition
The density of plants in a crop determines the occurrence of numerous processes of

interference between individual plants. As plant density is varied the environment
of each plant is altered in terms of light intensity and quality and availability of
resources (water, nutrients).
Thus, as plant density is higher, intercepted radiation and the availability of
water and nutrients are reduced for each individual, which limits their ability to
grow. Light quality is changed fundamentally in the relationship between red
(670 nm) and far red radiation (760 nm) (ratio R: FR). On average, sunlight has
a ratio R: FR of 1.15. As red is absorbed by pigments, light transmitted or
reflected by vegetation has lower R: FR. For example, R:FR below crops may
range from 0.1 to 0.5. As the density increases the ratio R:FR is reduced, and
this reduction is detected by phytochromes, inducing morphological changes in
many species (increased height growth, reduction in the formation of side
branches or stems).
Fig. 12.3 Time course of
DRY MATTER PARTITIONING

dry matter partitioning for
leaves in sunflower for low
and high planting densities
TO LEAVES
low density
high density
TIME
The main crop responses to increased density are:

– Reduction of expansive growth and mass (per plant). Plant leaf area and thus
radiation interception per plant are reduced.
– Increase or decrease in harvest index (in some species): For very low density,
biomass per plant may be very large while seed growth is limited by the number
of seeds and potential seed growth rate, implying a fall in HI. In other cases
(e.g. maize) very high densities increase the percentage of sterile plants.
– Reduction of the number of seeds per plant and/or mass of single seeds.
– Changes in the distribution of dry matter among plant organs: increased alloca-
tion to stem and reduced allocation to leaves (Fig. 12.3). In general height
increases while stem diameter decreases, which leads to a notable increase of
its slenderness ratio and therefore a higher risk of lodging in adverse situations
(for instance, strong winds).
– Depressed branching in dicots (e.g. soybean) and tillering in cereals.
– Acceleration of leaf senescence: In plants under high density leaf senescence
starts sooner and the rate of senescence is higher, which seems to occur in
response to very low levels of radiation.
– Changes in crop quality: In some cases quality may improve with high density
(e.g. percentage of oil in sunflower seed) and in other cases it may decline
(e.g. sunflower seed size for direct consumption). In general the size of harvest-
able organs is reduced as density increases (grains, tubers, bulbs, etc.).
12.6 Yield and Planting Density
Throughout the twentieth century great attention was paid to the relationships
between plant density and crop yield. These relationships are important from the
practical standpoint of defining if there is an optimum density.
Two types of relationships between yield and density are usually found: asymp-
totic and parabolic (Fig. 12.4). In the first type, yield increases with density, reaches
a plateau and does not decrease for very high densities. This case is also found when
Fig. 12.4 Generic

parabolic and asymptotic parabolic
response curves of yield
versus density
hyperbolic
YIELD
PLANT DENSITY
Fig. 12.5 Changes in

high fertility
parabolic response curves of
yield versus density when
resource availability is
changed low fertility
YIELD
PLANT DENSITY
we plot biomass versus density. In the parabolic case, yield reaches a maximum at a
given density and decreases for densities above or below.
The biomass production of any crop follows an asymptotic relationship with
density. This is the response predicted by the Reciprocal Yield Law (Eq. 12.2). In
crops harvested for their seed it used to be common to observe a decrease in yield
for high densities (parabolic response), which implies that the HI is reduced. This
was due to a direct effect of density (e.g. barrenness in maize) or as a result of
another limitation of resources such as water or nutrients. Thus, in situations of
water deficit, the highest densities are at increased risk of not having sufficient
water during grain filling. Evidence that a parabolic curve is the product of a
limitation other than density may come from the observation that the density for
maximum yield increases with increasing water or nutrient availability (Fig. 12.5).
Additionally, within a species we can find various yield-density curves for the
different cultivars, especially if they differ in cycle length. Very short cycles
produce less biomass and yield at low densities. The maximum yield is achieved
at higher density if the cycle is shorter than if it is long. A long cycle can fully
exploit the available resources with low densities. It compensates the low plant
density with a longer vegetative period which implies a higher growth potential for
the single plants. This is illustrated in Fig. 12.6 for the case of two cultivars
Fig. 12.6 Response of

sunflower yield to density long cultivar
for short and long cycle
genotypes
YIELD
short cultivar
PLANT DENSITY
of sunflower in Cordoba, Spain. In winter cereals, plant breeders have incorporated

substantial plasticity in the response to density, and yields are the same over a wide
range of densities provided that resources are not limiting (fully irrigated and
fertilized conditions). Even in maize, modern varieties are quite resistant to
barreness and yields do not decline until very high densities are achieved
(>150,000 plants/ha).
Using the Reciprocal Yield Law Eq. 12.1 we can deduce at what density a
particular fraction of maximum biomass (r ¼ B/Bmax) is achieved:
r b1
D p ðr Þ ¼ ð12:3Þ
ð1 r Þb2
Example 12.2 Two cultivars of sunflower differ in cycle. Under very low
density the shorter cultivar produces 360 g/plant and the longer 1400 g/plant.
In this environment maximum biomass production is 1200 g m2 and 1600 g
m2, for the short and the long cultivar, respectively. We will calculate the
densities required for these varieties to reach 90 % of maximum biomass.
For both varieties:
Short: b2 ¼ 1=1200 ¼ 8:33 104 m2 =g

Long: b2 ¼ 1=1600 ¼ 6:25 104 m2 =g
Short cycle cultivar:
1=W b1 when Dp is very small ! b1 ¼ 1=W ¼ 1=360

¼ 2:78 103 plants=g
(continued)

r b1 0:9 2:78 103
D p ðr Þ ¼ ¼ ¼ 30 plants m2
ð1 r Þb2 ð1 0:9Þ 8:33 104
Long cycle cultivar:
1=W b1 when Dp is very small ! b1 ¼ 1=W ¼ 1=1400

¼ 7:14 104 plants=g
rb1 0:9 7:14 104
Dp ðrÞ ¼ ¼ ¼ 10 plants m2
ð1 rÞb2 ð1 0:9Þ6:25 104
We see that to achieve yields close to the maximum a much higher density
is required for the short cultivar.
The yield-density relationships obtained experimentally should be used with

caution because they depend on the limitations of water and nutrients and the
cultivar considered. In any case it should be noted that there will be a minimum
density which should be increased for short growing cycles. Furthermore, if water
or nutrients are scarce, we must avoid high densities that could reduce the harvest-
able fraction of biomass and therefore yield. The high density may cause other
undesirable effects such as increased risk of crop lodging by wind which may
decrease yield dramatically.
Parabolic responses to density are also observed for some horticultural crops where
the product price is closely related to the size of the harvestable organ (e.g. garlic,
onion, carrot). High densities lead to smaller and therefore lower selling price, leading
to yield-density curves of parabolic type, when we express the yield in income/ha.
The relationship between density and yield mentioned above are obtained in
experimental plots in which the crop density is uniform across the plot. However in
a commercial plot, plants are not distributed evenly across the field. There is a
spatial variability in density so that there are areas where the density is high and
areas where it is low. This may be due to variability in soil conditions (compaction,
initial water), the presence of pests and soil diseases or poor seed distribution at
sowing.
In general we may expect that the larger the variability in the density of plants
the larger the decrease in yield. However, it depends strongly on the size of gaps
and the possibility for compensation, as indicated in Sect. 12.3. The negative effect
of variability should be more pronounced when the response is parabolic as both
lower and higher densities reduce yield below the optimum. In this case if we
cannot avoid a very high variability (soil or machinery problems, for example) we
may reduce planting density below the theoretical optimum.
Bibliography
Harper, J. L. (1977). Population biology of plants. London/New York: Academic Press.
Chapter 13
Radiation Interception, Radiation Use
Efficiency and Crop Productivity
Abstract Photosynthesis is the set of processes whereby radiant energy is

converted and stored as chemical energy in most plants, algae and cyanobacteria.
This process depends on radiation, temperature and CO2 concentration. The max-
imum efficiency of the process is 6 % but it is usually well below. The leaf-level
photosynthesis can be described mathematically, and this analysis can be extended
to the calculation of crop photosynthesis, as a function of its leaf area index, the
coefficient of extinction as a shortcut to represent canopy architecture, and leaf
photosynthetic parameters.
The respiration of the crop can be decomposed into a maintenance component
that is dependent on biomass and temperature, and a growth component which is
proportional to gross photosynthesis.
Crop yield can be expressed as the product of three factors, the amount of
intercepted radiation, radiation use efficiency (RUE) and harvest index (HI). RUE
is smaller for C3 than C4 crops, and is smaller in crops with oil- and protein-rich
seed in comparison to cereals. Radiation interception depends on incident radiation,
leaf area index and the extinction coefficient accounting for canopy architecture.
The HI depends on the species and its use. The main cause of genetic yield
improvements in the past has been the increase in HI, which has been a remarkable
success in the case of cereals; maize is the main expectation to this trend. In some
cases, such as high-yielding wheat in northern Europe, HI is reaching a biophysical
limit, hence further improvements in yield would require increasing RUE and
biomass.
V.O. Sadras (*)


Agriculture, DOI 10.1007/978-3-319-46116-8_13
13.1 Introduction
The input of energy to ecosystems is based on the process of photosynthesis by

which sunlight is converted into chemical energy. Photosynthesis is the primary
process in producing carbon compounds necessary for the construction and main-
tenance of crop biomass. Here, the focus is leaf-level photosynthesis with a brief
scaling-up exercise to crop level photosynthesis, which links with crop traits
developed in Chap. 3. When we measure the rate of CO2 exchange of a single
leaf, this rate is the net photosynthesis (Pn), which results from the total or gross
photosynthesis (Pg) and respiration losses (R). We use a convention that fluxes from
air to leaf are positive, whereas fluxes from leaf to air are negative. Emphasis is
placed on how radiation, CO2 and temperature modulate leaf photosynthesis. Other
factors affecting photosynthesis, such as plant water status and nitrogen nutrition
are discussed in other chapters.
In the second part of this chapter we discuss the relation between biomass
production and the amount of radiation absorbed by photosynthetically active
tissues, i.e. the efficiency in the use of radiation for biomass production. Absorbed
radiation is closely related to intercepted radiation, i.e. the difference between
incoming radiation and that reaching the soil surface (Chap. 3). Radiation use
efficiency varies between crop types in relation to both photosynthetic metabolism
(C4-C3) and seed composition (cereals, legumes and oilseed crops); there is also
moderate variation between cultivars that can be exploited in plant breeding. Plant
age, source-sink ratio, dry matter and nitrogen allocation also affect crop radiation
use efficiency. Nitrogen, water and temperature are major environmental factors
with effects on both radiation interception and radiation use efficiency.
13.2 Leaf Level Photosynthesis
Most (85–90 %) of the dry matter accumulated in a crop derives from photosyn-
thesis, which can be decomposed into the following three processes:
(a) Diffusion of CO2 from the atmosphere to the chloroplasts, following the
concentration gradient:
Pn ¼ gsc ðCa Ci Þ ð13:1Þ
where Pn (μmol m2 s1) is the net flux of CO2 entering the leaf, gsc is stomatal
conductance for CO2 (molm2s1) and Ca and Ci are the concentrations
(μmolmol1) of CO2 in the air surrounding the leaf and in the substomatal
cavity, respectively.
(b) Absorption of light by the photosynthetic pigments and photolysis of water. The
amount of radiation absorbed depends on the concentration of pigments, mostly
chlorophylls, present in the chloroplasts. In this stage, O2 is released and energy
13 Radiation Interception, Radiation Use Efficiency and Crop Productivity 171
compounds (ATP and NADPH) are generated. This process does not depend on
the temperature or the concentration of CO2.
(c) Reduction of CO2 using the compounds generated in the photolysis of water.
Between 8 and 12 light quanta are required for each molecule of CO2 reduced.
The reduction can occur in the dark and is very sensitive to temperature.
In summary, the photosynthesis of a leaf of a healthy, well watered and well
fertilized plant depends on irradiance, CO2 concentration and temperature.
The energy efficiency of photosynthesis, i.e. the ratio of energy stored in
chemical form and incoming solar radiation, has a maximum value around 6 %,
but actual efficiency in agricultural crops usually does not exceed 2–3 %. We can
calculate the relative importance of photosynthesis in the energy balance equation
(Chap. 7) as follows. Net radiation (Rn) above a crop is equivalent to 60–80 % of
solar radiation (Rs) depending mostly on cloud cover. On a clear day we can assume
Rn ¼ 0.6 Rs so that the energy stored with 6 % of efficiency represents 10 % of net
radiation. However, the energy spent in photosynthesis also includes a fraction that
is lost by respiration. If that fraction is one third, then 15 % of net radiation may be
spent in photosynthesis. Therefore the common assumption of neglecting energy
use in photosynthesis may be wrong in particular when productivity is high
(e.g. greenhouse crops) and irradiance is low.
13.3 Plant Types According to Photosynthesis Mechanisms
Higher plants have developed three different photosynthetic systems (C3, C4 and
CAM) that have distinct chemical and anatomical features. Terrestrial plants
evolved from algae and initially were all C3. Subsequently there has been a shift
towards C4 and CAM systems. Agricultural species and natural flora present mostly
the C3 system. Few but important cultivated species have the C4 system (maize,
sorghum, millet, sugar cane, and some tropical grasses such as Paspalum spp),
whereas the CAM system is rarer and less important in crops (agave, pineapple). C3
plants originate mostly in high to intermediate latitudes and high altitude, whereas
C4 are more typical of subtropical to tropical regions. C4 photosynthesis is a
complex evolutionary trait that resulted from a major reorganization of leaf anat-
omy and metabolism leading to a CO2-concentrating mechanism that counteracts
the inhibitory effects of low atmospheric CO2 on photosynthesis. The C4 pathway
evolved independently at least 66 times within the past 35 million years. The main
features of the three systems are as follows:
(a) C3 plants. The first compound formed in the process is phosphoglyceric acid
(three carbon atoms) by combination of ribulose diphosphate (5C) with CO2.
The enzyme responsible is the ribulose-bisphosphate carboxylase (Rubisco).
Although primarily serving for carboxylation, Rubisco can also act as
oxygenase. Thus, in the presence of light, O2competes with CO2 at the enzyme
active sites which leads to a loss of CO2 (photorespiration). In addition to its
enzymatic role, the large amount of Rubisco in plants means it plays a major
role as reserve of reduced nitrogen. This is most evident during the grain filling
and senescence of annual crops, where a significant part of the nitrogen, stored
as Rubisco, is transported and contributes to the protein content in grain.
(b) C4 plants. The first compound formed in the process is oxaloacetic acid (four
carbons) by combining phosphoenol pyruvate (PEP) with CO2. The enzyme
responsible is the phosphoenol-pyruvate carboxylase. C4 plants have higher
photosynthesis per unit energy and per unit water than their C3 counterparts,
and this contributes to their adaptation to dry environments.
(c) CAM plants (Crassulacean Acid Metabolism). This system is predominant in
the Crassulaceae family that includes numerous cactuses. CO2 fixation occurs
during the night by formation of PEP which is converted to organic acids that
are stored in the vacuoles. During the day malate enters the chloroplasts where
PEP is regenerated. CAM plants behave as C3 if the water supply is adequate.
In drought situations CAM plants have reduced rates of water use as compared
to C3 and C4 plants.
13.4 Effects of Environmental Factors on Photosynthesis
The main environmental factors affecting photosynthesis rate are solar radiation,
temperature and CO2 concentration. Diseases, water and nutrient deficit may limit
strongly photosynthesis which is discussed in Chaps. 14 and 15.
13.4.1 Radiation
The net photosynthesis of a leaf (Pn) responds to irradiance (I) as shown in

Fig. 13.1. For I ¼ 0, the leaf loses CO2 at a rate Rd (dark respiration). Assimilation
is zero at the so called light compensation point where gross photosynthesis is equal
to respiration rate. With higher I, Pn grows rapidly to a maximum in which the
system reaches light saturation. For many species, Pn is saturated well below the
typical radiation of clear days, this is particularly true of most C3 plants. The
Fig. 13.1 Relations

NET PHOTOSYNTHESIS
between leaf net

C4
photosynthesis and
irradiance for C3 and C4
plants C3
IRRADIANCE
maximum value of Pn varies greatly between the C3 and the C4 groups and also
within each group. However, the initial slope of the relationship Pn ¼ f (I) is
relatively constant for all species.
If Rd is constant as I increases, gross assimilation is defined as Pg ¼ Pn þ Rd. The
gross assimilation rate increases with irradiance along a curve which can be fitted to
a hyperbola of the type:
ε I Pgx
Pg ¼ ð13:2Þ
ε I þ Pgx
where Pg is the rate of gross photosynthesis (μmol m–2 s–1), I is irradiance (W m-2),
ε is the initial slope of the curve Pg ¼ f (I) and Pgx is the asymptotic value of Pg when
I tends to infinity. C3 plants have values of Pgx in the range 10–40 μmol m2 s1
while C4 show a range 18–55 μmol m2 s1. The initial slope of the curve is similar
for C3 and C4 species with a value around 0.076 μmol J1. When I is expressed as
absorbed PAR (mol photons m2 s1) the initial slope is called quantum efficiency
of photosynthesis, which is a measure of the intrinsic efficiency of the photosyn-
thetic system. In the range 20–25 C both C3 and C4 plants have quantum
efficiencies around 0.06 mol CO2 E1. For lower temperatures C3 plants perform
better and for higher temperatures the opposite occurs.
C3 plants reach their Pgx with much lower irradiance than C4 plants. Under
optimal conditions C4 plants show higher Pgx than C3 plants, although the differ-
ences are attenuated as we scale up from leaf to plant community. Thus, the
maximum biomass produced by C3 and C4 crops differs much less than the
maximum rates of photosynthesis at leaf level.
The irradiance under which leaf growth occurred also affects its response to
radiation. When a leaf has grown in the shade, Pgx is lower than when it has grown
under high radiation. This process of acclimation is due to increased accumulation
of proteins (photosynthetic enzymes) in the leaves under high irradiance.
13.4.2 Temperature
The maximum photosynthesis responds to temperature following a bell shaped

curve with a maximum between 15 and 25 ºC for C3 plants and between 25 and
35 ºC for C4 plants (Fig. 13.2). Very few C4 species perform well under low
temperature, whereas most of these species suffer irreversible damage with tem-
peratures between 10 and 12 ºC (chilling injury). Many C3 plants such as cotton and
sunflower perform well with high temperature (30–40 ºC) although some C3 are
sensitive to low temperatures (e.g. banana). However leaves of most of C3 plants
can withstand temperatures down to 0 ºC.

maximum gross
photosynthesis to
MAXIMUM GROSS
PHOTOSYNTHESIS
(arbitrary scale)
temperature for C3 and C4
crops
C4
C3
0 10 20 30 40
TEMPERATURE (°C)
13.4.3 Concentration of CO2 and Endogenous Factors
Photosynthesis increases as CO2 concentration increases because of the larger

concentration gradient between the atmosphere and the leaf mesophyll (Eq. 13.1).
Net photosynthesis responds linearly to the [CO2] for C3 plants. Leaves of C3
species lose CO2 (Pn is negative) for CO2 concentrations below 50–100 ppm. The
concentration of CO2 for which Pn ¼ 0 is called CO2 compensation point. C4 plants
show little response to CO2 concentration in the range 200–400 ppm.
In addition to the environmental effects, photosynthesis rate can be affected by
the existence of sinks capable of accumulating carbohydrates. For example, sup-
pression of the tubers of potato plants leads to a decrease in the rate of leaf
photosynthesis.
13.5 Respiration
Respiration (oxidation of carbohydrates and other compounds to produce energy) in

plants can be decomposed into two main categories: growth respiration (Rg) which
is proportional to the gross photosynthesis and maintenance respiration (Rm),
proportional to crop biomass (B):
Rd ¼ Rg þ Rm ¼ a Pg þ b B ð13:3Þ
The growth respiration is the energy cost involved in the synthesis of plant tissues
from glucose, which is considered the building block of plant tissues. For example,
1 g of glucose allows building approximately 0.65 g of leaves (or vegetative tissues
in general). The energy cost depends mainly on tissue composition: synthesis of fats
and proteins involve a higher Rg than carbohydrates (see 13.11).
Maintenance respiration is the energy cost associated with maintaining the
organization and functioning of the crop tissues. The fundamental processes in
which Rm is inverted are protein turnover and keeping active ion transport
Fig. 13.3 Leaf respiration
RELATIVE RESPIRATION RATE

1
at 30 C as a function of leaf
N concentration per unit
leaf area
0
0 1 2 3
LEAF N CONTENT (g N m–2)
mechanisms. For this reason, N-deficient plants or older plants with lower nitrogen
concentration have lower maintenance respiration per unit of dry matter as illus-
trated in Fig. 13.3. The Rm per unit biomass increases exponentially with temper-
ature up to 40–50 ºC, depending on the species. For higher temperatures irreversible
damage occurs. Indicative values of Rm at 20 ºC range 0.01 and 0.035 g glucose/g of
dry matter/day.
13.6 Crop Photosynthesis
The response of photosynthesis of a crop to environmental conditions is in principle

more complex than that of its individual leaves. A crop is a set of leaves of different
age and nitrogen concentration, subjected to different radiation intensities that
change throughout the day. As will be seen later, in many cases crop photosynthesis
is linearly related to irradiance (or better, to intercepted radiation).
Considering the extinction of radiation within the canopy (Chap. 3) and the
response of leaf assimilation to irradiance (Eq. 13.2) we can integrate assimilation
for all the leaves and arrive to an equation for the rate of net photosynthesis of the
crop:

Pgx ε k I 0 þ 1:2 Pgx
Pnc ¼ ln L Rd ð13:4Þ
k ε k I 0 ekL þ 1:2 Pgx
where I0 is incident radiation, k is the extinction coefficient (see Chap. 3), L is the
Leaf Area Index and Rd is dark respiration per unit leaf area. The main determinants
of canopy photosynthesis are discussed using Eq. 13.4 in the examples below.
Example 13.1 Figure 13.4 shows the net photosynthesis of C3 crops with
extinction coefficients 0.5 (vertical leaves) and 0.9 (horizontal leaves) as a
function of LAI. The maximum crop photosynthesis occurs for LAI around
3 for horizontal leaves and LAI close to 5 for vertical leaves. When LAI is
low, horizontal leaves are more efficient for crop photosynthesis. Vertical
leaves show a clear advantage for large LAI.
25
CANOPY PHTOSYNTHESIS (μmol m–2 s–1)
20
15
10
k = 0.5
5 k = 0.9
0
0 1 2 3 4 5 6 7 8
LEAF AREA INDEX
Fig. 13.4 Net photosynthesis of C3 crops with extinction coefficients 0.5 (vertical leaves)
and 0.9 (horizontal leaves) as a function of LAI
Example 13.2 Figure 13.5 shows the net photosynthesis of C3 crops with
LAI¼1 and LAI¼4 as a function of incoming radiation. The response is
closer to linear than that shown by leaf assimilation to irradiance
(Fig. 13.1). When irradiance is low, the crop with LAI¼1 assimilates
more CO2 which is explained by the smaller respiration loss.
(continued)

25
CANOPY PHOTOSYNTHESIS (μmol m–2 s–1)
LAI = 1
20
LAI = 4
15
10
–5
–10
0 200 400 600 800 1000
IRRADIANCE (W m–2)
Fig. 13.5 Net photosynthesis of C3 crops with LAI¼1 and LAI¼4 as a function of
incoming radiation
Example 13.3 The effect of leaf photosynthetic capacity on canopy photo-

synthesis is shown in Fig. 13.6 for crops of different LAI. The range in
maximum leaf gross photosynthesis goes from the low range in C3 species
to the high range of C4. We have assumed that respiration is 10 % of Pgx and
solar radiation of 500 W m2. It is clear that the importance of leaf photo-
synthetic capacity is reduced as the canopy grows. For very high LAI, crop
photosynthesis is almost independent of Pgx. This may explain why breeding
for high leaf Pgx has not been successful in enhancing biomass production.
(continued)

30
NET CROP PHOTOSYNTHESIS (μmol m–2 s–1)
25
20
15
10
LAI = 1
LAI = 2
5
LAI = 4
0
0 10 20 30 40 50 60
MAX LEAF GROSS PHOTOSYNTHESIS (μmol m–2 s–1)
Fig. 13.6 Effect of leaf photosynthetic capacity on net canopy photosynthesis. The range in
maximum leaf gross photosynthesis goes from the low range in C3 species to the high range of
C4. We have assumed that respiration is 10 % of Pgx and solar radiation is 500 W m2
13.7 Intercepted Radiation
We analyzed radiation interception by crops in Chap. 3. Actual interception

depends on LAI, leaf angles and the geometrical distribution of incoming radiation
(sun angle, fraction of diffuse radiation). Models of radiation interception may be
quite complex and allow considering also the actual 3D distribution of foliage
elements in the canopy which is required for tree orchards or forests.
A more simple approach uses the relationship between crop coefficients and
radiation interception seen in Chap. 10. The FAO method for calculating Kc divides
the crop cycle into four stages (initial, rapid growth, maximum, declining). Rapid
growth starts when ground cover is around 0.2 and ends at ground cover 0.8–0.9.
Full radiation interception corresponds to Kc around 1.2 which leads to the simple
model:
f i ¼ K c 0:3 ð13:5Þ
where fi is the fraction of intercepted PAR. This should be valid for the third and
fourth stages. In the initial stage, when we move from zero to 20 % interception we
Fig. 13.7 Time course of kc

crop coefficient and the
fraction of intercepted PAR
for an annual crop. A: f
sowing to 20 % ground
cover, B: 20 % to 80 %
Kc or f
ground cover, C: 80 %
ground cover to start of leaf
senescence, D: start of
senescence to harvest
A B C D
TIME
may assume an average value of fi ¼ 0.1. Values of interception for the rapid growth
stage may be calculated by linear interpolation (Fig. 13.7). Note that in this simple
model intercepted radiation in the fourth stage, when leaves are old and the canopy
is senescing, is a surrogate for absorbed radiation by pigments. In other words, dead
leaves do not contribute to photosynthesis or transpiration, so the estimated value of
intercepted radiation refers to functional leaves.
13.8 Radiation Use Efficiency
The relationship between crop growth rate (CGR: increase in above ground biomass
per unit time) and intercepted radiation is approximately linear. Therefore the
relationship between biomass production (B) and the sum of intercepted radiation
should be linear:
X
harvest
B ¼ RUE f i Rsp ð13:6Þ
emergence
where the proportionality coefficient is the RUE (Radiation-Use Efficiency, units of

g/(MJ PAR)) and Rsp is incoming PAR. The first to propose such a relation was
Monteith in 1977, who plotted biomass versus accumulated solar radiation
intercepted by different C3 crops (sugar beets, potatoes, barley, apples) and found
similar linear relationships with slope around 1.4 g dry matter/MJ solar radiation
despite the obvious differences among these crops. The RUE of C3 plants inside
greenhouses is higher than outdoors, with values in the range (2.2–2.5 g/(MJ PAR),
which is explained by the higher proportion of diffuse radiation, the favorable
thermal environment and lack of wind that promotes higher shoot:root ratio
(Fig. 13.8).
1.4
1.2
PHOTOSYNTHESIS (mg CO2m-2s-1)
1.0
0.8
0.6
0.4
Morning
0.2 Afternoon
0.0
0 100 200 300 400 500
-2
PAR (W m )
Fig. 13.8 Relationship between hourly crop photosynthesis and intercepted PAR in a cotton crop.
Cordoba, Spain. June 23, 2003. For the same radiation photosynthesis is higher during the morning
than during the afternoon
13.9 Crop Potential Productivity
We can calculate crop yield (Y) as the product of biomass and harvest index (see
below), and further expressing biomass as the product of RUE and total intercepted
PAR (Eq. 13.6):
X
harvest
Y ¼ HI RUE f i Rsp ð13:7Þ
emergence
Assuming crops grown under ideal conditions that maximize capture and efficiency
in the use of radiation and harvest index, we can calculate the potential dry matter
yield of a crop in a given environment. Abiotic and biotic stress factors can reduce
harvest index (see below), and the components of crop biomass (Sects. 13.10 and
13.11). It is important to keep in mind that a fraction of mass harvested by farmers is
water. Therefore the commercial or fresh yield can be calculated using Eq. 13.7 as:
1 X
harvest
Y fresh ¼ HI RUE f i Rsp ð13:8Þ
1w emergence
where w is the fraction of water over fresh mass. Values of w for different species
are shown in Table 24.1. The difference between Y and Yfresh is small for crops that
are harvested dry (cereals, grain legumes) but it is very large for some species
(e.g. potato, tomato).
The fraction of biomass that is harvested is called Harvest Index (HI ¼ Y/B).
Biomass usually refers only to the above ground fraction, unless the harvested
organ is underground, in which case it is included in the biomass. The HI of fodder
crops is very high (up to 0.9), as we may harvest and use almost all the aerial
biomass (see Table 13.1). Small grain cereals like wheat can reach HI near 0.50, and
many legume and oilseed crops show HI in the range 0.30–0.40. Crops with
subterranean harvestable organs show high HI like sugar beet (up to 0.7) and potato
(up to 0.8). It should be remembered, however, that the values shown in Table 13.1
correspond to “normal” cropping conditions, and that in extreme situations the HI
may go down to zero, for example if a crop fails to set grain due to untimely frost
or heat.
13.10 Dry Matter Partitioning and Harvest Index
The dry matter accumulated during the crop cycle is partitioned among the different
plant organs. In determinate annual crops there is a distinct vegetative growth phase
where assimilates are partitioned to leaves, stems and roots. This phase is followed
by a reproductive phase when inflorescences, flowers, seeds and supporting struc-
tures grow while vegetative growth stops. In plants with an indeterminate growth
habit, vegetative and reproductive growth overlap for much of the cycle. The
difference between determinate and indeterminate growth is relative in the sense
that there will always be some overlap between vegetative and reproductive growth.
The shorter the overlap is, the more determinate the growth habit. Interestingly
genetic variability exists for growth habit which has been exploited to develop
determinate cultivars that allow mechanical harvesting (e.g. processing tomato).
The fraction of dry matter that goes to each plant organ is called the partition
coefficient. Part of the dry matter may be remobilized later and be transported from
one organ to another. This typically occurs with reserve carbohydrates in the stem,
which support partially seed growth during the seed filling period. In crops like
sunflower and wheat, stem reserves can contribute significantly to grain yield, in
particular under conditions that constraint photosynthesis during grain filling such
as drought and foliar diseases.
The general trend over the twentieth century was the increase in Harvest Index of
the different crops, which has enabled significant increases in yield without major
improvements in biomass production. One example is that of the wheat cultivars
obtained by the International Center for Improvement of Maize and Wheat
(CIMMYT) in which improved HI was associated to short stature plants which
enabled increasing N fertilization without increased risk of lodging. Maize is an
exception to this trend, as improvement in yield has been primarily achieved by
increasing crop biomass mediated by tolerance to crowding in turn allowing higher
sowing densities. For the last 50 years, plant breeders have focused their improve-
ment programs on increasing HI to the point that the major crops are now
approaching their biophysical limits in terms of HI.
Table 13.1 Harvest index (%) of different crop species

HImin HImax
Alfalfa 90 90
Apple 60 70
Barley 40 50
Cassava 30 65
Chickpea 30 36
Citrus 50 60
Cotton 25 40
Cowpea (determinate) 44 60
Cowpea (indeterminate) 15 30
Bean (dry seed) 25 45
Faba bean 25 40
Garlic 50 55
Grain sorghum 40 50
Grapevine 40 60
Lentil 35 50
Lupin 30 50
Maize 40 55
Oats 40 50
Oil palm 40 45
Olive 50 70
Onion 60 85
Opium 30 40
Pea 35 45
Peanut 35 42
Pepper 40 60
Potato 50 80
Rapeseed 30 35
Rice 40 55
Safflower 25 35
Soybean 35 50
Sugarbeet 50 70
Sugarcane 70 80
Sunflower 25 40
Tobaco 60 80
Tomato (industry) 45 60
Triticale 45 47
Vetch 40 50
Wheat 40 50
The typical intervals for commercial crops are shown. These values may be taken as representative
of crops not suffering from extreme biotic or abiotic stress.
Table 13.2 Effect of nitrogen supply on harvest index (relative to a maximum value of 0.47) of
wheat at three locations in Australia
N rate (kg N/ha) Ginninderra Pucawan Wagga
0 0.94 0.89 0.72
200 1.00 0.85 0.55
Adapted from van Herwaarden et al. (1998) Austr J Agri Res 49, 1067–1082
Harvest index depends mainly on the developmental pattern of the crop, on the
distribution of assimilates among plant organs and on the ability to translocate
assimilates to the harvestable organ. In particular in determinate crops
(e.g. sunflower) the reproductive phase is clearly separated in time from the
vegetative phase. Failure in setting seeds (or harvestable organs in general) may
be a primary cause for reduced HI. Later, low post-anthesis assimilation or poor
translocation of assimilates may reduce further the HI. Therefore several critical
periods for determining the HI may exist (flower initiation, flowering, pollination).
A greater flexibility of the HI may be expected for indeterminate crops (e.g. cotton)
as both vegetative and reproductive growth overlap during most of the crop cycle.
Nutrient and water deficit have variable effects on harvest index. This relates to
the definition of harvest index as a ratio, yield to biomass. Stress may reduce both
biomass and yield in equal proportion with no consequence for harvest index, or
reduce biomass proportionally more than yield hence increasing harvest index, or
reduce yield proportionally more than biomass, thus reducing harvest index. The
actual response of harvest index to stress is thus contingent to the nature of the
stress, and its timing, intensity and duration. Table 13.2 illustrates all three possi-
bilities. In comparison to unfertilized crops, well-fertilized crops had similar, lower
or higher harvest index depending on location-specific growing conditions. The
seasonality of rainfall in dryland systems often has an impact on yield mediated by
harvest index. In Mediterranean environments with winter rainfall that favors
biomass growth and scarce rainfall during reproductive stages, harvest index is
often much lower than its potential. In irrigated systems, strategies of deficit
irrigation may be used to improve HI by reducing water use during specific periods
that favor reproductive over vegetative growth (see Chap. 21).
13.11 Crop Factors Affecting Interception and Efficiency

in the Use of Radiation
Chapter 3 discussed crop traits affecting radiation interception, including optical

and geometrical properties of leaves and canopies. The RUE, which is the amount
of dry matter produced per unit of intercepted PAR, depends on the crop species.
Under optimal temperature and water supply C3 plants produce between 2 and
2.5 g/MJ PAR and C4 plants produce 2.5–3.0 g/MJ PAR.
Differences in seed composition explain the higher RUE of cereals in compar-

ison to both oilseed crops and grain legumes. Following the analyses of Penning de
Vries, the conversion factor for formation of dry matter (CVF: g dry matter/g
glucose) can be calculated as:
1=CVF ¼ ð1:24 FC þ 1:70 FP þ 3:11 FF þ 2:17 FL þ 0:93 FO þ 0:05 FMÞ

ð13:9Þ
where FC, FP, FF, FL, FO and FM represent the fractions of carbohydrates, protein,
fat, lignin, organic acids and minerals in the dry matter being formed. With these
values we derive an approximate method to calculate RUE of a crop based on
biomass composition, assuming that the harvested product consists of carbohy-
drates, proteins and fats, and the crop residues contain only carbohydrates:
RUEc
RUE ¼ ð13:10Þ
ð1 HI Þ þ HIðFC þ 1:4 FP þ 2:5 FFÞ
where RUEc is the RUE for production of carbohydrates. Appendix 32.1 (Chap. 32)
presents data of composition of harvested parts of many crops species.
Example 13.4 Winter cereals have a seasonal RUE around 2 g/(MJ PAR)
which may be taken as the reference RUEc. Then for oilseed sunflower with
HI ¼ 0.35 and seeds containing 45 % fat and 20 % protein, RUE will be:
2
RUE ¼ ¼ 1:58 g=ðMJ PARÞ
ð1 0:35Þ þ 0:35ð0:35 þ 1:4 0:2 þ 2:5 0:45Þ
Example 13.5 We will calculate the potential yield (without limitation of

water and nutrients) of castor bean (Ricinus communis) in Cordoba, (Spain).
The crop is sown on April 1 and harvested on September 30. The average
fraction of intercepted radiation is 0.2 in April, 0.4 in May, 0.9 in June, July
and August and 0.2 in September. Harvest index is 0.25. Seed water content is
5 %. The castor seed contains 50 % fat and 15 % protein (dry mass basis). We
also assume that temperature does not affect crop productivity. Solar radia-
tion data and PAR interception calculation are presented in Table 13.3.
Seasonal intercepted PAR is 1106 MJ m2.
First we calculate RUE using Eq. 13.10 and assuming RUEc ¼ 2 g/
(MJ PAR), which yields
RUE ¼ 1:66 g=ðMJ PARÞ
(continued)

Table 13.3 Calculation of intercepted radiation for a castor bean crop in Cordoba (Spain)
Month Rs PAR Days/month f Intercepted PAR per month
MJ/m2/day MJ/m2
1 8.8 4.0 31
2 9.9 4.4 28
3 11.7 5.2 31
4 16.9 7.6 30 0.2 45.7
5 20.1 9.1 31 0.4 112.3
6 23.8 10.7 30 0.9 289.7
7 25.1 11.3 31 0.9 315.4
8 23.5 10.6 31 0.9 295.4
9 17.8 8.0 30 0.2 48.0
10 10.9 4.9 31
11 8.0 3.6 30
12 6.3 2.8 31
Season 1106.4
And now we calculate yield applying Eq. 13.8:
1 X
harvest
1
Y fresh ¼ HI RUE f i Rsp ¼ 0:25 1:66 1106 ¼ 483 g m2
1w emergence
1 0:05
¼ 4830 kg=ha
Variations in RUE due to the composition of the harvested product and loss of
photosynthetic capacity that occurs in many crops during the final phase of the cycle
reduce seasonal RUE below the maximum values. Thus, for seasonal calculation we
propose intervals of RUE of 1.6–2.0 g/(MJ PAR) for non-leguminous C3 plants,
1.5–1.8 for C3 leguminous plants and 2.2–3.0 g/(MJ PAR) for C4 plants. These RUE
values represent “well adapted” crops under normal cropping conditions, so RUE
values include the effect of temperature and assume good supply of water and
nutrients. However, RUE is expected to change under cooler or hotter environments,
and when water and nutrients are in short supply, as explained in the next section.
Some species show low RUE due to their low leaf photosynthesis; for example
in olive RUE is 1.3 g/(MJ PAR) for non bearing trees and 0.9 g/(MJ PAR) for
bearing trees. This difference is basically due to the higher cost of biomass
production of olive bearing trees as fruits accumulate oil. In citrus RUE of adult
trees is 1.3 g/(MJ PAR).
The intra-specific variation in RUE is smaller than the differences between crop
types associated with photosynthetic metabolism and seed composition.
3.0
RADIATION USE EFFICIENCY (g MJ–1)

2.5
2.0
1.5
UK
Australia
1.0
1955 1965 1975 1985 1995 2005
YEAR OF RELEASE
Fig. 13.9 Selection for yield indirectly improved the pre-flowering radiation use efficiency of
wheat varieties in UK and Australia (Adapted from Evans JR (2013) Plant Physiol
162:1780–1793)
Nonetheless, selective pressure for yield has improved the radiation use efficiency
of wheat in both favourable (UK) and stressful (Australia) environments (Fig. 13.9).
The improvement in pre-flowering RUE of Australian wheats was associated with
higher nitrogen uptake (i.e. greener leaves), and changes in canopy architecture that
favoured greater PAR penetration in the profile; effectively, modern varieties have
a greater proportion of leaves contributing to total crop photosynthesis, whereas the
photosynthesis per unit leaf area remained unchanged. The role of nitrogen on crop
photosynthesis is discussed in the next section.
A single RUE value is useful for comparisons between contrasting crops types
(C3 v C4, cereal v oilseed) and also captures large environmental effects, such as
shortage of water or nitrogen (next section). However, RUE changes with crop age,
developmental stage and the pattern of nitrogen and dry matter allocation. For
example, studies of RUE in sunflower distinguished three phases: establishment
(sowing to 47 days after sowing), rapid growth (47 days after sowing to anthesis)
and grain filling (anthesis to maturity). Radiation use efficiency was highest during
rapid growth (2.4 g/MJ) and lower during establishment (1.0 g/MJ) and grain filling
(1.3 g/MJ). The low RUE during establishment was associated with a large pro-
portion of leaves exposed to high radiation, and the intrinsically lower efficiency of
leaves at saturated light regimes. The low RUE during grain filling was related to
high respiration and high synthesis cost of oil and protein in the seed (see Eq. 13.10)
and to leaf senescence.
The concept of source and sink is useful to understand the physiology of the
crop, despite some problems in definitions. A mature leaf is a net source of carbon,
whereas the growing seed is a sink. A leaf however transitions from sink at early
stages of development to source at later stages. Likewise, wheat and sunflower
stems are sinks of carbohydrates during early growth stages but became sources for
grain filling. The source:sink ratio can be manipulated experimentally by increasing
source activity (e.g. increasing ambient CO2 concentration or radiation), reducing
source activity (e.g. defoliation, shading), reducing sink activity (e.g. removal of
maize ear, cooling of potato tubers), or increasing sink activity (e.g. applying
gibberellic acid). Experiments where the source:sink ratio is diminished often
show increase in photosynthesis, whereas increasing source:sink ratio may reduce
photosynthesis. With high source:sink ratio, the plant has a relative excess of
photosynthates, and the accumulation of starch in leaves can trigger the feedback-
inhibition of photosynthesis usually mediated by reduction in stomatal conduc-
tance. Conversely, leaf photosynthesis may be stimulated if the capacity of the
source to meet the carbohydrate requirements of the sink is restricted. These source:
sink interactions are agronomically important. For example in pasture species such
as alfalfa or grasses, animal browsing reduces the size of the canopy relative to root
biomass, hence the reduced source:sink ratio. After browsing, the rate of photo-
synthesis of remanent leaves may increase in response to low source:sink ratio,
hence contributing to re-growth. This “compensatory photosynthesis” has also been
recorded in plants after damage by insects that reduce source:sink ratio
(e.g. defoliators). In wheat varieties developed in CIMMYT (International Centre
for Improvement of Maize and Wheat), increased seed number in modern varieties,
compared with older ones, has increased RUE during grain filling; this increase in
crop photosynthesis has been interpreted in terms of reduced source:sink size, as
grain number increased without a proportional increase in canopy size. Effectively,
more grains are “pulling” photosynthesis up.
13.12 Environmental and Management Factors Affecting

Interception and Efficiency in the Use of Radiation
Effects of temperature on biomass production and its components, radiation inter-

ception and RUE are twofold. First, and most important, temperature changes the
duration of the period from sowing to harvest, setting the limits of integration in
Eq. 13.7. At high latitudes in Europe, Asia and North America, warming over recent
decades has extended this period, with positive implications for crop growth and
yield (Chap. 11). In contrast, increasing temperature in subtropical and temperate
environments may shorten season length, with potentially negative effects for crop
yield. The early observation by Westlake thus remains relevant in this context:
“. . .adaptation of agricultural techniques to increase the proportion of the year with
the ground completely covered by leaves is probably the most rewarding change
that could be made to increase the world’s food supplies. . .” Second, RUE is
non-linearly related to temperature, an effect that is mediated by the effects of
temperature on lower-level processes such as leaf gross photosynthesis, respiration
Fig. 13.10 Relationship 14

between percentage yield
increase in response to 12
increase in radiation
YIELD INCREASE (%)

interception due to 10
reduction in row spacing for
maize, soybean and 8
sunflower (Adapted from
Andrade et al. (2002) Agron 6
J 94:975–980)
4
0
0 5 10 15 20 25
RI INCREASE (%)
and dry matter partitioning. For example, field pea RUE is reduced with average
ambient temperatures below 12 C or above 22 C.
Row spacing is manipulated to improve capture and efficiency in the use of
resources, and to accommodate practices such as weed control and tillage
(Chap. 17). The best row spacing for a combination of crop, soil, climate and
cropping system is usually determined empirically. Alternatively, we can apply the
physiological concepts developed in this chapter to predict the responses of crop
yield to row spacing as illustrated in Fig. 13.10. For the well-watered crops in this
study, no yield gains are expected from narrowing the space between rows when
wide-row crops intercept 90 % of incident radiation in the critical period of yield
determination but gains up to 12 % can arise if wide-row crops intercept 60–90 % of
radiation. The gain in yield with narrowing row distance is proportional to the gain
in intercepted radiation (Fig. 13.10). In recent decades, even though row spacing
has been diminishing to increase radiation interception, in practice sowing machin-
ery and tillage/traffic practices are the major determinants of row spacing.
Bibliography
management in agricultural systems (2nd ed.). New York: Cambridge University Press.
Hay, R., & Porter, J. (2006). The physiology of crop yield (2nd ed.). Oxford: Wiley-Blackwell.
Monteith, J. L. (1977). Climate and the efficiency of crop production in Britain. Philosophical
Transactions of the Royal Society of London B, 281, 277–294.
Sadras, V. O., & Calderini, D. F. (2014). Crop physiology (2nd ed.). San Diego: Academic Press.
Chapter 14
Effects of Water Stress on Crop Production
Victor O. Sadras, Francisco J. Villalobos, Francisco Orgaz,

and Elias Fereres
Abstract Water stress is due to a low water potential in the plant as a result of low
soil water potential, high evaporative demand and/or a substantial resistance to
water flow through the plant. The water deficit affects many processes in the crop,
although most of the effects are related to the reduction in growth, the most
sensitive process, and to stomatal closure. Mild to moderate deficits do not affect
harvest index, and in some species they may increase it. Instead severe water
deficits reduce the HI. The effect of water stress on crop yield can be quantified
by Stewart’s equation which establishes that the relative reduction in yield is
directly proportional to the relative reduction in ET, with an empirical coefficient
(Ky) which ranges between 0.8 and 1.5. More mechanistic type models may be used
to characterize the yield responses to variable water supply, but they need to be
locally calibrated for accuracy.
14.1 Introduction
Cultivated plants require a continuous supply of water to replace that lost by

evaporation from the aerial organs, particularly from the leaves. This requirement
is due simply to the fact that the leaves are exposed to a high evaporative demand
(solar and thermal radiation and warm, dry air) while the substomatal cavities are
saturated with water vapor. The gradient in water potential between the substomatal
V.O. Sadras (*)

F. Orgaz

Agriculture, DOI 10.1007/978-3-319-46116-8_14
cavity and the leaf surface thus drives the evaporation process. Leaf cuticle
however is a significant barrier for evaporation, leading to a mainstream transpira-
tion flow through the stomata. For carbon dioxide to enter into the leaves, the
stomata must be open, allowing water vapor to escape in response to the vapor
pressure gradient. To maintain water flow without tissue desiccation, terrestrial
plants have developed elaborate systems for the uptake and transport of water.
Water flows from the soil into the root system and then is carried by xylem vessels
to the leaves, where they replace the water evaporated into the atmosphere. Thus,
from a purely physical point of view, the plants transport water from a source, the
soil, to a sink, the atmosphere. These systems are capable of transporting large
quantities of water, equivalent, in a typical summer day to a water depth of 6–8 mm
in the field, which involves several times the total weight of the plant. However, a
small, hardly detectable imbalance in the transport process in response to changes
in the supply of water from the soil or in atmospheric demand, causes a water deficit
in the plant. These mild deficits that are often harmful to crop growth and yield may
occur in spite of the large amounts of water used by the plants. In this chapter, we
outline the concepts of energy status of water in the plant, and its role in driving
water flows. Then we look at the responses of crop processes to water deficits and
the mechanisms of crop regulation of transpiration. After establishing these prin-
ciples, we conclude with an analysis of the important links between water and crop
production from an agronomic viewpoint.
14.2 Energy Status of Water in the Plant
As we have seen for the soil (Chap. 8), the energy status of water in the plant may be
characterized in terms of water potential. To do this we must go down to the cellular
level, since water status varies between subcellular compartments.
Schematically a plant cell is a protoplast (nucleus, cytoplasm and vacuoles)
surrounded by a membrane (plasmalemma) which presses on a semi-rigid cell wall.
This pressure, called turgor, reflects turgor potential (Ψp). The osmotic potential
(Ψo) due to the presence of solutes is the other main component of total water
potential (Ψ) in the protoplast. In the cell wall, a porous structure composed of
microfibers and polysaccharides, the solute concentration is much lower, so Ψo is
very small and Ψp is zero. Thus, the major component of potential outside the
protoplast is the matrix potential (Ψm) due to adsorption forces that the porous cell
wall matrix exerts on water. Another component of the potential, the gravitational is
negligible in plants with the exception of very tall trees.
Example 14.1 In an equilibrium situation the water potential in the vacuole

is 0.5 MPa with components Ψp ¼ 0.2 MPa and Ψo ¼ 0.7 MPa. The total
potential in the cell wall is also 0.5 MPa. Its components are Ψm ¼
0.48 MPa and Ψo ¼ 0.02 MPa.
14 Effects of Water Stress on Crop Production 191
Outside the cell, in the xylem, for the same water potential, its components will
be different than in the cell. The Ψo is much lower and Ψp is negative since water in
the xylem is under tension. In fact, turgor in the protoplast and tension outside
creates a very large pressure gradient that would tear the plasmalemma if not for the
rigidity of the cell wall containing it.
The pressure potential is the main driver of expansive growth of plant shoots and
leaves and is largely responsible for the proper functioning of some basic processes
for growth and crop production.
14.3 Causes of Water Deficits
From soil to substomatal cavity, water moves in liquid phase whereas the flow from
substomatal cavity to atmosphere is in vapour phase. Transpiration (Ep) is the water
vapor flux from the substomatal cavities into the atmosphere following a vapor
pressure gradient. This loss of water from the plant is compensated by the inflow of
water absorption from the soil by the root system. The water flow can be expressed
in terms of the water potential gradient between the soil and the leaves:
Ep ¼ ðΨs Ψl Þ=Rsl ð14:1Þ
where Ψs is soil water potential, Ψl is leaf water potential and Rsl is the resistance to
flow between the soil and the leaves. Rearranging Eq. 14.1:
Ψl ¼ Ψs Ep Rsl ð14:2Þ
That is, the leaf water potential, which is always lower than Ψs, is a function of soil
water potential, the flow resistance and transpiration rate. A low leaf water potential
is required for water to move from the soil, so leaves are under a water deficit as
long as transpiration occurs. However, it is considered that a plant is water stressed
when the water potential in its tissues decreases enough to affect physiological
processes. Thus, the causes that can lead to low leaf water potential and therefore to
water deficit are:
(a) Low soil water potential (low water content and/or high salt concentration in
the soil solution).
(b) High evaporative demand (high Ep)
(c) High resistance to water flow (high Rsl) in the soil (low hydraulic conductivity,
dry soil) or in the plant (low root length density, vascular diseases, etc.)
These processes can act simultaneously and in the same or in opposite direction.
For instance, low soil water potential causes an increase in the soil-root resistance,
while increases of transpiration may be accompanied by a reduction in resistance.
Water deficits may occur in the short term in plants with good water supply
during the middle of the day, in response to increased evaporative demand. By
WATER UPTAKE
WATER FLUX
TRANSPIRATION
SUNRISE SUNSET
Fig. 14.1 Diurnal variation of transpiration and water uptake
Stomata
Atmosphere
Cuticle
Xylem
Root (axial)
Soil
Soil Root
(radial)
Fig. 14.2 Diagram of the soil-plant-atmosphere water flux
contrast, long term water deficits are commonly associated with progressive deple-
tion of soil water.
We have seen then that the water status of the plant is the result of interactions
between atmospheric demand (Ep) and the ability of the plant to meet this demand,
which depends on the water content in the soil and the flow resistance. Transpira-
tion rates and soil water uptake follow similar diurnal patterns with a maximum in
the middle of the day, due to the high evaporative demand, but are offset in time
(Fig. 14.1).
The role of Rsl is critical in regulating the water flow between the soil and the
atmosphere. Between the soil and the leaves the resistance can be separated into
several, as indicated schematically in Fig. 14.2. The resistance between the soil and
Fig. 14.3 Cereals DAY AFTER SOWING

(e.g. wheat) have a much 0 50 100 150 50 100 150
larger root length density
4 Cereals Dicots
than dicots (e.g. lupin) in
crops of similar leaf area 3
index (Adapted from
LAI
Hamblin and Tennant 2
(1987). Australian Journal
of Agricultural Research, 1
38, 513–527)
0
TOTAL ROOT LENGHT (km root/m2)

2
10
12
14
the root depends on the soil hydraulic conductivity (Kh), root length density (Lv),
and root resistances. As the soil dries, Kh decreases exponentially and Rsl increases.
The root length density (root length per unit volume of soil) usually decreases
exponentially with depth. A higher Lv implies a shorter average distance between
the soil and the root surface and thus lower Rsl values.
Cereals have higher root length density than dicots for crops of similar size
(Fig. 14.3), reaching Lv between 3104 and 10104 m m3 in the surface layers
(0–30 cm). The low root length density of dicots, which can support similarly large
canopies, is compensated by larger and more abundant metaxylem vessels and
smaller axial resistance to water flow. Numerous studies have shown that Lv around
1104 m m3 is enough for full soil water extraction down to the Permanent Wilting
Point. In fact, some crops such as sunflower are capable of full extraction of subsoil
water with Lv of about 0.5104 m m3. It is likely that the adaptive value of higher
Lv is related to the extraction of immobile nutrients as P, rather than to the
absorption of water.
Once water penetrates the surface of a root, it must overcome a significant radial
resistance before reaching the xylem. This resistance is due to the suberization of a
layer, the endodermis, which prevents water from flowing through the apoplast
hence forcing water flow through cell membranes. The lipid bilayer of cell mem-
branes is a significant resistance to water flow in roots. However, membranes
contain a particular type of proteins, called aquaporins. Effectively, aquaporins
can control water flow across membranes by changes in their abundance, by
opening and closing the channel (gating), or both; for these properties an analogy
can be drawn between aquaporins and stomata. In some cases, there is a good
correlation between expression of aquaporins and physiological traits,
e.g. aquaporin expression follows a day/night cycle consistent with the daily
changes in root hydraulic conductivity coupled with transpiration. After reaching

the xylem, the axial resistances from the root to the base of the stem and from the
stem to the leaves are smaller (Fig. 14.2).
Water in the leaves finds two resistances in parallel: the stomatal resistance of
variable magnitude depending on stomatal opening and the cuticular resistance,
much greater than the stomatal resistance, but highly variable among crops, with
rice having half the value of corn or sorghum.
Although the crop is a simple intermediary for water transport between the soil
and the atmosphere, the modulation of the resistances described above allow the
matching, within certain limits, of soil water supply to the evaporative demand of
the atmosphere.
14.4 Effects of Water Deficits
Water stress can affect virtually all morphological and physiological processes of
the crop if the duration and severity of stress are intense enough. The general
response is reflected in a reduction in plant size, leaf area and harvestable yield.
The main effects of water stress on crops can be explained largely through the
effect on two processes: expansive growth and stomatal functioning.
14.4.1 Expansive Growth
Turgor is responsible for the plant form and is a prerequisite for cell expansion,
since the increase in size requires both new cell wall synthesis and turgor pressure
against the wall. Relative growth rate of the cell may be related to Ψp by the
following equation:
dV 1
¼ ε Ψ p Ψ pu ð14:3Þ
dt V
where V is the volume of the cell, ε is the cell wall extensibility and Ψpu is the
threshold turgor potential, below which the expansion is stopped. The value of Ψpu
is high which means that with a small reduction in Ψp, expansion may cease, and
this would occur much sooner than wilting (Ψp ¼ 0).
In general if the crop is subjected to a progressive water deficit, acclimation
occurs by increasing ε and/or reducing Ψpu that conditions the future plant
responses to water deficit. These acclimations occur differentially in the expansion
of roots and aerial parts, so that the growth of the shoot is much more sensitive to
water stress than root growth. In roots Ψpu is reduced rapidly and solutes accumu-
late so that Ψo is reduced (osmotic adjustment), allowing root growth to continue
under low soil water potential.
The great sensitivity of expansive growth to water deficits leads to expansion

reductions for all organs that are within their growth phase when the water deficit occurs,
which causes varying effects at the crop level depending on the timing of water deficit.
14.4.2 Stomata Functioning
In terms of evolution, stomatal closure prevents desiccation and death of the plant
in the aerial environment, in special when water is scarce. The tradeoffs of stomatal
closure are the reduction in the flux of CO2 into the leaves and thus in assimilation,
and the increase in canopy temperature. Stomatal closure under high vapor pressure
deficit or drying soil favors a higher photosynthesis/transpiration ratio which is
related to water use efficiency as discussed below (Sect. 14.5).
The stomata are formed by two guard cells of kidney shape which are welded at
their ends. The stoma opening occurs when the guard cells are turgid. If their turgor
is reduced the guard cells approach and the stoma closes. Therefore stomatal
resistance is a function of pressure potential of the guard cells which is regulated
by leaf water potential and other factors in a rather complex way. The complexity in
stomatal behavior arises from several causes. First, stomata are responsive to many
environmental (CO2 concentration, temperature, vapor pressure deficit, radiation,
light quality) and plant factors (e.g. leaf age, source-sink ratio, see Chap. 10).
Second, the process of stomata regulation is highly dynamic; for example stomatal
resistance increases when evaporative demand increases, but the increased resis-
tance improves hydration and thus leaf water potential, which in turn stimulates
stomata opening. In short-time scales (minutes) the size of the stomata pore changes
continuously, and there is also variation among stomata in the same leaf. Third,
there are genotype-dependent differences in stomata response to soil and atmo-
spheric dryness. In some species such as maize and soybean, stomata tend to close
when leaf water potential drops; this allows for maintenance of leaf water potential
and for this reason this strategy has been called “isohydric”. In other crops such as
sunflower, stomata is more likely to remain open at low water potential; this
response is called “anysohydric”. There is, however, intra-specific variation in
these strategies and a wide range in responses between the extreme isohydric and
anysohydric behaviors (Fig. 14.4). Fourth, there are adaptive trade-offs. To under-
stand these trade-offs, we need to recognize three functions of stomata: regulate the
ratio between photosynthesis and transpiration, prevent cavitation, and regulate
canopy temperature. Stomata closure in drying soil or under high evaporative
demand favors the photosynthesis/transpiration ratio, and prevents cavitation.
However, stomata closure increases canopy temperature. This trade-off means
that stomata physiology cannot always be optimized. In wheat, cotton and Shiraz
grapevines, heat stress favors stomata opening and evaporative cooling at the
expense of photosynthesis/transpiration, suggesting that keeping the canopy cool
is, under some conditions, more important than saving water (Fig. 14.5). The role of
stomata in regulating canopy temperature is particularly important in warm regions
where heat waves, compounded with dry soil, can cause severe damage.
Fig. 14.4 Stomatal

conductance declines with
RELATIVE CONDUCTANCE
increasing water deficit, but
stomata are more
responsive in Grenache
(continuous line) than in
Shiraz (dotted line)
(Adapted from Schultz
(2003). Plant, Cell and Syrah
Environment, 26, Grenache
1393–1405)
PRE-DAWN WATER POTENTIAL
Control
Heated
100%
Morning Afternoon Morning Afternoon Morning Afternoon

CONDUCTANCE TRANSPIRATION PHOTOSYNTHESIS
Fig. 14.5 Comparison of stomatal conductance, transpiration and net photosynthesis in well-
watered Shiraz vines grown under ambient (control) and elevated (heated) temperature. Elevated
temperature increased stomatal conductance and transpiration, and to a lesser extent photosynthe-
sis. Evaporative cooling maintained difference in canopy temperature below 1 C between heated
and control vines, at the expense of lower photosynthesis/transpiration ratio. M morning,
A afternoon (Adapted from Soar et al. 2009. Functional Plant Biol 36, 801–814)
14.4.3 Crop Regulation of Transpiration
Crop transpiration (Ep) can be analyzed as the product of three factors: total
transpiring area, approximated by the leaf area index (LAI), intercepted radiation
(IR) per unit leaf area index, and transpiration per unit intercepted radiation:
Ep ¼ LAI IR=LAI Ep =IR ð14:4Þ

Table 14.1 Relative importance rated from greatest (+++) to negligible () of mechanisms
regulating transpiration of sunflower in pre- and post-anthesis
transpiration
RELATIVE VALUE
expansion
summer
expansion
spring
0
0 1
SOIL WATER CONTENT
Fig. 14.6 Relations between expansive growth or transpiration and soil water content for sun-
flower in spring and summer. Soil water content is relative to the values in Permament Wilting
Point (0) and Field Capacity (1) (Adapted from Sadras et al. 1993. Agron J 85, 564–570)
A small canopy is the most effective means of transpiration control. Indeed, LAI
components including tillering or branching, and area of individual leaves are very
responsive to water deficit. For a given LAI, the amount of radiation intercepted,
which provides the energy for water evaporation, depends on canopy architecture.
Figure 13.6 (Chap. 13) shows how wilting can transiently reduce IR/LAI and
therefore reduce transpiration. Transpiration per unit intercepted radiation is related
to the conductance of crop canopies, which is related to both its aerodynamic
properties and stomatal conductance (Box 14.1).
The relative importance of the factors in Eq. 14.4 changes with crop develop-
ment (Table 14.1). In general tissue expansion is much more sensitive to water
deficit than stomatal conductance, and regulation of LAI is the main control of
transpiration. Figure 14.6 illustrates this idea with experimental data of sunflower in
Cordoba in 1991. In summer the expansion rate decreases when relative soil water
content is 0.8, while stomatal conductance is affected when relative soil water is
0.2–0.3. In spring expansion is reduced after half of soil water is depleted, which is
explained by the better water status of plants under lower evaporative demand.
Thus, during the period of canopy expansion, reduction in LAI is the main control
of transpiration. However, in annual crops LAI peaks around flowering, when the
number and individual size of leaves is fixed; canopy expansion is therefore no
longer relevant during grain filling. In this stage, leaf senescence is the only
possible regulation of LAI, and often water deficit accelerates this process
(Table 14.1).
Box 14.1 Canopy-Atmosphere Coupling

Figure 14.2 emphasizes the flow of water from soil to atmosphere at the level
of the plant. At the level of the crop, however, there is a further resistance
between the canopy and the atmosphere. This is often called aerodynamic
resistance. With increasing aerodynamic resistance, the canopy and the
atmosphere became increasingly decoupled. In these cases, the impact of
stomatal resistance in modulating crop transpiration is less than in cases
where the aerodynamic component represents a smaller resistance. In tall,
aerodynamically rough vegetation such as conifer forests, the aerodynamic
resistance is small, canopies and atmosphere are closely coupled, and stomata
regulation has significant impact on transpiration. In contrast, in short and
aerodynamically smoother vegetation, such as grassland, the aerodynamic
resistance is higher, canopy and atmosphere are less coupled, and stomata
regulation has less impact in regulating transpiration.
14.4.4 Effects on Other Processes and Interactions
The effects of water stress depend on its timing, duration and severity. A very
severe stress affects almost all processes of the crop, and may lead to total crop loss.
Crops do not usually suffer such severe deficits as to affect their survival, so we will
only consider the effects of mild to moderate water deficits. Effects can be defined
and investigated at different levels of organization. For example, water deficit can
reduce the rates of cell division and expansion, lead to abnormal cell differentiation
processes, which can result in abortion of flowers and/or a reduction in the number
of flower primordia. Water stress can cause hydrolysis of proteins and accumulation
of amino acids which, at the leaf level, reduces CO2 fixation by a direct effect of
stress on photosystems. Water deficit changes increases levels of abscisic acid and
ethylene and reduces levels of indole acetic acid, cytokinins and gibberellins. These
hormonal changes can lead to accelerated senescence. Water stress has little effect
on the respiratory processes, the transport of assimilates and the rate of
development.
Water stress in real field situations, however, does not occur in isolation. In arid
and semi-arid environments, low soil fertility interacts with low rainfall in leading
to water-nutrient co-limitation. Water and nitrogen co-limitation is particularly
Table 14.2 Ear damage (%) caused by heat stress in rainfed wheat crops grown under two sowing
densities and several nitrogen rates. No ear damage was observed when the same treatments were
applied under irrigation
N rate Half density Normal density
None 10 22
Low 23 32
High 33 60
Adapted from Rodriguez et al. 2005. Aust J Agric Res 56, 983–993
important for the yield of cereals in Mediterranean environments. In these environ-

ments, water stress is also combined with heat stress. In wheat crops in south-
eastern Australia, ear damage caused by a heat episode close to flowering was
significant in rainfed crops but not in irrigated crops (Table 14.2). Conditions that
favored more vigorous growth and faster depletion of soil water, including exces-
sive nitrogen supply and higher sowing density, compounded the effects of heat in
the rainfed crops (Table 14.2). The sensitivity of crops to insects and diseases can
increase, diminish or remain unchanged in water-stressed plants. Water-stressed
cotton, for example, was found to be less susceptible to spider mites than their well-
watered counterparts. This was because water deficit increased leaf thickness and
hardness, making them less palatable to mites. In choice-tests, mites showed a
marked preference for leaves from well-watered plants. Root diseases can com-
pound the effect of water deficit.
14.5 Coupling of Photosynthesis and Transpiration
The flow of CO2 into the leaf follows the same path as the flow of water vapor out of
the leaf (transpiration). Any opening or closing of stomata affects the two
processes.
The flux of transpiration (Ep, mol m2 s1) may be expressed as:
ei ea
Ep ¼ 1:6gsc ð14:5Þ
Pat
Where gsc (mol m2 s1) is the stomatal conductance for CO2, ei and ea (both in
kPa) are the vapor pressure in the substomatal cavities and the air outside the leaf,
respectively, and Pát (kPa) is atmospheric pressure. The factor 1.6 is the ratio of
diffusion coefficients for water vapor and CO2 in air at 25 C. A similar equation
(Eq. 14.1) is used for photosynthesis.
Air within the substomatal cavities is usually close to saturation (ei ¼ es at leaf
temperature), so adding and subtracting es at air temperature we may write:
ei ea ¼ es ðTc Þ ea þ es ðTa Þ es ðTa Þ ¼ VPD þ ΔðTc Ta Þ

where VPD is the Vapor Pressure Deficit (Chap. 5), Tc and Ta are the leaf (crop) and
air temperature, respectively, while Δ (kPa K1) is the slope of the saturation vapor
pressure function versus temperature.
Therefore the ratio of net photosynthesis and transpiration will be:
Pn ðCa Ci ÞPat
¼ ð14:6Þ
Ep 1:6½VPD þ ΔðT c T a Þ
Several authors have shown that the ratio Ci/Ca tends to fairly constant values of
0.7–0.8 (C3 plants) and 0.3–0.4 (C4 plants). Assuming that leaf and air temperature
are equal:
Pn Ca ð1 Ci =Ca ÞPat αw
¼ ¼ ð14:7Þ
Ep 1:6VPD VPD
This equation implies that the photosynthesis/transpiration ratio (also called Water-
Use Efficiency, WUE) depends mainly on the vapor pressure deficit: the amount of
carbon fixed per unit of water lost will decrease as the atmosphere is drier or
warmer. This is illustrated in Fig. 14.7 that shows hourly values of cotton WUE
as a function of VPD during summer days in Cordoba, Spain. Here WUE was
calculated as the ratio of canopy photosynthesis and ET. Equation 14.7 also
indicates that WUE increases as Ca increases.
For a current value of Ca ¼ 397 μmol mol1, the coefficient αw would be
5,111–7,667 μmol mol1kPa for C3 plants and 15,334–17,890 μmol mol1kPa
for C4 plants. If we convert these values to biomass production of carbohydrates
and taking into account the relation between daytime VPD and mean VPD, the
16
14
12
WUE (mg CO2/g H2O)
10
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
VPD (kPa)
Fig. 14.7 Relationship between WUE and VPD for cotton in Cordoba (Spain) during summer
days of 2003. WUE was calculated as the ratio of canopy photosynthesis and ET
coefficients would be 5.2–7.7 g dry matter L1kPa for C3 and 15.5–18.09 g dry
matter L1kPa for C4. If we consider also night respiration, the coefficients would
be reduced to around 60 % of the previous values for seasonal estimates.
Therefore, biomass accumulation (g m2 day1) can be calculated as:
αw
ΔB ¼ Ep ð14:8Þ
VPD
The same equation applies to isolated trees or plants, but then transpiration is
expressed in L/tree/day and the increase of biomass in g/tree/day.
Example 14.2 Typical seasonal values of WUE (g dry matter/kg water tran-
spired) for various C3 species in Southern Spain are: olive (3.4), sunflower
(3.5) and wheat (5.3).
The higher value of wheat is due in principle to lower VPD during its cycle
from December to May (average 0.7 kPa) compared with sunflower
(February to July, 1.3 kPa) and olive (January to December, 1.2 kPa). The
theoretical estimates, taking Ci/Ca ¼ 0.75, would be:
Wheat: 6.5 g dry matter L1kPa/0.7 ∙ 0.6 ¼ 5.6 g dry matter/kg water
Olive: 6.5 g dry matter L1kPa/1.2 ∙ 0.6 ¼ 3.25 g dry matter/kg water
Sunflower: 6.5 g dry matter L1kPa/1.3 ∙ 0.6 ¼ 3.0 g dry matter/kg water
Example 14.3 In Example 9.5 we calculated transpiration of an olive tree

with radius 0.5 m in Cordoba, Spain on 21 March with ET0 3 mm day1.
Calculated Ep is 2.07 L/day.
If average VPD is 1.5 kPa the increase of biomass would be:
αw L 6:5 g kPa g
ΔB ¼ Eptree ¼ 2:07 L
¼ 9
VPD tree day 1:5 kPa tree day
14.6 Quantifying the Impact of Water Deficit on Crop

Production
The impact of water deficit on crop production can be quantified using models of
different complexity. Here we illustrate the simple model of Stewart, and the more
complex AquaCrop model.
14.6.1 Stewart Model
The model is based on the empirical relations found between crop evapotranspira-
tion and yield which derives from the association between transpiration and pho-
tosynthesis. These two processes are linked for two reasons. First and most
importantly, both processes are driven by radiation that provides the energy for
water evaporation and CO2 reduction. Smaller canopies capture less radiation, fix
less carbon and transpire less. Second, the flows of both water and CO2 between
crop and atmosphere are partially regulated by stomata. These two mechanisms
account for the correlations between biomass and transpiration. Therefore, if the
harvest index (HI) is known, yield could be calculated as a function of seasonal
transpiration:
X
t¼harvest
Y ¼ HI WUE Ep ð14:9Þ
t¼emergence
where WUE is the biomass produced per unit transpiration (g dry matter/kg water),
which depends mainly on VPD and the crop species (see Eq. 14.10). For simplicity
we will assume that WUE is constant throughout the season and that HI is not
affected by water deficit, but see Eq. 13.10 for a more realistic discussion. As
evapotranspiration is the sum of soil evaporation and transpiration:
!
X
t¼harvest X
t¼harvest
Y ¼ HI WUE ET Es ð14:10Þ
t¼emergence t¼emergence
We arrive to a linear relationship between yield and total evapotranspiration (ETsea).

The intercept is negative, and the x-intercept is an approximation to soil evaporation
while the slope is the product of HI and WUE.
We are interested in computing yield as a function of water used (ETsea). A
simple approach that requires a priori estimates of ET and yield in unstressed crops
(ETxsea and Yx) is the model of Stewart which calculates the relative reduction in
yield as a linear function of the relative reduction in ET:

Y ET
1 ¼ Ky 1 ð14:11Þ
Yx ET x
where Ky is a coefficient of sensitivity to water stress. If HI is not affected by water

stress and we change ETsea while keeping soil evaporation constant, Ky will depend
only on the ratio soil evaporation: transpiration:
Essea
Ky ¼ 1 þ ð14:12Þ
Epsea
On the other hand, if soil evaporation if proportional to ET, Ky should tend to

1. These relations are not valid for severe water stress that reduces HI (see
Eq. 13.10). In general, the empirical values reported for Ky are in the range of
0.8–1.5.
The Stewart model which was adopted by the FAO Manual 33 of Doorenbos and
Kassam, has been used as a first approximation for calculating yield as a function of
ET but has a number of limitations:
– Variation in WUE: For a given environment the actual mean WUE will depend
on when actual transpiration occurs due to the inverse dependence of WUE on
Vapor Pressure Deficit. Under water stress the pattern of water use will change
as compared to the unstressed situation which may lead to important changes
in WUE.
– Calculation of unstressed yield: In the absence of local, reliable data a good
alternative would be to use a crop simulation model, calibrated and verified for
local conditions.
– Irrigation method: the irrigation method affects evaporation from the soil sur-
face. Obviously different Es leads to different ET for the same transpiration (the
variable which is directly related to biomass and yield). The Ky coefficient of
Eq. 14.6 can be considered independent only if Es and Ep are affected equally by
a reduction in the supply of irrigation, which is very unlikely.
– Acclimation: crops responses to water stress are conditional to the previous
history of stress. In addition, the effect of water deficit in perennials can be
carried over from one season to the next.
These limitations are common to any empirical model used to generate universal
production functions. However, their use can be advantageous in those cases where
a high accuracy is not required, and in many cases these empirical models are the
only available alternative for practical purposes.
14.6.2 The AquaCrop Simulation Model
AquaCrop (www.fao.org/nr/water/aquacrop.html) is a simulation model that pre-

dicts yield as a function of water supply. It is based on the relation between B and T
and on the constancy of WUE (for a given species) if normalized by the evaporative
demand (reference ET). After seedling emergence, the model calculates the rate of
canopy expansion, and the T and E components of ET by computing soil water
extraction on a daily basis. Based on the amount of T, the model then calculates the
increment of B which is accumulated until harvest. As the canopy develops and the
crop grows driven by thermal time, different developmental stages occur and,
depending on the crop, the harvest index is built up during the pertinent period
and yield is computed at harvest date.
Water stress affects the computations via reductions in canopy expansion first,
followed by reductions in stomatal conductance and acceleration of canopy
senescence. Different crop-specific thresholds in root zone water content are

defined for each process, and the HI is also modulated by water deficits, generally
decreasing under severe stress. The model gives an estimate of water-limited
potential yield without any other limiting or yield-reducing factors. It has a soil
fertility module to make adjustments to non-optimal fertility conditions which are
common in many agricultural systems.
To obtain accurate predictions, AquaCrop should be calibrated and validated
with experimental data in the environment where it will be utilized. Its description,
together with its numerous applications may be found in FAO 66.
Bibliography
Fereres, E., Orgaz, F., Gonzalez-Dugo, V., Testi, L., & Villalobos, F. J. (2014). Balancing crop
yield and water productivity tradeoffs in herbaceous and woody crops. Functional Plant
Biology, 41, 1009–1018.
Gonzalez-Dugo, V., Durand, J.-L., & Gastal, F. (2011). Water deficit and nitrogen nutrition of
crops. Sustainable Agriculture, 2, 557–575.
Sadras, V. O., & Richards, R. A. (2014). Improvement of crop yield in dry environments:
Benchmarks, levels of organization and the role of nitrogen. Journal of Experimental Botany,
65, 1981–1995.
Tanner, C. B., & Sinclair, T. R. (1983). Efficient water use in crop production: Research or
re-search. In H. M. Taylor et al. (Eds.), Limitations to efficient water use in crop production
(pp. 1–27). Madison: ASA.
Chapter 15
Limitations to Crop Productivity
Abstract Environmental factors including water stress, nutrient deficiency, high or

low temperatures, chemical (Al-toxicity, salinity) and physical soil constraints
(e.g. compaction), and biotic factors reduce crop yield. Deficit of water and
nitrogen, and soil constraints generally have a larger impact on canopy size and
duration, hence growth reductions are closely linked with reduced intercepted
radiation; radiation use efficiency is generally less responsive to stress. Extreme
temperatures at critical stages usually reduce harvest index and yield. Other stresses
can be neutral, positive or negative for harvest index; this depends in particular on
the nature, timing, intensity and duration of stress and its relative impact on total
and harvestable biomass. Yields are also reduced by the action of biotic agents such
as pests, diseases and weeds. Potential yield losses to biotic stresses may be quite
high but are decreased by crop protection practices.
15.1 Introduction
In Chap. 13 we have analyzed crop yield as a function of temperature and radiation

interception. This “potential” yield will be higher than actual yields as other biotic
and abiotic factors come into play. The main abiotic factors that reduce crop yields
are water stress (already reviewed in Chap. 14), nutrient deficits, adverse chemical
conditions (salinity, acidity), water excess and meteorological events (extreme high
or low temperatures, hail, wind). Biotic factors are weeds, animal pests and
diseases.
V.O. Sadras (*)


Agriculture, DOI 10.1007/978-3-319-46116-8_15
In this chapter we’ll revise the main impact of these factors on crop productivity.
Some of them will be treated in more detail in specific chapters of this book
(salinity, nutrient deficits, frost).
15.2 Nutrient Deficiency and Soil Related Limitations

of Productivity
Deficit of water and nutrients reduce biomass by primarily reducing leaf area index
and radiation interception; under severe stress radiation use efficiency is also
reduced. This is because tissue expansion is more sensitive to both water and
nutrient deficit than leaf photosynthetic rate. Indeed, a common short-term response
of water and nitrogen stressed plants is to accumulate carbohydrates as the restric-
tion in expansion is more severe than the restriction in photosynthesis, leading to
transient excess of reduced carbon.
The effects of nitrogen supply on crop growth and yield can thus be explained in
terms of its effects on interception and efficiency in the use of radiation. Nitrogen
deficit reduces crop LAI by reducing tillering or branching, and leaf expansion
(Fig. 15.1). Reduced leaf size of nitrogen-deficient crops is associated with reduced
rates of cell division and expansion (Table 15.1). Nitrogen deficit can accelerate
leaf senescence (Fig. 15.2), further contributing to reduced radiation interception
and photosynthesis. Rubisco and light-harvesting proteins involved in photosyn-
thesis represent 60 % of the leaf N content, hence the link: shortage of nitrogen !
less Rubisco ! less photosynthesis.
30 600
MAXIMUM PHOTOSYNTHESIS ( μmol/m2/s)
25
FINAL LEAF AREA (cm2) for LEAF15
500
20
400
15
300
10
200
5 Photoshyntesis
Leaf area
0 100
N1 N2 N3 N4 N5
NITROGEN SUPPLY
Fig. 15.1 Nitrogen deficit reduces photosynthesis and leaf size. Sunflower plants where grown
under five nitrogen regimes, with extreme treatments receiving 0.25 (N1) and 7.5 g N per plant
(N5) (Adapted from Connor et al. (1993) Aust J Plant Physiol 20: 251–263)
15 Limitations to Crop Productivity 207
Table 15.1 Nitrogen deficit Number of cells Area per cell Leaf area
reduces cell number and size,
106 μm2 cm2
hence leaf area of sunflower
High N 56 554 302
Low N 33 443 147
Data for leaf number 10 at full expansion under high or low N
supply are shown
Adapted from Trapani et al. (1999) Ann. Bot. 84:599–606
Fig. 15.2 Nitrogen
LEAF CHLOROPHYLL (%)

deficiency accelerates leaf Fertilized
100
senescence of wheat. Leaf
chlorophyll, measured in
the youngest expanded leaf, 80 Not fertilized
is expressed as percentage
of that of fertilized crops
(Adapted from Caviglia and 60
Sadras (2001) Field Crops
Res 69:259–266)
40
40 60 80 100 120
DAYS AFTER SOWING
The effects of water supply on crop growth and yield (Chap. 14) can also be
explained in terms of interception and efficiency in the use of radiation. In response
to water deficit, plants have largely irreversible responses such as reduced tillering
or branching, and reduced leaf expansion. Reduced stomatal conductance and
wilting are transient crop responses to water deficit that reduce both crop water
use and photosynthesis. Figure 15.3 illustrates the saw tooth pattern of radiation
interception in alternating dry-wet periods with transient wilting and recovery after
irrigation. In addition to the individual effects of nitrogen and water, these resources
often interact in complex ways. In wheat crops growing under a combination of
irrigation and fertilizer regimes, RUE was 1.8 g/MJ in rainfed, unfertilized crops
and increased to 2.1 g/MJ with nitrogen fertilization. Weekly irrigation did not
improve RUE in unfertilized crops, but irrigation combined with nitrogen fertili-
zation increased efficiency to 2.5 g/MJ.
In common with water and nitrogen supply, the reduction in crop growth in
response to physical and chemical soil constraints is mostly mediated by reduced
canopy size and interception of radiation, whereas radiation use efficiency is less
responsive to stress. This is illustrated in two examples. Soil compaction is a
common problem, often caused by tillage, machinery traffic, and loss of organic
matter, which is reflected in increased soil bulk density. Its effects on the crop are
twofold: it hinders the emergence and establishment of seedlings and slows down
growth of the root system and depresses shoot growth. The effect of compaction on
root growth is quantified by the penetration resistance, which can be measured with
a penetrometer. For a particular soil, penetration resistance is directly proportional
to the apparent density and inversely proportional to the water content of the soil.
0.8
FRACTIONAL INTERCEPTION (f)
0.7
0.6
0.5
0.4
0.3
0.2
0.1
Irri Irri
0.0
60 70 80 90 100 110 120
DAYS AFTER SOWING
Fig. 15.3 Dynamics of fractional radiation interception of peanut crops. The saw-tooth pattern of
radiation interception results from transient wilting and recovery after irrigation. During this
period LAI increased from 0.5 to 2. Arrows show time of irrigation (Adapted from Matthews
et al. (1988) Exp Agric 24:203–213)
Al-tolerant
INTERCEPTED PAR
1.0 1.0
RELATIVE RUE
0.8 Al-tolerant 0.8

FRACTION
0.6 0.6
0.4 0.4 Al-sensitive
0.2 Al-sensitive 0.2
0.0 0.0
0 1 2 3 0 1 2 3
EXCHANGEABLE SOIL Al (cmol kg–1)
Fig. 15.4 Effects of soil aluminium toxicity on (a) intercepted radiation and (b) radiation use
efficiency of wheat crops (Adapted from Valle et al. (2009) Field Crops Res 114:343–350)
Therefore the effects of compaction are more severe when the soil is dry (Chap. 17).
In the Mallee region of south-eastern Australia, sandy soils often develop a
compacted layer at 0.2–0.3 m depth, thus restricting root proliferation, and water
and nitrogen uptake below this depth. Crops in compacted soil were compared with
crops in soils were deep ripping was used to remove the constraint. Wheat yield in
compacted soil ranged from 1.2 to 2.9 t/ha, and ripping improved yield up to 40 %.
The reduction in yield associated with compaction was fully accounted for by the
reduction in leaf area index and intercepted radiation, whereas radiation use
efficiency was unaffected by soil condition. Aluminum toxicity reduces yield in
acid soils (pH<5.8), which represent about 30 % of agricultural soils worldwide.
Growth analysis of Al-tolerant and Al-sensitive wheats in southern Chile showed a
marked reduction in intercepted radiation with increased concentration of Al in
soils, a largely unresponsive RUE in the tolerant wheat, and reductions in RUE in
sensitive wheat only at high Al concentrations (Fig. 15.4).
15.3 Climatic Accidents
We will consider the effect of extreme low temperatures in Chap. 29 and the effects
of wind in Chap. 27.
15.3.1 High Temperature
In general the main effect of extreme high temperature is a reduction in HI due to

reduced pollination or seed abortion. For instance, the number of grains in winter
cereals decreases with canopy temperature above 31 C. On the other hand, high
temperatures will promote leaf senescence and shorten the grain filling period.
In Mediterranean climates low water availability is coincident with high tem-
perature so low transpiration due to water stress amplifies heat stress as canopy
temperature increases above air temperature.
15.3.2 Hail
Hail is a form of solid precipitation consisting of ice balls or lumps of ice which
originate in strong thunderstorm clouds (cumulonimbi). The hailstones show diam-
eters typically between 5 and 12 mm. Hail occurrence is localized, with areas
affected from a few hectares to hundredths of has.
Occurrence of hail is more frequent in mid-latitudes (inland areas, elevated
regions) than in the tropics despite the higher frequency of thunderstorms. Some
areas where hailstorms are common are Northern India and some regions of China
and Central Europe. During the year hail will occur mostly during spring and
summer.
Damage to the crop is due to the impact of hailstones which velocity will be
proportional to size. For a 1 cm diameter hailstone, terminal velocity could reach
9 m/s while an 8-cm hailstone would fall at 48 m/s. Apart from the direct physical
damage that destroys leaves or reproductive structures, wounds facilitate the infec-
tion by pathogens. The effect of defoliation due to hail on crop yield depends on the
development stage when hail occurs and the level of defoliation. Partial early
defoliation may be compensated by increased dry matter allocation to leaves
resulting in a small reduction in yield. Full defoliation before anthesis may be
catastrophic for yield. Damage to reproductive structures leads to reduced Harvest
Index in proportion to the number of structures affected. Partial damage of fruits
reduces their commercial value.
High value crops may be protected by covering with anti-hail nets. In field crops
or extensive fruit production the main protection alternatives are:
– anti-hail rockets that use silver iodide
– ground generators that produce smoke with silver iodide
– anti-hail guns: they use shockwaves directed at the hail storm
– crop insurance
15.4 Waterlogging
High soil water content implies a limited supply of oxygen for root respiration, so
root functioning (absorption, growth) is impaired. If the soil is saturated the roots
decompose starting from the tips, plant stops growing and eventually it will die.
Most often crops will suffer from temporal waterlogging events and will recover
partially afterwards, but the poor root system will have a low capacity for water and
nutrient absorption. The overall effect will be reduced yield.
Waterlogging also contributes to N losses by denitrification (see Chap. 24), the
conversion of nitrate to volatile N compounds. Low N uptake will cause symptoms
of N deficiency.
Waterlogging may be prevented by improving the drainage system of the field or
using raised bed planting.
15.5 Biotic Factors
Biotic agents causing crop damage (generically known as pests) are weeds, animal
pests (arthropods, nematodes, gastropods, rodents, birds), pathogens (fungi, bacte-
ria) and viruses.
The effect of pests and diseases on crop growth can also be analyzed in terms of
capture and efficiency in the use of radiation, and harvest index (Eq. 13.3). Insects
that feed on reproductive structures, such as cotton bollworms, have a primary
effect on harvest index. In extreme cases of uncontrolled infestation for example,
cotton crops can accumulate large amounts of biomass with little fruit set, hence
reduced harvest index and yield. Defoliators reduce leaf area and intercepted
radiation, whereas some diseases can also reduce photosynthetic rate of individual
leaves and RUE. Spider mites for example, pierce the leaf epidermis with needle-
like mouthparts and feed on mesophyll and palisade cells, thus reducing leaf
photosynthesis and RUE (Fig. 15.5). Comparisons of wheat crops protected with
fungicides, and unprotected crops exposed to damage by foliar pathogens showed
that growth reduction was mostly associated with reduced healthy leaf area, with a
secondary contribution of reduced radiation use efficiency.
Fig. 15.5 Spider mites feed 1

on mesophyll and palisade
RELATIVE RUE
cotton leaves, reducing leaf
photosynthesis and
radiation use efficiency
(Adapted from Sadras and
Wilson (1997) Crop Sci
37:481–491)
0
MITE ABUNDANCE
15.5.1 Arthropods
Many species of insects and other arthropods are present in the agro-ecosystem but
only a few are important pests, and may cause complete yield loss. Insects are
six-legged invertebrates that usually undergo metamorphosis during development.
Adult insects have three body regions (head, thorax and abdomen), three pairs of
legs, one pair of antennae, complex mouthparts, and frequently two pairs of wings.
The skin of an insect is the exoskeleton, which covers the whole body.
All insects have an egg and an adult stage. Complete metamorphosis includes
four stages (egg, larva, pupa and adult). The most common, foliage-eating insect
pests are larvae of Lepidoptera (butterflies and moths) and larvae and adults of
Coleoptera (beetles). Aphids, leafhoppers and thrips not only feed on the crop but
are also the main vector for transmission of plant virus diseases.
Many insect species are predators or parasites of other insects and are, thus,
beneficial.
15.5.2 Plant Pathogens
Plant pathogens (fungi, bacteria, virus, nematodes, etc.) affect crop plants by
altering the following processes:
– photosynthesis: destruction of photosynthetic tissue, degradation of chloroplasts,
leaf senescence, yellowing, etc.
– water and nutrient transport: destruction of roots, formation of root galls and root
knots, impaired root absorption, destruction or blocking of xylem tissue, damage
to leaf cuticles or stomatal function (higher transpiration), altered phloem
transport.
– plant respiration increases after infection which contributes to depleting the
plant reserves.
– membrane permeability also increases causing leaf damage by loss of nutrients

and entry of toxins
– changes in transcription and translation of nucleic acids alter plant function and
structure and the synthesis of enzymes or substances involved in plant resistance.
The disease starts with a primary infection (first in the season) due to primary
inoculum (spores or fungal mycelium) that overwinters. Overwintering is the ability
of a pathogen to survive from one growing season to the next.
The probability of a disease epidemic is proportional to the amount of inoculum
and to the proximity to its host. Primary infection occurs when the pathogen is in
contact with a susceptible host under suitable conditions. The pathogens enter
directly through the surface of the plant or through wounds or natural openings.
Bacterial and fungal pathogens usually require free water for spore germination, so
infection is favored by wet periods with high air humidity and by wet canopies.
Dissemination of the pathogen from an inoculum source to a host can occur by
wind, splashing rain, runoff, insects, infected seeds or seedlings, etc. Fungi grow
and spread within their host by means of mycelium, and eventually produce spores
on or within the infected tissue. These spores lead to secondary infections during
the season. Bacteria spread in the plant by rapidly increasing the population. Then,
when fissures develop on infected tissue, the cells (secondary inoculum) are
exposed to the environment and thus, dissemination may proceed.
The secondary infection cycle can be repeated many times during the growing
season, depending on the biology of the pathogen and its host and environmental
conditions.
15.5.3 Yield Losses Due to Pests
Two crop yield loss rates may be differentiated. The loss potential characterizes the
risk that the agent exerts on crop yield in a no-control scenario. The actual losses are
those occurring despite the crop protection practices. The efficacy of the crop
protection practices may be evaluated as a percentage of potential losses prevented.
The potential and actual loss are quite variable depending on the crop species and
the region considered. Among crops the loss potential of all biotic factors world-
wide varies between 50 % (wheat) and more than 80 % (cotton). Actual losses are
estimated at 26–31 % for soybean, wheat, maize and cotton, and 37–40 % for
potatoes and rice, respectively (Table 15.2). Overall, weeds have the highest loss
potential (23–40 %) with animal pests and pathogens being less important (9–37 %
and 9–29 %, respectively). Although viruses cause serious problems in potatoes and
sugar beets in some areas, worldwide losses due to viruses average 6–7 % on these
crops and <1–3 % in other crops. The efficacy of crop protection lies between
43 and 65 % for the different crops. Global efficacy in weed control (67–80 %) is
Table 15.2 Yield losses (%) due to pests at a global scale. Potential and actual yield losses are
those occurring in a no-control and a current control scenario, respectively. The efficacy of control
is the percentage of losses prevented by current control measures
Wheat Rice Maize Potato Soybean Cotton
Weeds Potential 23 37.1 40.3 30.2 37 35.9
Actual 7.7 10.2 10.5 8.3 7.5 8.6
Efficacy 67 73 74 73 80 76
Animal pests Potential 8.7 24.7 15.9 15.3 10.7 36.8
Actual 7.9 15.1 9.6 10.9 8.8 12.3
Efficacy 9 39 40 29 18 67
Pathogens and viruses Potential 18.1 15.2 12.3 29.3 12.4 9.3
Actual 12.6 12.2 11.2 21.1 10.1 7.9
Efficacy 30 20 9 28 19 15
Total Potential 49.8 77 68.5 74.9 60 82
Actual 28.2 37.4 31.2 40.3 26.3 28.8
Efficacy 43 51 54 46 56 65
Adapted from Oerke (2006)
much higher than that of animal pests (9–67 %) or diseases (9–30 %). These values
have to be taken only as indicative as they are based on estimates of reference yields
(not affected by pests).
Bibliography
Denison, R. F. (2012). Darwinian agriculture: How understanding evolution can improve agri-
culture. Princeton: Princeton University Press.
Oerke, E. C. (2006). Crop losses to pests. Journal of Agricultural Science, 144(1), 31–43.
Chapter 16
Sowing and Planting
Francisco J. Villalobos, Francisco Orgaz, and Elias Fereres
Abstract Successful crop establishment depends on several factors at the time of

sowing (soil water content, soil structure and soil temperature, seed viability,
presence of pests). Therefore decisions regarding the date and depth of sowing,
the planting density, the spatial arrangement of plants and other cultural techniques
(irrigation, fertilization, application of pesticides) will be critical. Sowing date
should match the growing cycle to the best possible environmental conditions for
the crop. Early spring sowings improve water use efficiency in Mediterranean areas
but increase the risk of attacks by biotic factors. Seeding rate is a function of single
seed mass, desired planting density, seed viability and expected fraction of emerged
plants. Trees and some annuals species are sown in nurseries where they grow for
some time until they are transplanted to the field. The best time for planting trees is
autumn when they are dormant and the risk of desiccation is minimal.
16.1 Introduction
By sowing or planting the farmer intends to ensure good crop establishment and get
the right conditions for growth, development and yield. For many crops the
establishment phase (germination, emergence and early seedling growth) is the
most critical phase of the cycle. To succeed, the farmer must make a series of

F. Orgaz

Agriculture, DOI 10.1007/978-3-319-46116-8_16
decisions related to the amount of seed to be used, the method of planting and the
spatial distribution of seeds, the planting date, the application of pesticides or
performing additional tasks like irrigation.
16.2 Crop Emergence
Some time after sowing the seeds, the seedlings will emerge from the soil. The
duration of this period and the success, i.e. the fraction of seeds leading to emerged
seedlings, depend on several factors:
– Seed viability: A viable seed is one able to germinate under suitable conditions.
Viability decreases with time from the harvest of the seeds, and occurs in parallel
to the loss of reserve substances (e.g. lipid oxidation in sunflower seeds). In some
species there are mechanisms that delay seed germination, such as the presence
of germination inhibitors or waterproof coats. For these cases germination may
be improved by scarification (mechanical abrasion or chemical treatment with
acids to improve the permeability of seed coats) or stratification (placing the
seeds between layers of cold (1–5 C) moist soil).
– Soil water content: Germination is a process that begins with water uptake by the
dry seed (imbibition). If the soil is dry or the contact seed-soil is loose the
transport of water from the soil to the seed is prevented and germination does not
proceed. After germination, the radicle expands, which contributes to guarantee
the supply of water to the seedling. The increase in depth of the radicle occurs in
advance to hypocotyl growth, so that, at the time of emergence, root depth
normally exceeds 10 cm.
– Temperature: Along with depth, temperature will determine the duration of the
sowing-emergence period. If this period is too long, the likelihood of attacks by
pathogens or soil insects increases. Table 16.1 shows the average values of
thermal time from sowing to emergence and its base temperature for a series
of annual crops.
– Soil structure: the presence of a surface crust or excessive soil compaction above
the seed makes emergence difficult as they prevent the expansion of the hypo-
cotyl, especially if the soil is dry (see Chap. 17). A greater amount of seed or
wetting the soil can contribute to mitigating the effects of the surface crust.
– Presence of pests or pathogens: during emergence and initial seedling growth,
attacks by insects or soil fungi can often lead to a failure of crop establishment.
To avoid this problem fungicides (seed treatment) and/or insecticides (seed
and/or soil treatments) are applied.
– Oxygen concentration in the soil: the processes of germination and emergence
use energy derived from the seed reserves through respiration, which is an
aerobic process. This is why the fraction of emerged plants can be greatly
reduced in waterlogged soils.
16 Sowing and Planting 219
Table 16.1 Base Thermal time Tb

temperature (Tb, C) and
Cd C
thermal time (TT, Cd)
required to complete the Amaranthus 32 11.7
phase sowing-emergence for Barley 120 0
a series of crop species Bean 52 10.6
Buckwheat 37 11.1
Castor bean 95 12.5
Chickpea 94 4.5
Cotton 60 15.5
Cucumber 40 15.5
Faba bean 200 1.2
Lentil 90 1.4
Linseed 89 1.9
Maize 75 8
Melon 52 15.5
Millet (finger) 40 13.5
Millet (foxtail) 42 10.9
Millet (pearl) 40 11.8
Millet (proso) 45 10.4
Oats 132 1.6
Pea 110 1.4
Peanut 76 13.3
Pepper 135 13
Rapeseed 79 2.6
Rye 91 2.2
Ryegrass 130 2
Safflower 70 7.4
Sesame 21 16
Sorghum 74 8
Soybean 70 9.9
Sunflower 67 7.9
Tomato 57.5 9.3
Trifolium spp 150 0
Watermelon 55 15.5
Wheat 110 0
The values shown in this table have been obtained experimentally
under field conditions with high water content in the soil and
planting depth of about 3 cm
Adapted from Angus et al. (1980) Field Crops Res 3:365–378 and
Moot et al. (2000) New Zealand J Agric Res 43:15–25
16.3 Decisions Related to Sowing
The farmer has to make a series of operational decisions before sowing the crop,
such as the date of sowing, the amount of seed (seeding rate), the sowing depth, the
planting pattern (row distance, plant spacing within the row). Additional operations
may be required such as fertilization, irrigation, application of pesticides or tillage.
The method of sowing which may imply the selection of the planting machinery
(e.g. seed drill) is a key strategic decision for the farmer.
16.4 Sowing Date
The choice of sowing date has to ensure that the crop cycle matches the most
suitable period for growth and yield. A first limitation on the growth of a crop is its
ability to survive when exposed to low temperatures. In mid latitudes, this limita-
tion allows classifying crops into two categories:
(a) Autumn sown crops: species able to withstand frost and grow at low temper-
atures (winter cereals like wheat or rye, rapeseed, flax, faba beans, beets, etc.).
They have a low base temperature.
(b) Spring planting: species with high base temperature (corn, cotton, soybeans).
They are damaged even by low temperatures above freezing.
Regardless of the type of crop, early plantings have several advantages:
(a) The Water Use Efficiency is inversely proportional to Vapor Pressure Deficit
(Chap. 14). If crops are grown in a period of low evaporative demand (early
planting date) they produce more biomass and will require less water in
irrigated conditions. This is especially important in Mediterranean climates as
rainfall decreases and evaporative demand increases from spring to summer.
(b) The grain filling process is more efficient and longer if the temperature is not
too high. If we avoid this process to occur under very warm conditions the
harvest index will increase and so yield will do.
(c) In some spring-summer crops the cycle may be terminated by cold temperatures
in autumn (e.g. cotton). Therefore early sowing favors crop maturing before the
low temperatures stop crop development. Additionally, autumn precipitation
may adversely affect crop quality.
(d) In some horticultural crops the price is directly proportional to precocity
(melon, watermelon, etc.) thus early sowing allows increased revenue.
Early planting may be limited by possible negative effects later in the
growing season. For winter cereals it is extremely important to avoid frost
during anthesis, and this is achieved by preventing excessively early plantings,
and/or using longer season cultivars (winter types with large vernalization
requirement). The environmental conditions at the time of sowing may also
restrict early sowings:
– If the temperature is too low, the crop takes a long time to emerge and
establish, allowing the attack of biotic agents and leading to a significant
reduction in the fraction of established plants. Also under low temperature
conditions, weed competition will be more severe. In general we must sow
when the temperature is such that the sowing-emergence period does not
exceed 15–20 days. To calculate the time from sowing to emergence as a
function of temperature for different crops, we can use the information
presented in Table 16.1.
– Water content in the soil at the time of sowing should ensure seed imbibition
and the water supply to the seedlings after they emerge. In Mediterranean
areas the autumn sowing is usually delayed until the rains have been sufficient
and evaporative demand is low. Sowing over partially dry soil may cause
relatively early emergence (with high evaporative demand) but seedlings may
die if the rains do not continue.
16.5 Seeding Rate and Planting Density
The amount of seed to be applied per unit area (seeding rate) depends on the cost of
the seed and on planting density desired.
(a) Cost of seed: In general the cost of seed is a low fraction of total cultivation
costs. However, the consequences of using a small amount of seed or low
quality seed may be extremely negative (see the importance of obtaining a
suitable planting density in Chap. 12). In general we should use high quality
seeds. The indices used to characterize the quality of the seed are viability
(germination percentage) and purity (proportion of the seed belonging to the
acquired cultivar). The seed should be free of pests, diseases and weed seeds
and present a suitable size. The probability of emergence and the growth rate of
the seedling afterwards are both proportional to seed size.
(b) Density: The amount of seed used must be sufficient to ensure the emergence
and establishment of a sufficient number of seedlings. The excess of seed
applied depends on various factors (seed viability, soil structure, pathogens,
water content, etc.).
To calculate the seeding rate (g m2) we can use:
pu Dp
QS¼ ð16:1Þ
f1 f2
where pu is the mass of each seed (g/seed), Dp is the desired planting density (plants
m2), f1 is the viability (fraction) and f2 is the fraction of viable seeds that become
established plants. The values of pu can be measured directly by weighing a known
number of seeds or estimated using Table 16.2, which also shows intervals of Dp for
various crops. Viability (f1) depends largely on the quality of the seed, and is
Table 16.2 Mass per seed, planting density and minimum amount of seed required for different crop species
222
Mass (fresh) per seed Planting density Seed rate

mg/seed Plants/m2 kg seed/ha
Crop Min Max Average Min Max Average Min Max Average
Alfalfa 2.1 2.3 2.2 400 500 450 8 12 10
Cotton 100 120 110 5 10 75 5 12 8
Lupin (white) 200 320 260 30 60 45 60 192 117
Lupin (yellow) 100 130 115 30 60 45 30 78 52
Oat 30 45 37.5 130 250 190 39 113 71
Safflower 30 50 40 25 50 37.5 8 25 15
Barley 30 50 40 150 230 190 45 115 76
Rye 25 35 30 140 250 195 35 88 59
Rape 2.5 4 3.25 50 180 115 1 7 4
Chickpea 50 70 60 25 45 35 13 32 21
Sunflower (oil) 100 140 120 6 12 9 6 17 11
Sunflower (seed) 150 180 165 4 7 5.5 6 13 9
Pea 125 300 212.5 35 100 67.5 44 300 143
Faba bean 350 800 575 15 50 32.5 53 400 187
Bean 130 500 315 25 10 17.5 33 50 55
Lentil 20 80 50 100 150 125 20 120 63
Flax 5 7 6 100 400 250 5 28 15
Maize 350 400 375 7 10 8.5 25 40 32
Sugar beet 20 20 20 6 9 7.5 1 2 2
Soybean 100 200 150 15 60 37.5 15 120 56
16
Sorghum (grain) 20 30 25 10 13 11.5 2 4 3
Sorghum (forrage) 20 30 25 80 140 110 16 42 28
Clover 0.6 0.8 0.7 500 900 700 3 7 5
Wheat 30 45 37.5 150 250 200 45 113 75
Triticale 30 50 40 180 220 200 54 110 80
Vetch 20 30 25 200 300 250 40 90 63
Potato 23,000 30,000 27,000 7.5 13.3 10 1725 3990 2700
Sowing and Planting
Garlic 1500 6000 3500 20 28 24 300 1680 840

Garlic white type 1000 8000 5000 20 28 24 200 2240 1200
Onion 4 10 7 60 90 80 2.40 9.00 5.60
Pepper 5 10 7 4 6 5 0.20 0.60 0.35
Tomato 2.2 3.3 2.8 2 6 4 0.04 0.20 0.11
Melon 29 47 38 0.4 1 0.7 0.12 0.47 0.27
Watermelon 35 60 47 0.25 1 0.75 0.09 0.60 0.35
Leek 2.5 3.2 2.8 16 38 25 0.40 1.22 0.70
223
usually above 0.9 for certified seed. The value of f2 depends greatly on the state of
the soil at planting, sowing depth and environmental conditions after planting.
Under adverse conditions f2 will be proportional to seed size.
Example 16.1 We will calculate the seeding rate for wheat to obtain a density
of 150 plants m2 if the percentage of viable seeds that emerge is 90 % and the
seed has a viability of 0.95.
In Table 16.2 we see that wheat seed mass is between 30 and 45 mg. If we
take an intermediate value (37 mg/seed) the seeding rate should be:
37 103 150
QS ¼ ¼ 6:5 g m2 ¼ 65 kg=ha
0:9 0:95
16.6 Sowing Depth
The more appropriate sowing depth depends on the conditions of temperature and
water content of the soil. In general, soil water content increases with depth while
temperature and oxygen availability decrease. The greater the sowing depth the
greater expansion of the hypocotyl required to reach the soil surface. If the depth is
excessive reserves would be exhausted before emergence. Therefore, larger seeds
allow deeper sowings. Very large seeds (faba beans, beans) allow depths up to
about 15 cm, while the medium sized seeds (winter cereals, sunflower, cotton)
should not exceed around 10 cm sowing depth and small seeds (onion, carrot) allow
less than 3 cm. In the latter case it is difficult to ensure adequate soil water content
in the surface layer, requiring irrigation for successful emergence.
The rules about sowing depth may show remarkable exceptions. For instance, in
the very dry inland area of the Pacific Northwest of the USA winter wheat is sown
in late summer at depths as large as 20 cm to ensure water supply to the seed and
crop emergence.
16.7 Planting Pattern and Sowing Method
Crop plants are usually sown in rows at spacing between 0.15 and 0.20 m (cereals,
rapeseed) and 1 m (e.g. cotton). Wide separations between rows were in many cases
required for mechanical control of weeds. The appearance and use of herbicides has
allowed reducing the inter-row spacing, which contributes to increase the radiation
interception.
In many species yield is relatively independent of the distance between plants
(e.g. winter cereals). In some (garlic, beets) excessive crowding can lead to a
reduction in yield or quality of the harvested product.
The sowing method to use depends on the type of crop, soil conditions and
available machinery. The methods used are:
(a) Broadcasting: the seeds are randomly distributed in the field. The application
can be done by hand, using centrifugal fertilizer broadcasters or an airplane, as
in the case of rice. Usually a broadcasting sowing requires further operations to
bury the seeds, and carries a high cost of seed, poor distribution uniformity and
irregularity in the sowing depth.
(b) Sowing in furrows: It is done by opening furrows in the soil and depositing the
seeds inside, which is usually performed with a seed drill. This drill has the
necessary equipment to open the furrow (shoe type, hoe type or disk), deposit
the seeds and close the furrows (plank, disks). In some cases, to allow post-
emergence tillage and achieve optimum planting density, crop rows are distrib-
uted non-uniformly as in the case of the paired lines, in which the lines are
grouped in close pairs, separated from the next pair by enough distance to allow
inter-row tillage operations.
Precision seed drills provide a better distribution of seeds which saves seed, and
in some cases, avoids the need for plant thinning (e.g. sugar beet).
16.8 Additional Operations
At the time of sowing, other farming operations that contribute to crop establish-
ment may be performed:
(a) Irrigation: The application of irrigation may be needed to ensure germination
and emergence. In soils that form surface crust, more than one irrigation
application may be needed to prevent hardening of the crust.
(b) Tillage: The pre-planting tillage should contribute to the formation of a suitable
seedbed. This entails small aggregates in the surface and sufficient soil water
content in the upper layer. In some methods of sowing, the seed once deposited
in the soil should then be covered by harrowing. In other cases it may be
necessary to slightly compact the soil surface to ensure water supply to the
seed (e.g. roller pass in small seed crops). Soil compaction after sowing also
contributes to soil warming (Chap. 6) and thus to faster emergence.
(c) Fertilizers and pesticides: It is common to apply fertilizers (P, K and some N)
and other products (e.g. pesticides) while preparing the seed bed. Some seed
drills allow localized fertilizer application at sowing time which can be of great
interest in poor soils, especially for P and K (Chap. 26). Other products (soil
insecticides, pre-emergence herbicides) may be also applied along with
the seed.
16.9 Transplanting of Annuals
In the case of some horticultural crops, the seedlings are grown at a place (nursery)
during some time and then they are transplanted to the field. The need for trans-
plantation may be due to:
(a) High seed cost, poor germination success and/or delicate seedlings: The con-
ditions in the nursery can be manipulated to provide a suitable environment for
the seedlings. This can be achieved by soil heating or using plastic or glass
covers. The associated cost will be acceptable in the nursery because of its
small size.
(b) The need to shorten the cycle: In the nursery we can maintain proper conditions
of moisture and temperature, hastening the crop cycle in a time when external
conditions are unfavorable.
The structures used as nurseries range from natural shelters to greenhouses. In
the nursery the seeds are planted at high density. The seedlings are maintained in
the nursery until its final transplanting to the field. During that time some thinning
of plants may be required to avoid etiolation.
16.10 Transplanting and Grafting of Trees
Trees of agricultural interest are usually fruit trees with planting densities ranging
from 50 to more than 1000 trees/ha. Some species with industrial interest
(e.g. rubber tree, cork oak) and ornamental trees and shrubs may be also of
commercial value. Plantations of fruit trees are established to last for long, typically
more than 15–20 years. Fruit tree species may be evergreen (citrus, olive) or
deciduous (pome fruits, stone fruits, walnut).
Plantations are established by transplanting young trees grown in nurseries. In
principle trees may be planted at any time of year, provided that temperatures are
not too low. However, successful establishment requires an adequate balance
between root water uptake and water loss by transpiration to avoid desiccation
and death. The best time for planting is autumn when the tree is dormant, the air
temperature is low and the soil temperature is still warm, which enhances root
growth. Leafless trees of deciduous species may be planted as bare-root trees in
autumn or spring. Evergreen species require using trees with active root systems
and are then transplanted directly from the pots where they have grown in the
nursery.
A typical limiting factor for tree growth is soil compaction so breaking any
compacted soil layers by deep vertical tillage is a common practice before
transplanting.
Although not specifically planting operations, pruning allows renovating the tree
structure while grafting is used to change the cultivar (scion) which constitutes most
of the tree shoot while keeping the rootstock.
Bibliography
Kirby, E. J. M. (1993). Effect of sowing depth on seedling emergence, growth and development in
barley and wheat. Field Crops Research, 35, 101–111.
Chapter 17
Tillage
José A. Gómez, Francisco Orgaz, Helena Gómez-Macpherson,

Abstract Tillage has been developed in farming to improve soil conditions in

relation to the water balance and crop growth, to incorporate crop residues, and for
weed control and preparation of the seedbed. The effects of tillage depend greatly
on the water content and the characteristics of the soil. Clay soils will not be usually
found under the more suitable conditions for tilling, which are easier to find on
medium or coarse textured soils. The main undesirable effects of tillage are soil
compaction, which leads to a reduction in crop yields, and soil degradation,
particularly due to water erosion (Chap. 18). Erosion of the surface soil layers
reduce the natural fertility and the water retention capacity of soils.
17.1 Introduction
From the point of view of farming, the soil has often been viewed as a mere medium
on which the crop grows. Thus, the soil structure should be suitable for the
germination of the seeds and the growth of the roots and must have characteristics
that enhance the storage and supply of water, nutrients, gases and heat to the crop.
From this perspective tillage is inseparable from agriculture. The transformation of
a natural ecosystem into an agroecosystem requires necessarily a mechanical
intervention on the soil. Since hoe tillage, and later the Roman plow, followed by
the appearance of the moldboard plow, and finally the development of mechanical
traction, tillage and crop cultivation have been virtually synonymous.
Each soil-crop-climate system presents specific problems that require the use of
different tillage operations, which has led to the development of diverse machinery
J.A. Gómez (*) • F. Orgaz • H. Gómez-Macpherson

e-mail: [email protected]; [email protected]; [email protected]

Agriculture, DOI 10.1007/978-3-319-46116-8_17
230 J.A. Gómez et al.
whose engineering is well known. Unfortunately, much less is known about the
effects of tillage on the physical, chemical and biological properties of the soil, and
ultimately on the effects on crop yield. This limited knowledge is often translated
into tillage practices whose only rationale is the habit or tradition from the times
when tillage was performed using animal power.
In western agriculture traditional or conventional tillage, which is characterized
by a large number of operations, using diverse implements and powerful tractors, is
increasingly challenged by the high cost of energy expenditure and by soil degra-
dation resulting in different environmental problems in numerous agricultural
areas. The rationalization of tillage requires considering the soil as a valuable
resource and should be based on a better understanding of the effects of tillage on
soil properties and on crop production, which is the objective of this chapter.
17.2 Objectives of Tillage
Traditionally tillage had three main objectives, which were seedbed preparation,
improving soil conditions for crop growth and controlling weeds. But the goals
have changed with the appearance of new technologies (herbicides, seed drills),
new issues or problems (compaction, erosion, offsite contamination) and a better
understanding of the relevance of some soil functions (such as carbon sink or
natural filter of water). In rain fed agriculture tillage is also an essential tool to
modify the water balance so as to improve the availability of water for the crop.
17.2.1 Seedbed Preparation
This is a process that often requires the removal of the residues of previous crops.
Removal can be done by burning the residues or burying them with certain
operations. The burning of stubble is a fast and cheap method that has been widely
used in the past and has some clear advantages, such as the elimination of weed
seeds, the destruction of propagules of pathogens and insect eggs and larvae, and
the immediate release of some nutrients. But burning causes a loss of organic matter
and N (lost as volatile N oxides), contributes to air pollution and to anthropogenic
CO2 emissions, and overall soil degradation and increases the fire risks.
After clearing the residues, we may proceed to preparing the seedbed, which
ideally consists of a surface layer of granular structure with a high percentage of
aggregates smaller than the seed. In general this objective is achieved only when
tillage is performed with a soil water content close to field capacity and is called
optimum soil water content for tillage (OPT, e.g. Dexter and Bird 2001) (see 17.3).
In some cases it is necessary to compact slightly the seedbed with a roller
17 Tillage 231
compactor to enhance seed-soil contact and hydration of the seeds. Furthermore,

below the depth of planting, the soil must have a lower bulk density to allow root
growth without restrictions, which is often achieved by previous deeper tillage.
17.2.2 Weed Control
Before the advent of herbicides, tillage was the only effective method for control-
ling weeds. The control may be direct by destroying the plants, cutting the roots or
the stem or burying them. The control may be indirect, by changing the position of
weed propagules or changing the environmental conditions of the weed seed bank.
For example, the moldboard plow buries many seeds below a certain depth making
them unable to emerge. In other cases, such as weeds that propagate through
underground organs (tubers, rhizomes), tillage contributes to cutting such organs
and therefore enhances the dispersion of the weed when the soil is wet but it may
have the opposite effect when the soil is dry as the weed propagules would
desiccate.
In this tillage strategy for controlling weeds it is essential to till immediately
before planting to minimize weed-crop competition. This operation can reduce the
water content of the seedbed and therefore cause poor seedling emergence in dry
areas. This occurs, for example, in spring sown rain fed sunflower. This negative
effect can be solved by replacing the tillage operation by a pre-plant herbicide
application.
In those crops for which there are no selective herbicides or when they are not
very effective against the weed community, inter-row tillage operations will be
required after crop emergence. Although it is possible to control weeds only with
tillage, herbicide use, at least partially, is often a much more effective alternative in
terms of cost and time.
17.2.3 Modification of Water Balance
In rain fed crops the main objective of tillage is to improve the water balance to
maximize the availability of water for the crop. In a natural ecosystem, with the soil
covered by vegetation, the macropores formed by the roots and mesofauna allow
high and stable infiltration rates even at relatively high bulk density. The situation is
very different in an agroecosystem as seedbed preparation involves traffic of
machinery and the surface of the soil is kept bare. This kind of tillage results in a
low bulk density but in a pore system in which many of the pores are less stable than
the macropores formed by vegetation or soil fauna or are even occluded with no
connection to the soil surface (Fig. 17.1). In this situation the impact of raindrops
breaks soil aggregates and causes the sealing of pores, thus reducing the infiltration
rate. This is reversed after tilling, by breaking the surface crust and increasing the
Fig. 17.1 Conceptual Intensive tillage No-till

model of soil porosity in
tilled and long-term no till
systems (Adapted from
Ontario Ministry of
Agriculture and Food)
Plow pan
soil porosity and surface roughness, but the effect is temporary and the infiltration
rate is reduced after new rain events occur. The velocity of the surface sealing
process depends not only on the amount of precipitation but also on soil character-
istics, especially its structure (closely related to soil texture and organic matter
content), being at its best state in undisturbed natural soils under grassland or forest
plant cover.
Tillage thus serve to break the surface crust, but also increases porosity, which
improves water retention capacity and aeration, which in turn favor root growth.
In many rain fed areas farmers consider tillage an effective method for reducing
evaporation from the soil surface (Es). Indeed, in soils prone to cracking, sealing of
the surface cracks by tillage helps reducing Es. Large cracks increase evaporation in
relatively deep sections of the subsurface soil in desert areas providing that trans-
port of the pore water from the sediment matrix to the crack walls is not a limiting
factor. This would explain the adoption of inter row cultivator passes in summer
crops (e.g. cotton) in expansive soils. In many cases, however, tillage may increase
soil evaporation. Evaporation from a dry soil surface is very small (second stage
evaporation). If the soil is tilled the dry surface layer mixes with moist soil from
below thus increasing evaporation. In Fig. 10.3 (Chap. 10) an increase in evapora-
tion on day 157 was due to a pass of cultivator for weed control. In the long run, the
impact of tillage on Es is of very limited magnitude as most the Es takes place
before the tractor with its tillage implements can enter the field. Thus, with the
exception of expansive soils, the effect of any tillage system on soil evaporation has
much less impact on the water balance than on infiltration.
17.2.4 Other Goals
Tillage can serve other purposes such as the modification of the energy balance, or
incorporation of fertilizers or soil amendments. For instance, the reduction of bulk
density and the soil water content after tilling, increases thermal diffusivity which
favors a faster warming of the soil surface.
17 Tillage 233
Sometimes we find some tillage practices whose objectives are unclear and may
be due more to tradition or aesthetic reasons. Traditionally good farmers kept their
fields “clean” of weeds at all times and this has led to excessive tillage and the
adoption of unsustainable practices.
17.3 Influence of Soil Water Content on the Effects

of Tillage
The energy required for tilling the soil and the effects of tillage depend on the soil
water content. In medium or fine textured soils the cohesive strength of soil
aggregates decrease with increasing water content. The adhesion forces between
the soil and the tools increase with water content up to a maximum, in which the soil
passes from the plastic, i.e. moldable, to the liquid state. In the liquid state, tillage
causes the dispersion of the soil particles, and the soil loses its structure. The
coherent state of the soil occurs with low water content and does not allow
deformations without breaking of the aggregates. In this state tillage generates
large blocks of aggregates (lumps) with large gaps between them. Between the
coherent and the plastic state there is a point in which the sum of both of adhesion
and cohesion forces is minimal, which occurs with a medium content of water
below the upper limit. At this state, which is called optimum soil water content for
tillage (OPT, e.g. Dexter and Bird 2001) the soil crumbles after tillage (Fig. 17.2).
The water content of the soil not only determines the effects of tillage on soil
conditions but also the degree of soil compaction due to traffic of machinery. Soil
compaction occurs mainly in the plastic state but it is not likely to be important in
drier soil, since in this case the force causes breakage of aggregates. Within the
plastic state two zones may be distinguished above and below the Adhesion Point:
above that point the soil will adhere to a smooth surface cutting it, as is the case with
RESISTANCE OF TOOLS
FORCE
ADHERENCE
COHESION
COHERENT FRIABLE PLASTIC LIQUID

WATER CONTENT
RISK OF COMPACTION: LOW MEDIUM HIGH LOW
Fig. 17.2 Coherence and adhesion forces between soil and tillage tools as a function of soil water
content
implements. This implies a high energy expenditure for tilling and the danger of
cementation once the soil dries.
The best condition for tilling occurs in a range of water contents below field
capacity. This occurs approximately 2–3 days after rain or irrigation for medium
texture soils. In this zone the energy required for tillage is minimal and a granular
structure is achieved, which is desirable for the seedbed while the risk of compac-
tion is moderate. Medium and sandy textured soils drain well, and have a narrow
range of plastic state, so they move quickly to the plastic state and reach the OPT
quickly after a rain or irrigation, staying that way for a long time. That state,
however, is not easy to achieve in clay soils where drainage is slow and the plastic
state interval is wide. If we till in that state the soil is cut into slices, which get
extremely hard when dry and the tillage operation barely increases soil porosity.
Besides, the risk of compaction is maximum when a soil is tilled in the plastic state.
To promote drainage in clay soils it is necessary to till before the rainy season. Only
so it will be possible to get a proper moisture condition before planting. The
drawback is that tilling in the dry season, when the soil is in coherent state, requires
much energy and generates large clods which then must be disaggregated by
additional secondary tillage operations. That subsequent disaggregation can be
very difficult when rainfall is low. The disadvantages of tilling when the soil is
dry do not occur in sandy soils as they do not show a coherent state.
17.4 Conventional Tillage
Maintenance of the infiltration rate, weed control and seedbed preparation require
performing several tillage operations that vary widely across geographic areas, soil
types and crops. This set of operations, which we call conventional tillage, can be
classified according to different criteria. In conventional agriculture it is common to
distinguish between primary and secondary operations. The primary operations are
performed with a moldboard plow or a disc plow sometime after harvest and serve
to incorporate crop residues and to improve soil conditions. The moldboard plow
cuts, lifts and turns the soil down to 40 cm deep at most. This process improves
infiltration, incorporates crop residues, and buries the weed seed bank. For the
rupture of compacted deep layers subsoilers are used which can achieve greater
depths (60–70 cm), performing a vertical cut so that residues are not incorporated.
Subsoilers have replaced the moldboard plows in many areas, leaving 50–80 % of
the residues on the soil. The chisel plow performs a similar but shallower (less than
30 cm) vertical tillage than the subsoilers.
Secondary tasks are performed with harrows, cultivators and other implements,
affecting only the surface 10–20 cm. They serve to refine the soil before sowing
(reducing the size of the aggregates on the surface) and to control weeds. The
primary operations often result in large aggregates which are then broken down by
harrowing. The finer structure is achieved with cultivators, which are also used for
weed control before and after sowing (passes between rows). To finish the
17 Tillage 235
shredding of the aggregates and/or for compacting the soil surface layers, various
tools can be used (e.g. compactor, harrow tines).
Example 17.1 Sunflower conventional tillage in a wheat-sunflower rotation

in a Mediterranean area. After burning the wheat stubble in summer a
moldboard pass results in large aggregates and many gaps. After the first
rains we may use a cultivator or a disk harrow to reduce the size of the
aggregates. Then two additional passes of cultivator are needed in autumn-
winter to remove weeds, and another before planting. Additionally one can
have one to two passes of cultivator between rows to control weeds during the
campaign. Tillage costs in this case may account for over 60 % of the total
production costs of the sunflower crop.
17.5 Compaction and Plow Pan
The existence of compacted layers may be due to natural causes (e.g. petrocalcic
horizons) but it is a widespread phenomenon due to tillage. Compaction can occur
in the uppermost layer (of a few mms width) of the soil due to the impact of rain.
This surface crust hinders seedling emergence, especially if the soil is dry, and
reduces infiltration (Fig. 17.3). Secondary tillage favors the formation of surface
crusts when it leaves very fine aggregates on the surface.
Another kind of compacted horizons are those within the soil profile, which not
only delay or prevent the growth of the root system, but also lead to reduced growth
of the aerial part of the plant and finally, to yield losses (Fig. 17.4), even when the
supply of water and nutrients is not limited. Figure 17.5 shows the relationship
between penetration resistance and soil water content for a loamy soil with or
without compaction. Taking into account that with a penetration resistance of
3.2 MPa root growth is considerably reduced, one can deduce that the growth
conditions for the root in the compacted soil were greatly restricted. Compaction
has other side effects such as the development of waterlogging conditions, which
promotes denitrification (see Chap. 24), root anoxia and a higher incidence of soil
diseases (e.g. Phytophthora).
IMPACT CRUSTING
Fig. 17.3 Schematic representation of the formation of a depositional crust on a soil with sand, silt
and clay-sized particles (Adapted from Cattle et al. 2002 In: SuperSoil 2004: 3rd Australian-New
Zealand Soils Conference, Dec 2004, Univ of Sydney, Australia (published on CDROM))
Fig. 17.4 Time course of 100

LAI and biomass of cotton
CROP HEIGHT
on non-compacted (solid
line) and compacted (dotted
line) soil (Adapted from
Coelho et al. 2000)
Non compacted
Compacted
0
20 60 100 140 180
100
LEAF AREA INDEX
0
20 60 100 140 180
100
AERIAL BIOMASS
0
20 60 100 140 180
DAYS AFTER EMERGENCE
Fig. 17.5 Relationship 6

PENETRATION RESISTENCE (MPa)
between soil bulk density,

soil water content and soil 5
resistance to penetrometer.
Sandy loam soil (Adapted
4
from Coelho et al. 2000)
3
2
High density
1
Low density
0
0.00 0.10 0.20 0.30 0.40
SOIL WATER CONTENT (m3 m–3)
17 Tillage 237
Compaction is caused by the weight of the implements (plow, disk) at the depth
on which they act resulting in a plow pan, and/or by the wheels of the tractor or
other machines that compact the entire surface horizon. In either case the magnitude
of the compaction is dependent on the pressure applied (regulated by axel load and
tire type and pressure) and water content of the soil at the time of the operation, thus
traffic should be avoided if the soil is in the plastic state. This is why compaction
risk is very high in clay soils. Also in this case the plastic state which promotes
compaction is transmitted in depth. By contrast in medium textured soils the risk of
compaction is usually lower and generates a compacted layer below the tilled depth
(Fig. 17.1). This is why subsoiling to relieve compaction problems is more effective
in medium textured soils than in clay soils.
17.6 Energy Requirements of Tillage
Primitive tillage was based only on human power with hand tools. It first evolved to
the ard with draft animals and then to the plow until now, when tillage is carried out
with powerful tractors that require external energy sources (fossil fuel).
Apart from the risk of compaction, tillage has been questioned because it
increases soil erosion risk and for its large energy expenditure. These negative
effects have fostered the adoption of reduced tillage systems (see Chap. 18). In Sect.
7.4 we discussed the classification of energy inputs in agricultural operations. In the
case of tillage only two components are relevant to calculate energy requirements
per unit land area:
Ereq ¼ Edir þ Eind ¼ Efuel þ Eind ðtractorÞ þ Eind ðimplementÞ ð17:1Þ
The direct component is the energy in fuel consumed and the indirect component
is that corresponding to manufacturing, maintenance and repair of the machinery
(tractor and implement or machine). The calculation of indirect energy require-
ments (MJ ha1) may be performed with the following equation:
Mtractor EMtractor Mmachine EMmachine

Eind ðtractorÞ þ Eind ðimplementÞ ¼ þ ð17:2Þ
Ltractor MFC Lmachine MFC
where M is mass (kg), EM is the ratio of energy required and mass (MJ kg1), L is
the useful life (hour) and MFC is the machine field capacity (ha hour1). Using
average values of these parameters we have calculated the indirect energy require-
ments of different operations shown in Table 17.1. Typical values of direct require-
ments are also presented for comparison. These data should be taken as an example.
Actual direct requirements will be higher when tillage is performed under conditions
Table 17.1 Energy requirements of different agricultural operations calculated with Eqs. 17.2
and 17.3
Indirect Direct Total Labor Total
Tractor Machine Total Inc. labor
Operation MJ ha1
Plow 80 133 213 1000 1213 80 1293
Sprayer 13 5 18 68 86 13 98
Spreader 11 2 13 73 86 11 97
Sow rows 28 103 131 200 331 28 360
Roller 29 48 77 200 277 29 306
Disc harrow 69 20 90 264 354 69 423
Cultivator 20 27 47 220 267 20 287
No till drill 42 135 177 200 377 42 419
Combine 0 646 646 500 1146 51 1197
The value of EL, the daily energy required per human has been taken as 1000 MJ day1 which
corresponds to farmers of developed countries. WD the work duration has been assumed 10 h
day1
departing from optimal (OPT). Actual indirect requirements may differ if the
parameters M, EM, L or MFC change. The consumption of energy increases with
the use of high power tractor for small operations and with deeper tillage, and depends
on the shape of teeth or disks of implements.
Many studies on energy requirements of agricultural practices ignore the energy
associated to human labor. Table 17.1 also shows the energy required for labor
calculated as:
EL
Elabor ¼ ð17:3Þ
WD MFC
where EL is the daily energy required per human (MJ day1) and WD is the work
duration (hours day1). The value of EL depends strongly on the standards of living
of the farmer (Chap. 7). Even using a very high value (EL ¼ 1000 MJ day1) the
fraction of energy due to labor is small (5–15 %).
The relative importance of tillage in the energy requirements of farming is
limited. In Example 7.2 (Chap. 7) we saw a case of rain fed wheat farm producing
2500 kg d.m. grain ha1 with total energy requirement of 14,779 MJ ha1 (1500 for
tillage, 1900 for sowing, harvest and other operations, 9613 for fertilizer, 1050 for
seed and 716 for pesticides). In this case tillage is limited to one plowing and one
pass of cultivator and the fraction of energy used in tillage is only 10 % of the total
thanks to the use of herbicide. Even with more intensive tillage, its share of energy
is small as compared to the energy inputs in fertilizers. Therefore the adoption of
reduced or no tillage should be primarily promoted for improving soil conditions
rather than for saving energy.
17 Tillage 239
Bibliography
Aluko, O. B., & Koolen, A. J. (2000). The essential mechanics of capillary crumbling of structured
agricultural soils. Soil and Tillage Research, 55, 117–126.
ASAE Standards. (2000). D497, agricultural machinery management data. St. Joseph: ASABE.
Botta, G. F., Tolon-Becerra, A., Tourn, M., Lastra-Bravo, X., & Rivero, D. (2012). Agricultural
traffic: Motion resistance and soil compaction in relation to tractor design and different soil
conditions. Soil and Tillage Research, 120, 92–98.
Coelho, M. B., Mateos, L., & Villalobos, F. J. (2000). Influence of a compacted loam subsoil layer
on growth and yield of irrigated cotton in Southern Spain. Soil and Tillage Research, 57,
129–142.
Dexter, A. R., & Bird, N. R. A. (2001). Methods for predicting the optimum and the range of soil
water contents for tillage based on the water retention curve. Soil and Tillage Research, 57,
203–212.
Chapter 18
Soil Conservation
Helena Gómez-Macpherson, José A. Gómez, Francisco Orgaz,

Abstract Tillage serves to improve soil conditions in relation to the water balance
and crop growth, to incorporate crop residues, to control weeds and to prepare the
seedbed. However, tillage significantly increases the risk of soil erosion. These
problems have led to the development of conservation tillage techniques which
typically rely on the maintenance of plant residues on the ground and in a substantial
reduction in tillage operations. When conservation tillage is combined with the use of
crop rotations is termed conservation agriculture. Conservation tillage requires the
use of herbicides and specific direct drills for crop sowing. The transition from
conventional to conservation tillage should be gradual as additional problems may
arise (e.g. compaction) in some soil types. Sporadic tillage or controlled traffic could
then be adopted. In tree orchards many options for soil management are available,
from conventional tillage to the no-tillage with bare soil and herbicide applications. A
particular case is that of rain fed tree orchards in Mediterranean areas, whereas the
problems of no-till (gully erosion, reduced infiltration) can be partially alleviated by
temporal cover crops that protect the soil during the rainy season and are killed in
early spring to avoid competition for water with trees.
18.1 Introduction
Soil erosion is the main threat to the sustainability of agricultural systems in many
parts of the world. The development of powerful tractors in the last century allowed
rapid mechanization of tillage operations but also resulted in a reduction of ground
cover by vegetation and stubble, a decrease in soil organic matter, and a
H. Gómez-Macpherson (*) • J.A. Gómez • F. Orgaz

e-mail: [email protected]; [email protected]; [email protected]

Agriculture, DOI 10.1007/978-3-319-46116-8_18
242 H. Gómez-Macpherson et al.
deterioration of soil structure, and therefore, in an increased risk of water and wind
soil erosion. The “Dust Ball” period in USA in the 1930s exemplifies the relevance
of this problem.
Conserving the soil in agricultural lands is not a straightforward task. Success
requires adapting various soil management practices to local conditions as well as
considering costs and profitability. The recommended approach is to develop a set
of practices based on adapting and adopting a package of agronomic technologies
known as Conservation Tillage guidelines that allow for minimum disturbance of
the soil, maintenance of soil cover with vegetation and/or residues, and spatiotem-
poral diversification of cropping systems.
18.2 Soil Erosion
Erosion is the process of soil loss. Firstly it requires energy for removing the
particles, and then some transport medium. The energy is obtained from the impact
of raindrops and the transfer of momentum from water (surface runoff) or wind.
The transport medium will be the fluid (water or air). The water erosion is
proportional to runoff which depends on the relationship between precipitation
and infiltration (Chap. 8) and a parameter (Erodibility) that reflects the ease with
which the soil is eroded. The erodibility depends mostly on the structural stability of
the soil which is related to the organic matter content, as well as soil texture. In bare
soils with fine aggregates erodibility is highest.
Erosion has two major effects on the agricultural system: soil loss results in a
decrease of soil depth which in turn involves a reduction of the water storage
capacity, and therefore a reduction in long-term yield. Moreover the surface soil
lost is often richest in nutrients, so that erosion involves a loss of fertility (and yield
potential), and causes an environmental problem (sediment and pollutants accumu-
late in surface waters). Unfortunately it is not easy to control erosion due to its
ephemeral nature, and often, most of the erosion occurs in a few episodes of
torrential rain. In any case, excessive tillage destroys the soil structure and main-
tains the soil surface exposed to the wind and rain for prolonged periods, making it
the leading cause of erosion in many agricultural systems. Although it initially
increases water infiltration, its effect is temporary and, in some lime-fine texture
soils, it can favor the rapid formation of surface crust after the subsequent rainfall.
On a larger scale, erosion is a process with positive feedback: soil loss implies a
reduction of vegetation which favors intensified erosion, which ultimately leads to
desertification (the process by which a dryland region becomes increasingly arid,
typically losing its bodies of water as well as vegetation) that is advancing in some
areas of the planet, being soil erosion one of its major drivers.
Several methods have been proposed to quantify soil erosion among which the
most popular is USLE (Universal Soil Loss Equation) and its revised version,
18 Soil Conservation 243
RUSLE (Renard et al. 1997). Both were designed to predict the effect of different
soil management on average annual erosion rates at hillslope scale. According to
the RUSLE, the average loss of soil by erosion (SLE, t/ha/year) is calculated as the
product of six factors:
SLE ¼ R1 K1 L1 S1 C1 P1 ð18:1Þ
where R1 is the average annual rainfall erosivity, which is a measurement of the

rainfall energy available for water erosion. It depends on amount and intensity of
the rainfall and its calculation is explained in detail by Renard et al. (1997). There
are several equations that can be used as a regional approximation, such as Eq. 18.2
for the USA, based on the monthly rainfall distribution ():
X
R1 ¼ 4:17 Pi 2 =P 152 ð18:2Þ
where Pi is the precipitation (mm) of month i, and P is the annual precipitation.

K1 is the soil erodibility factor, and is equivalent to soil loss (t/ha/year) that
would occur in standard conditions, that is, if the land is kept as clean fallow, and
has a slope of 9 % and a length of 22 m. It depends on soil texture and soil organic
matter. The values can be approximated from values in Table 18.1.
L1 and S1 are two factors that include the effect of the slope length (L) and
steepness (S). In its most basic form (USLE) L1 and S1 are combined to represent
the ratio of erosion in the specific situation and that happening in standard condi-
tions (22 m long 9 % steepness), and their product is calculated as:

L1 S1 ¼ 0:065 þ 0:0456 pt þ 0:006541 pt 2 ðlt =22:1ÞNT ð18:3Þ
where pt is the slope (%) of the land, lt is the length (m) and NT is a factor that
depends on the slope steepness (see Table 18.1).
C1 is the factor that reflects the effect of cover and management and their
interaction with the rainfall erosivity distribution during the year. It can be approx-
imated from tables developed for local conditions as presented in Table 18.1.
Finally, the factor P1 indicates the effect of several agronomic measures for
erosion control (Table 18.1) such as contour plow.
The value calculated using Eq. 18.1 is compared with the tolerable soil loss rate
(Table 18.2), i.e. the maximum value before the long-term natural soil productivity
is severely affected. The tolerance level varies depending on the type and rooting
depth of soil. Generally, deep soils not previously eroded are assumed to have a
higher tolerable soil loss rate than shallow and/or previously eroded soils.
Table 18.1 Parameters for calculating soil loss using the USLE
K1
Soil texture Low OM Medium OM High OM
Clay 0.2 0.17 0.13
Fine sand 0.16 0.14 0.1
Fine sandy loam 0.35 0.3 0.24
Loam 0.38 0.34 0.29
Loamy fine sand 0.24 0.2 0.16
Loamy sand 0.12 0.1 0.16
Loamy very fine sand 0.44 0.38 0.3
Sand 0.05 0.03 0.02
Sandy clay loam 0.27 0.25 0.21
Sandy loam 0.27 0.24 0.19
Silt loam 0.48 0.42 0.33
Silty clay 0.25 0.23 0.19
Silty clay loam 0.37 0.32 0.26
Very fine sand 0.42 0.36 0.28
Very fine sandy loam 0.47 0.41 0.35
Tillage and cropping practice Crop sequence C1
Forest Permanent 0.0005
Pasture Permanent 0.005
Rotation 1/6 C-G-M-M-M-M 0.011
Rotation 2/5 C-S-G-M-M 0.027
No till cover crop after soybean C-S 0.0027
Chisel, 50 % residue on contour C-S 0.16
Chisel, little residue C-S 0.35
Moldboard plow, spring C-S 0.35
Moldboard plow, fall C-S 0.39
Bare soil None 1
Slope % NT
<1 0.2
1–3 0.3
3–5 0.4
>5 0.5
Direction of tillage P1
Same as slope 1
Contour lines 0.5
Sources: For K1 and NT: Stewart et al. (1975) US EPA Report No. 600/2-75-026 or USDA Rep
No. ARS-H-5-1
For C1 and P1: Franzmeier et al. (2009) Indiana Soils. Evaluation and Conservation. Purdue
University
C corn, M meadow (forage crop), G small grains, S soybean
Table 18.2 Soil loss Soil erosion class Potential soil loss (t/ha/year)
tolerance rates for comparison
Very low (tolerable) <6.7
with values derived from
USLE or RUSLE Low 6.7–11.2
Moderate 11.2–22.4
High 22.4–33.6
Severe >33.6
Example 18.1 Let us calculate the average erosion on a farm in Flint, Mich-
igan with 5 % average slope and slope length 50 m. The soil is loam with
average OM (2 %) and the crop is a rotation of corn and soybean with
moldboard plow in the fall, which is made on contour.
Monthly rainfall values are:
Month 1 2 3 4 5 6 7 8 9 10 11 12 12
Precipitation mm 35 32 55 75 67 81 69 89 90 55 66 53 53
Therefore, rainfall erosivity is:

X
R1 ¼ 4:17 Pi 2 =P 152 ¼ 136
Now, in Table 18.1 we see that for a loam soil with medium OM content
K1 ¼ 0.34 and with slope 5 % NT ¼ 0.4. As pt ¼ 5 and lt ¼ 50 we can calculate
the product of L1 and S1 as:

L1 S 1 ¼ 0:065 þ 0:0456 pt þ 0:006541 pt 2 ðlt =22:1ÞNT

¼ 0:065 þ 0:0456 5 þ 0:006541 52 ð50=22:1Þ0:4
¼ 0:63
According to Table 18.1 with contour tillage we have P1 ¼ 0.5 and the maize-
soybean rotation with fall plow has C1 ¼ 0.39. Therefore the estimated soil
loss due to erosion is:
SLE ¼ R1 K1 L1 S1 C1 P1 ¼ 136 0:34 0:63 0:39 0:5

¼ 5:7 t=ha=year
According to Table 18.2 this value is considered to be very low soil erosion.
Note that for this specific field if tillage direction was that of the slope, the
estimated soil loss would double as P1 would be 1 instead of 0.5. Then the
estimated soil loss (11.4 t/ha/year) would be classified as moderate.
18.3 Conservation Tillage
As discussed above, conventional tillage has a number of disadvantages:

(a) The increased infiltration due to tillage is only temporary and for some soil
types it enhances the formation of a superficial crust.
(b) Soil compaction is increased by the formation of a plow layer and by the
frequent traffic of machinery.
(c) Tillage prevents the accumulation of organic matter in the soil surface which is
necessary to protect the soil and improve and stabilize its aggregate structure.
Organic matter generates aggregating agents (especially polysaccharides) that
promote cohesion of the aggregates.
(d) Tillage has high economic and energetic costs and promotes the emission of
greenhouse gases from soils. The use of fossil fuels also contributes to anthro-
pogenic CO2 emissions.
(e) In general tillage favors erosion, although some operations may reduce it
temporarily.
All of these problems and the emergence of new technologies have allowed for
alternative systems for managing agricultural soils. These systems range from
reducing the number of operations (reduced tillage) to the complete elimination
of tillage (no tillage) but not all of them contribute to soil conservation. In order to
avoid confusions with the terminology, Conservation Tillage has been defined by
ASAE as tillage operations (or no tillage) that leave enough residues to cover 30 %
of the surface after sowing, and at least 110 g/m2 of organic material during the
critical periods of erosion risk. Conservation Agriculture (CA) adds a third com-
ponent to minimum soil disturbance and maintenance of residues: the use of more
than one crop in the rotation (FAO 2016).
Compared to conventional tillage, adopting no-tillage may have more negative
than positive effects on crop productivity (Fig. 18.1). A major concern is the
impossibility to decompact the plough layer, particularly when heavy direct-drill
seeders enter in a field with wet soil. Compacted soil will reduce root and plant
growth, whereas compacted superficial soil will result in lower water infiltration
and soil water content; waterlogging and even seedlings death may then occur.
Another major concern when adopting no-tillage is the potential increase of weeds,
diseases and pests. Adopting no-tillage requires attentive weed control and
increased herbicide use. On the other hand, the major advantage of no tillage is
the possibility to enter earlier in the field for sowing the crop. This is particularly
relevant to adjust the crop to a narrow cropping season window and for crop
sequence intensification.
Most negative effects may be counteracted by maintaining crop residues on the
soil surface after harvest (Fig. 18.1). Crop residues protect the soil from the direct
impact of rain and wind and improve soil structural stability. The presence of
decomposing roots from the previous crop also favors the structural stability of
the soil surface layers and thus reduces erosion. Another positive effect of the
Fig. 18.1 Main pathways through which a change in management from conventional to conser-
vation agriculture (zero-tillage with mulching and crop rotation) may impact key drivers
(highlighted in green boxes) of crop yields. From the cropping system performance perspective,
a single dark green arrow and a double red arrow indicate positive (beneficial) and negative
(constraining) effects, respectively, of a conservation tillage management on yield drivers and
component attributes. A dotted line indicates a beneficial effect expected to only accrue over the
long term (Adapted from Brouder and Gómez-Macpherson (2014))
presence of residues is the increased surface roughness, which implies an increase

in the water held in the surface that is retained until it infiltrates. Leaving crop
residues has other positive effects in the longer term (2–5 years) such as an increase
in organic matter content, and in the activity of the soil mesofauna (generating
macropores) and the soil flora which are enhanced as surface temperature fluctua-
tions are reduced. All these processes contribute to the improvement of the infil-
tration capacity. Furthermore the favorable microclimate near the soil surface
promotes root proliferation which competes favorably with direct evaporation
from the soil surface.
The effect of the presence of residues on evaporation from the soil surface
depends on the frequency of rainfall. If they are frequent, the soil is kept in
evaporation stage 1, so the residues will reduce evaporation in proportion to the
radiation they intercept. However, if rainfall is infrequent, the soil remains in phase
2 (limited by hydraulic conductivity) so the presence of residues has little effect on
evaporation. In short, if the amount of residues is sufficient to cover the ground, the
water balance is expected to improve due to enhanced rainfall infiltration and, to a
lesser extent, to less evaporation. In fact, CA has had more positive results by
increasing yields (relative to conventional tillage systems) in rainfed systems in the
semiarid zones than in other environments where water is not limiting.
Maintaining residues may also have negative short term effects on plant growth
(Fig. 18.1). In the initial phase of adopting no-tillage and mulching, high amounts
of residues may result in N immobilization. Greater N fertilizer amounts will then
be required to compensate for the immobilization until soil fertility is increased and
the system is balanced. Additionally, non-mobile soil nutrients, like phosphorus,
cannot be incorporated into the soil unless the fertilizer is placed next to seeds
during sowing. Residues also reduce radiation absorption by the soil which delays
soil warming during early establishment of spring crops when temperatures are low;
on the other hand, in the tropics, lower temperatures may benefit nutrient cycling
and plant growth. Leaving residues on the ground also require specific drills to sow
through them, and makes difficult flood or furrow irrigation or herbicide
application.
In conservation agriculture, crop rotation has the role to facilitate weed control
and to reduce the risk of pests and diseases incidence (Fig. 18.1), particularly in the
soil. For example, higher incidence of diseases caused by soil fungi in no-tilled
wheat monoculture systems were observed in Australia but were controlled when
rapeseed was included in the rotation following wheat. The rotation would also help
to maintain the optimum amount of residues in the system by combining high and
low-residue producing crops. The value of adopting a rotation under conservation
agriculture may be evident in the long term only.
There are differential responses to tillage systems in the different agricultural
systems. In a meta-analysis it was found that no-tillage reduces crop yield com-
pared to conventionally tilled systems but that these negative effects decrease if
residues are maintained. While CA has been widely adopted in North and South
America, so far it has been little adopted in Europe (except cover crops in orchards).
Reasons limiting CA adoption in Central and Northern Europe include technical
problems with crop establishment in cold and wet soils, high natural organic matter
content of many soils, flat topography and low erosion risk, and management
problems with crop residues and weed control.
Example 18.2 Intermediate system of conservation tillage (reduced tillage,

also called minimum tillage) for wheat-sunflower rotation includes vertical
tillage often with chisel and causes severe erosion also (Fig. 18.2):
– Sunflower is sown directly on wheat stubble using a special drill for direct
planting. Weeds are controlled with two applications of herbicide (glyph-
osate) and a residual herbicide applied before planting.
– The residues of sunflower are cut with disc harrow (2 passes crossed) or
burned. Then herbicide is applied or cultivator is passed before planting
wheat.
In this system it would be difficult to completely avoid tillage, unless a
no-till drill for cereals and sunflower is used, but in any case it improves soil
(continued)
Fig. 18.2 Severe rill erosion after chisel tillage (Photo by Juan Jose Perez)
protection and reduces costs compared to traditional tillage. However it may

require some deep tillage to break deep compacted layers, which can be done
by subsoiling every 4–8 years. At harvest, it will be convenient to cut the
cereal as high as possible and have a residue chopping and distributing system
located in the harvester, to avoid problems at planting. These systems in
which the residues are distributed simultaneously may favor the dispersal of
weeds in the field, as weed seeds are spread along with the crop residues
(Chap. 30).
In humid environments some forage crops with very small seeds such as alfalfa
may be sown without burying the seed, which is distributed at random on previous
crop residues or over the previous crop and is thus protected. If environmental
conditions are suitable (frequent rain, low evaporative demand) an acceptable plant
stand is achieved.
The transition from conventional tillage to conservation tillage should be a
gradual process and adjustments will be needed to cope with the specific problems
of the soil and crops and techniques to be adopted. For instance, no-till has fewer
problems adapting to medium or light textured soils because of their lower risk of
compaction. In heavy clay soils no-till may aggravate soil compaction and cause
yield reductions. Furthermore the performance of no-till seeders is usually worse in
heavy soils, which when cut with the hoe of the drill remain slightly open, so that
Fig. 18.3 Ridge tillage:

direct sowing of maize over
cotton stalks
the seed-soil contact is not good. These problems can be solved in part by changes
in the implements of the seeder. For example you can add several blades which cut
and remove the soil in a section of 10 10 cm in front of the sowing boot, which
favors the seed-soil contact. This localized tillage can be applied ahead of planting,
using a cultivator with blades spaced as boots on the planter. This procedure has
been called strip tillage, and allows the localized application of P and K fertilizers
and promotes soil warming of plant rows.
In ridge tillage or permanent bed planting (Fig. 18.3), the soil is left undisturbed
from harvest to planting except for nutrient injection. Ridges are rebuilt annually.
Planting is completed in a seedbed prepared on ridges with sweeps, disk openers,
coulters, or row cleaners. Residues are left on the surface between ridges. Weed
control is accomplished with herbicides and/or light cultivation. The beds or ridges,
on which the rows of plants are sown, have the advantage of drying sooner and
warming faster in spring.
18.4 Soil Conservation Systems in Orchards and Vineyards
In permanent woody crops the objectives of soil management are different than
those in arable crops because sowing is not required. Traditionally, tillage in
orchards targeted the elimination of weeds and the improvement of the water
balance. The wide availability of herbicides or mowers to control weeds allows
restricting tillage while ensuring an adequate water balance and erosion control.
The soil management options used in orchards are numerous but we can mention
the following:
– Conventional tillage maintains the soil bare by periodic surface tillage (cultiva-
tor, disc harrow) to increase infiltration and control weeds. It is the system used
traditionally and is still widespread in rain fed orchards (olive, almond,
vineyards). This system generates compaction problems, keeps soil organic
matter low and increases erosion risks.
– Minimum Tillage: The weeds are controlled by herbicides and tillage operations
are limited to a cultivator pass in summer-autumn to improve infiltration.
– No tillage with bare soil. The soil is kept bare with herbicides but the long-term
effect is a reduction in infiltration due to unavoidable traffic (application of
herbicides and fertilizers, harvesting). For rain fed systems the lower infiltration
usually involves a worsening of the water balance and a reduction in yield. In
sloping areas no tillage favors gully erosion due to higher runoff coefficients.
– Permanent cover crops: A cover crop is sown and managed by periodic mowing
when needed. It is not advisable under water-limiting conditions due to the
competition for water with the trees..
– Temporary cover crops: erosion problems in orchards have forced the search for
viable floor management alternatives in rain fed or irrigated areas where water
availability is limited. In these cases permanent herbaceous covers are not
feasible as they increase the ET and reduce the water availability to trees. In
those cases, temporary cover crops may be used. Depending on the farm
conditions the cover crop may be seeded or generated by the community of
weeds. In Mediterranean conditions the cover is established in the fall, before the
rainy season but is removed early in the spring to prevent competition for water
during spring and summer. This can be done with herbicides or by mowing, with
the residues providing mulch until the next fall. Mowing usually needs to be
repeated more than once during the season depending on the rainfall patterns.
Cover crops can occupy the whole area or just part of it (cover crops in strips).
The main problem to be solved with this method is the decision of when to
remove the cover crop, as if late it will mean a reduction of water available to the
trees, and therefore a loss in yield, but an early removal will result in less soil
protection and increased erosion risk. Although there is a need for annual
seeding of the cover crop, several options using perennial species are available.
Fig. 18.4 shows the cumulative evaporation from bare soil or from a grass cover
under an olive grove established in the autumn (15 October 2010). In this
example the cover below the olive trees does not evaporate more water than
bare soil, until mid March. By April 1, the water lost due to the cover is 9 mm.
By May 1 it is 37 mm. The appropriate date for killing the cover would be
around the time when the two curves start to diverge. Unfortunately, this date
varies with the conditions of each year for a particular orchard, i.e. in dry years
250
CUMULATIVE EVAPORATION (mm) GRASS EVAPORATION

200
BARE SOIL EVAPORATION
150
100
50
0
15-Oct 24-Nov 3-Jan 12-Feb 24-Mar 3-May 12-Jun
DAY OF YEAR
Fig. 18.4 Bare soil evaporation or grass evaporation below an olive orchard of 50 % ground cover
in Southern Spain. The grass cover is established in October 15, 2010
the competition between the cover and the trees would start much earlier. The
date also varies depending on tree spacing and size, soil type and rooting depth,
and cover crop species and extension. A combination of experiments and
modelling analyses is being pursued in different countries to reduce this
uncertainty.
18.5 Controlled Traffic
Conservation tillage may lead to soil compaction, particularly with the use of heavy
drills or harvesters in wet soils. The introduction of controlled traffic may reduce
this problem. Controlled traffic implies that traffic is restricted to the same rows all
the time so that the tractor wheels stay on the same tracks in the field for all
operations while the crop is cultivated in the zone within these tracks. This
approach has been facilitated with the availability of Global Positioning System
(GPS) guidance for field equipment. Track widths of commonly available equip-
ment dictate the width of the area without traffic. The adoption of controlled traffic
combined with conservation tillage can increase soil infiltration and reduce soil
erosion (Fig. 18.5), increase soil water content, and crop yields and farm profits.
Nevertheless, occasional deep ripping or subsoiling of the traffic lanes may be
needed.
Fig. 18.5 Time course of runoff rate and sediment load in runoff water during rainfall simulations
in conventional ridge system (CB) and in furrows of permanent ridge system (PB), with (þT) and
without (T) traffic
Example 18.3 In Southern Spain, a ridge planting system combined with

controlled traffic has been developed successfully to deal with soil compac-
tion and excessive residues produced by an irrigated maize-cotton rotation.
Irrigation was applied from a central pivot but ridges were formed to facilitate
controlled traffic and to have residues in the furrows rather than on the beds
where crops are sown. Applied irrigation was reduced by 17 % since the
introduction of the system, without yield loss, but most important to the
farmer, the costs were also reduced.
Bibliography
Bergsma, A. (1981). Indices of rain erosivity. A review. ITC Journal, 4, 460–483.

Brouder, S., & Gómez-Macpherson, H. (2014). The impact of conservation agriculture on small-
holder agricultural yields: A scoping review of the evidence. Agriculture Ecosystems and
Environment, 187, 11–32.
FAO. (2016). Basic principles of conservation agriculture. Food and Agriculture Organization of
the United Nations, Rome. http://www.fao.org/ag/ca.html
Renard, K. G., Foster, G. R., Weesies, G. A., McCool, D. K., & Yoder, D. C. (1997). Predicting
soil erosion by water: A guide to conservation planning with the revised universal soil loss
equation (RUSLE) (US Department of Agriculture, Agriculture handbook. No 703).
Washington, DC: U.S. Dept. of Agriculture, Agricultural Research Service.
Chapter 19
Irrigation Systems
Luciano Mateos
Abstract Irrigation methods are classified into surface irrigation, sprinkler irriga-
tion, and localized irrigation (drip/micro irrigation). Surface irrigation uses gravity:
the water is distributed over the field as it infiltrates. With sprinkler irrigation, water
is distributed across the field using pressurized pipes and sprinkled over the soil
through nozzles. Drip/micro irrigation systems are conceived to localize the water
to parts of the field and apply it frequently. The factors to be considered when
selecting an irrigation method are: project goals (maximize economic return,
minimize investment cost, conserving water and water quality), institutional and
social site conditions (financial, labour availability, durability and robustness),
physical site conditions (soil and topography). Irrigation performance assessment
is advisable as part of the processes of operation improvement, and to establish
system design criteria. The main irrigation performance indicators are irrigation
efficiency, application efficiency, adequacy, and distribution uniformity. They
assess different aspects of the irrigation process but have to be used jointly for a
comprehensive assessment.
19.1 Introduction
The history of irrigation parallels that of agriculture. Irrigation has been practiced
for more than 5000 years and was essential to early civilizations that developed in
arid and semiarid environments, where irrigation makes the difference between the
viability and non-viability of agriculture. Also in Mediterranean or sub-humid
environments, where rainfall is limited or non-uniformly distributed, irrigation is
responsible for an important part of the crop production. An estimated of 17 % of
global cultivated land is irrigated, and produces about 40 % of the world’s food.
L. Mateos (*)

Agriculture, DOI 10.1007/978-3-319-46116-8_19
256 L. Mateos
The twentieth century experienced a dramatic expansion of world irrigation. The

area equipped for irrigation worldwide is 308 million ha of which 83 % were
actually irrigated around year 2005. These figures represent about a sevenfold
increase from the beginning of the twentieth century. Sixty-two percent of the
area equipped for irrigation uses surface water, while 38 % uses groundwater and
only 0.1 % uses non-conventional water sources. About 70 % of the irrigation area
is in Asia and 17 % in America. The largest continuous areas of high irrigation
concentration are along the rivers Ganges and Indus; in the Hai He, Huang He and
Yangtze basins in China; along the Nile river in Egypt and Sudan; and in the
Mississippi-Missouri river basin in North America. Zones of high irrigation density
in Europe are along the Danube and Po rivers. The 3.8 Mha of land irrigated in
Spain concentrate along the main river plains, the Mediterranean coast, and over
aquifers in the central plateau.
Agriculture is the largest water-use sector worldwide, accounting for about 70 %
of water withdrawals from rivers and aquifers and 90 % of consumptive water uses.
The development of irrigated agriculture has boosted agricultural yields and con-
tributed to price stability, making it possible to feed the world’s growing popula-
tion. The future of agriculture in many countries relies on the possibility of
maintaining, improving and expanding the irrigated area. However, irrigation is
facing increasing competition from the domestic and industrial sectors as the
pressure on water resources increases, to the point that in many regions it is
becoming a threat to the environment. In order to fully understand some of the
irrigation management practices described in Chaps. 20 and 21 it is essential that
the reader reviews the main features of the different approaches, methods and
equipment used in irrigated agriculture.
19.2 Classification of Irrigation Systems
The earliest irrigation was by gravity diversion (from natural streams) and from
water lifters powered by humans, animals, or by the flow of water. The on-farm
irrigation systems were supplied either directly from the water source or through
channels supplying a number of farms. Water moved by gravity over the soil
surface was conducted by the irrigator to the crop plants or spread over level basins
limited by small ridges.
Current water distribution systems from the source to the farms use gravity or are
pressurized with pumps. They can be collective or serve single farms. Pressurized
systems use pipes while gravity systems use mainly open channels. The source of
supply may be surface water, groundwater, or both (conjunctive use). A variety of
surface irrigation methods, sprinkle, and drip/micro systems are used for the
application of water to the fields.
In collective distribution systems, delivery schedules determine when each
farmer will receive water and how much, thus affecting on-farm irrigation
19 Irrigation Systems 257
operations and performance. The delivery schedules may be on-demand, arranged,

and fixed rotation. Under on-demand delivery, the user decides when to irrigate,
how much water to apply and for how long. It is typical of modern pressurized
systems. Under fixed rotation, flow rate, frequency, and duration are fixed by the
water authority or agreed within the farmers’ community. Under arranged sched-
ules, rate, frequency, and/or duration are arranged between farmer and water supply
agency.
Irrigation can be applied to the land in several ways. The choice depends upon
many factors, including topography, economics, crop type, soil type, water avail-
ability, farming practices and others. The following major categories of on-farm
systems cover most of the variation:
– surface irrigation,
– sprinkler irrigation,
– localized irrigation (drip/micro irrigation).
A fourth, less common, category is sub-surface irrigation, that consists in
maintaining a saturated water table within reach of the crop roots.
Surface irrigation (also referred to as flood irrigation) consists on the application
and distribution of water over the field by gravity, wetting the entire soil surface or
most of it. The distinguishing feature of this irrigation method is that water moves
over the same medium where it infiltrates to fill the crop root zone. The rate of
infiltration and its spatial distribution are therefore controlled by the soil charac-
teristics. This feature makes it difficult to apply small depths of water, thus surface
irrigation is typically applied at long (from 1 week to more than 1 month) time
intervals. Although the area of surface irrigation is decreasing, it is by far the most
common form of irrigation throughout the world, accounting for about 70 % of the
total irrigated area.
Sprinkler irrigation consists on the application of water similarly to how rainfall
occurs. Water is distributed across the field through a system of pressurized pipes. It
is then sprayed into the air to wet the entire soil surface. The spray heads (sprin-
klers) breakup the water flow into small water drops which fall onto the ground.
Drip/micro irrigation systems apply water directly where the plant is growing,
wetting only a small part of the soil surface and sometimes only part of the root
zone. This is why localized irrigation is another term to call this method. Water is
distributed to the water emission outlets through polyethylene pipes, thus the
irrigation system needs to be pressurized. Water application is generally at a low
flow rate, in small amounts, and frequently, to keep a high water content in the
wetted zones. The water may either be applied above or below the soil surface.
Sprinkler and drip/trickle irrigation are expanding, with current areas of about
20 % and 5 % of the global irrigated area, respectively. The shift from surface to
pressurized systems is being faster in countries like Spain, where surface, sprinkler
and localized irrigation systems account now for 31 %, 22 %, and 47 % of the total
irrigated area, respectively.
258 L. Mateos
19.3 Selecting an Irrigation Method
Irrigation system selection depends on multiple physical and socioeconomic fac-

tors. It should be carried out by experienced professionals that interact with the
farmer or irrigation manager. The factors to be considered are:
(a) Project goals. As in most production activities, the main goal is maximum
economic return, although, for example, if capital is limited, the goal could be
minimize initial cost. There may be social and environmental (conserving water
and water quality) goals that should also be met.
(b) Institutional and social site conditions. The former include legal and political
aspects like financial incentives, land use regulations or water rights that may be
of primary importance. Social site conditions include availability of support for
irrigation equipment maintenance, and dependable labour availability. For
small-holder irrigation in developing countries, additional site conditions to
be considered are: divisibility (suitability of an irrigation method for a wide
variety of field sizes and configurations), skills and effort required for operation
and maintenance, and ruggedness (durability, robustness).
(c) Physical site conditions. These conditions may refer to the water supply and
quality, the land features (soil and topography), the cropping system, the
climate, and the energy availability.
– Water. The source, quantity, quality, reliability and delivery schedule for the
water supply may make some irrigation methods more preferable or impose
constraints to the selection of some of them.
– Topography and field configuration. Field slope, topographic irregularity,
field shape and physical obstructions may preclude the selection of some
irrigation methods.
– Soil characteristics. Soil texture, depth, heterogeneity, infiltration character-
istics, erodibility, salinity, drainability, and the presence of a shallow water
table must all be considered when selecting an irrigation method.
– Crops and cultural practices. Crop height, germination, root and foliage
diseases, and cultural practices (plant spacing, soil tillage and cultivation,
application of fertilizer and pesticides, crop rotation).
– Climate. Precipitation quantity and distribution will determine whether full or
supplemental irrigation is required (Chap. 21), making one or other irrigation
method more adequate and economically feasible. Wind is an important
factor when considering sprinkler irrigation. Irrigation can be used to modify
the crop climatic conditions: regulation of temperature and frost prevention,
regulation of humidity, but not all methods offer the same possibilities.
– Energy availability and reliability. If pumping is required, the energy source
must be dependable.
19.4 Performance of Irrigation Systems
Irrigation performance assessment is needed to evaluate the potential for improve-

ment, for meaningful comparison of irrigation systems and to establish system
design criteria. Several performance indicators have been defined. Although they
are related, they assess different aspects of the irrigation water use process.
The most comprehensive of the irrigation performance indicators is irrigation
efficiency, IE, that relates the amount of irrigation water that is beneficially used by
the crop to the total amount (beneficial plus non-beneficial uses) of irrigation water
that leaves the boundaries (outflow ¼ applied – Δ storage) of the system within a
specified time interval:
Irrig: water beneficially used

IE ¼ 100 ð19:1Þ
Irrig: water applied Δ storage of irrig: water
If at the end of the time period the water contained in the spatial domain is the same
as it was at the start, Δ storage of irrigation water is equal to zero. The system
(spatial domain) may be the soil root zone of a given field, an entire farm, an
irrigation scheme, or a watershed; while the time domain may be, for example, the
interval between two consecutive irrigations, or an entire irrigation season. The
water that is beneficially used is mainly crop evapotranspiration, although it also
includes salt leaching requirements, water needs for soil preparation, seed germi-
nation, seedling establishment and climate control (Chap. 28). Non beneficial uses
include deep percolation in excess of leaching requirements, surface runoff, weeds
transpiration and evaporation from reservoirs, sprinklers and wetted soil. Note that
the term “irrigation water” excludes rainfall.
The determination of IE is complex since it requires detailed quantification of
the fates of water that was applied at earlier dates. Usually the determination of IE is
done by computing a water balance, which most times requires assumptions to
estimate the different components.
Although knowing IE is often necessary to judge the performance of irrigation
systems, the difficulty of its determination is overcome by introducing other
efficiency terms that focus on parts of the water use processes. Application effi-
ciency, AE, evaluates how efficiently the irrigation water applied during a single
irrigation event meets the target irrigation depth:
Avg: depth irrig: water contributing to target

AE ¼ 100 ð19:2Þ
Avg: depth of irrig: water applied
Therefore, the concept of AE avoids the issues of establishing the beneficial use of
the applied water. The target depth may be the root zone soil water deficit, SWD, or
some smaller amount. If excess of water is required for salt leaching, then the target
depth should be greater than the SWD. Implicit in the definition of AE is that the
target depth is uniform over the field. If the target depth is equal to the sum of the
260 L. Mateos
expected beneficial uses, then AE provides an estimate for the potential IE. There-
fore, AE will typically be higher than IE. Furthermore, high AE may not imply high
IE, for instance if the irrigation event takes place at the end of the growing season.
It is easy to attain very high AE in a field by under irrigating. Therefore, AE
should be complemented with an indicator of the degree to which the target depth is
met. This indicator is irrigation adequacy, AD, that can be defined as the fraction of
the field that receives at least the desired amount of water (taking values between
0 and 1). Another expression for adequacy is the so called low-quarter adequacy,
ADlq, defined as:
Ilq
ADlq ¼ 100 ð19:3Þ
It
where Ilq is the mean irrigation depth in the quarter of the field area receiving the
smallest depths and It is the target depth. If meeting the average low-quarter depth is
used as scheduling criterion, then the objective is ADlq ¼ 1.0, meaning that about
one-eight of the field will remain under-irrigated. ADlq <1.0 implies under-
irrigation, whereas ADlq >1.0 implies over-irrigation.
AD and ADlq can be evaluated by plotting the distribution across the field of the
applied depth (Fig. 19.1). The curve is developed by ranking field measurements in
descending order, accounting for the field area that each measurement represents.
The point where the curve intersects the line for target depth indicates the fraction
of the field that is being adequately irrigated.
FRACTION OF THE FIELD

0.00 0.25 0.50 0.75 1.00
0
AD = 73
20 ADlq = 81
AE = 87
40 DUlq= 74
APPLIED DEPTH
60
D low quarter = 73 mm
80
D target = 90 mm
100 D applied avg = 98.3 mm
120
140
Fig. 19.1 Distribution of applied depth across the field. The distribution is ordered from larger to
smaller depth across the field area expressed as fraction of total area. Horizontal lines represent the
average applied depth, the target depth and the average depth of the low quarter
Figure 19.1 also reveals the importance of how uniformly the water is distributed
to the crop. A non-uniform distribution can deprive portions of the crop of needed
water and over-irrigate other portions of the field. The low quarter distribution
uniformity, DUlq, is an indicator of irrigation uniformity:
Ilq
DU lq ¼ 100 ð19:4Þ
Iavg
where Ilq has already been defined and Iavg is the average applied depth. DUlq may
be calculated using the distribution of applied depth like shown in Fig. 19.1.
The distribution of applied water with localized irrigation is based on measure-
ments of the volume of water applied to the soil by each emitter of a defined sample
of emitters. In sprinkler irrigation evaluations, applied water is sampled at the nodes
of a defined grid using catch cans usually placed above the crop canopy. In the
evaluation of furrow irrigation, the applied water is measured only at the furrow
inlet, and the distribution of water is typically derived from measurements of the
infiltration time used to estimate infiltrated water by means of an infiltration
function.
19.5 Design and Management of Irrigation Systems
19.5.1 Surface Irrigation
The surface irrigation process is described in four phases (Fig. 19.2), although not
always all of them take place. The water that is applied to one end of the field (the
high edge or point if the field is not levelled) advances over the soil until it spreads
across the entire surface or flow paths (furrows). This is the advance phase. Then, if
the field is open at its tail end, the water starts to run off; whereas if it is surrounded
by a dike or ridge, the water begins to pond. The interval between the end of the
advance phase and the inflow cut off time is the wetting phase. After water
application is stopped, the water on the surface begins to decline, infiltrating into
the soil and draining from the surface if there is an open field end. Two phases are
distinguished during the drainage period: the depletion phase (or vertical recession)
and the recession phase (horizontal recession). The depletion phase runs from cut
off to the appearance of the first bare soil under the water; the recession phase
begins then and ends when the surface is completely drained.
The infiltration opportunity time is the difference between the recession and
advance times, thus it can be calculated from the advance and recession trajectories
easily (Fig. 19.2). The infiltration rate initially decreases rapidly with time to reach
later a constant rate. Therefore, the variation of infiltrated water across the field is
less than the variation of opportunity time.
262 L. Mateos
RECESSION TIME
RECESSION PHASE
RECESSION CURVE
DEPLETION PHASE
ELAPSED TIME
TIME OF
CUTOFF WETTING PHASE
OPPORTUNITY TIME
ADVANCE CURVE
ADVANCE PHASE
FIELD LENGTH
0
0 L
DISTANCE FROM FIELD INLET
Fig. 19.2 Advance and recession trajectories of the water front in surface irrigation, indicating
(on the right) the phases of an irrigation event
Furrow irrigation is the most common surface irrigation method. Furrows are
small channels formed in the soil by means of a ridger-plough. This method avoids
flooding the entire field surface; water infiltrates through the wetted perimeter.
Furrows may be level, nearly along a contour line with a small slope, or down the
slope of the field. The flow into each furrow is independently controlled, using
siphons, gated pipes or perforated pipes. Furrow irrigation can therefore be used
with a large range of stream sizes by adjusting the number of furrows irrigated at the
same time. The furrow length and inflow rate should be regulated so that water will
flow to the end of the furrow rapidly, but without erosion. This will ensure good
infiltration uniformity, although the tail flow may be too high and thus runoff
excessive. Infiltration uniformity and application efficiency can be improved simul-
taneously by using a high initial inflow rate, until the advance phase is completed,
and cutting back the inflow afterwards. Another way of improving infiltration
uniformity is using surge flow: instead of applying a continuous stream of water,
the flow is intermittently applied through on-off cycles. By surging the water, some
soil surface sealing occurs, thus reducing infiltration and speeding advance during
subsequent surges over previously wetted portions.
Border irrigation requires construction of small earthen dikes (borders) separat-
ing evenly graded basins or strips typically 5–15 m wide. Water is released onto the
border strips through an outlet located near the centre of its top. The slope across the
width of the strips is graded to zero slope, thus the water moves along the
longitudinal gentle slope of the strip. The entire surface of the strips is flooded. A
variation of border systems is level basins, that are perfectly flat and surrounded by
check banks, meaning that all the water applied to a level basin infiltrates into the
soil and there is no runoff. Borders can then be unnecessary.
19.5.2 Sprinkler Irrigation
All pressurized irrigation systems should apply water at a rate below the infiltration
rate of the soil, thus avoiding runoff. Thus, here, the irrigation system controls the
rate of water infiltration while in surface irrigation the soil itself controls the
infiltration rate during irrigation. With sprinkler irrigation, water is distributed
across the field through a pressurized pipe and sprinkled over the soil using nozzles.
Sprinkler devices for agricultural use generally fall into two broad categories:
rotating head sprinklers and spray sprinklers. Rotating head sprinklers move them-
selves in a circle driven by different mechanisms. In the case of impact sprinklers,
this mechanism is an impact arm. The sprinkler head pivots on a bearing as the
impact arm repeatedly hits the water jet pushed by a spring. When the arm hits the
jet, water scatters watering the area around the sprinkler. Impact sprinklers can be
designed to rotate in a full or partial circle. They can have one or two nozzles and
many sizes, allowing flow from 2 to 280 L min1, radius of throw from 7 to 30 m,
with operating pressures between 140 and 690 kPa.
Spray and spinner sprinklers operate typically at pressure less than 200 kPa. The
water jet from a nozzle impinges on a plate that deflects the water in all directions.
The discharge plate can be smooth or serrated and can be flat, concave or convex,
producing different water distribution patterns. Water leaves a smooth plate in
small droplets; serrated plates create tiny streamlets with larger droplet sizes.
Typical radius of throw of spray sprinklers is in the range of 2.5–5 m.
Rotating spray plate sprinklers, commonly referred to as rotators, have features
of both impact and spray sprinklers. The water discharging from the nozzle
impinges onto a circular plate that rotates without the need of an impact arm. The
shape and configuration of the plate determines the multiple trajectories of the
water and its distribution pattern. The radius of throw can be up to 15 m. Rotators
typically operate at lower pressure than impact sprinklers (100–345 kPa) and can
accommodate lower flow rates without compromising performance.
These sprinkler devices are spaced to give a relatively uniform application of
water over the field being irrigated using a series of sets or a continuous move
system. There are several types of sprinkler systems. The most common are
described below.
Hand-move portable systems consist of one or more pipelines (laterals) with
sprinklers mounted on risers connected to the laterals at regular distances. The
height of the risers adapts to the crop height. When the desired amount of water has
been applied to the area covered by the set of laterals, the pipelines are dissembled
and carried to the next position. This operation is repeated until completing the
watering of the entire field. Distribution uniformity may be improved by using
264 L. Mateos
alternate sets: on every other irrigation cycle the pipelines are placed in intermedi-
ate positions. Most hand-move portable systems use rotating impact sprinklers, but
rotating spray plate sprinklers are also used. This system adapts to all topographies,
soils and crops, although moving the laterals when crops such as corn reach their
full size is arduous. Variations of hand-move portable systems are wheel line and
side move lateral systems. In the former system, the lateral line is mounted on
wheels with the pipe acting as axis driven by an engine that moves the whole lateral
from one position to another. In side move systems the lateral is supported by a
frame on which the wheels are mounted.
Stationary or solid-set systems are similar to portable systems except that both
the main line and the laterals remain in place permanently (if the main and laterals
are buried) or during the growing season. Rotating spray plate sprinklers are
becoming the most commonly used sprinkler type for stationary systems. Stationary
systems adapt well to crops that require frequent applications of water, to facilitate
germination and for frost protection. Compared to portable systems, solid-set
systems have high initial investment, but require little operation labour.
A center pivot system consists of a single lateral supported by wheeled towers
that are self-propelled, typically with electric motors, so that the whole lateral
rotates around the pivot point in the centre of the irrigated area. Water is supplied
also at the pivot point. For long center pivot laterals (e.g., 400 m) the period of
rotation may vary between 12 and 120 h, although 12 h to 3 day cycles are most
common, applying between 10 and 25 mm per cycle. Since the outer lateral
segments irrigate annulus of greater area than the inner segments, application
intensity must increase from the pivot point to the lateral distal end in order to
apply uniform water depth. This may generate runoff and erosion in some soils if
the system is not designed and managed properly. The sprinklers used in centre
pivots may be of any kind, but the trend is towards low pressure sprinklers. Centre
pivot systems adapt to most field crops and topographies. They may be unsuitable
for small fields or field shapes where circular geometries do not fit well. Moreover,
the fields must be free of obstacles. Several solutions have been developed to
overcome obstacles, and the installation of end guns (large sprinklers at the end
of the lateral) and corner systems (an additional arm that swings out on the corners
and tuck back in on the edges) allows irrigating part of the corners of squared fields
that would not be irrigated with the conventional centre pivot. Centre pivots can be
used for site-specific variable rate irrigation, by using solenoid valves that regulate
the application rate of each sprinkler.
Linear move systems are similar to centre pivots except that they do not rotate
but translate (move laterally). Water is supplied to the moving lateral using a
flexible hose or from an open ditch parallel to the translation direction. Contrary
to centre pivots, this system adapts well to rectangular fields and applies the water
with uniform intensity across the field. All types of sprinklers can be used in linear
move systems. However, the tendency is to use drop tubes, installed at short
distances along the lateral, and low elevation spray application (LESA) or low
energy precision application (LEPA). LESA systems use sprayers located near the
top of the crop canopy, while LEPA systems use low pressure nozzles located very
close to the soil surface. The soil is furrowed and the furrows are blocked at
regularly spaced intervals to prevent runoff and infiltration non-uniformity.
19.5.3 Drip/Micro Irrigation
Drip/micro irrigation systems are designed to localize the water only to parts of the
soil surface and apply it frequently. The water emitters may be micro tubes, orifices,
nozzles, or perforated pipes. Applications may be from daily to several times per
week, but sometimes (on sandy soils) daily needs are applied in several pulses
throughout the day. The systems may be located on or under the soil surface and are
permanent (solid set). Subsurface drip irrigation (SDI) is relatively new. The
laterals are installed typically at 0.3–0.6 m below the soil surface. This system
allows easier traffic and soil cultivation, reduced weed germination and minimal
soil evaporation, but may have the problems of root intrusion into the emitters and
impaired detection of system failures.
There are many types of emitters that are capable of supplying water directly to
the crop root zone. Drip irrigation generally refers to the use of emission devices
from which water drips onto the soil: drip tape or on- and in-line emitters (small
plastic devices inserted in or embedded in the lateral); while micro irrigation refers
to the use of microsprayers or microsprinklers. Microsprayers and microsprinklers
are connected to the lateral by means of spaghetti hose. Flow rate of these emitters
is very small. Drippers range from 0.5 to 8 L/h and microsprayers or
microsprinklers from 20 to 80 L/h.
Drip/micro irrigation is primarily used for wide-spaced or high-value crops such
as fruits, fresh vegetables and greenhouse crops. The drip/micro systems differ for
permanent crops, non-permanent crops, and greenhouse crops.
Permanent crops include all kinds of fruit tree orchards and vineyards. Drip/
micro systems for permanent crops may use drippers, microsprayers or
microsprinklers as emission devices. For instance, sandy soils or shallow rooted
trees are more effectively irrigated with microsprinklers/sprayers, whereas closely
spaced trees (hedgerow) are better suited for drip lines. Sometimes the tree
branches interfere with the microsprinklers/sprayers, and sometimes wetting only
a small part of the soil prevents diseases. In those two cases drippers are also
preferable. However, microsprayers and microsprinklers can provide some frost
protection. Drip/micro systems for permanent crops typically have one lateral per
plant row if the rows are closely spaced, or two if they are spaced more than, let’s
say, 5 m. The number of emitters per tree depends on the soil type, plant spacing
and type emitter. The idea is to have sufficient root zone wetted volume which
depends on annual rainfall, canopy cover, species and soil type. Drip/micro systems
are also used for non-permanent crops, particularly for row-crops such as vegetable
crops, and also for some field crops (e.g. cotton, tomatoes). After harvest, the
irrigation system is removed to allow land preparation. Drip/micro systems for
266 L. Mateos
row crops mainly use in-line drippers or drip tape as emission devices. Usually
there is one dripline or drip tape per row, but sometimes two are necessary or one
every second row is sufficient to provide water to each plant root zone, depending
on soil physical properties. Drip tapes last for one or two seasons. Drip lines last
longer (7–15 crop seasons) and, if buried (SDI), can remain in the field perma-
nently. Drip/micro systems for greenhouses may use all kind of emitters, depending
on the crop and soil substratum.
Drip/micro systems require clean water to avoid emitter clogging. Filters are
therefore essential components of drip/micro systems, and water filtration accounts
for a great part of the maintenance and operation efforts. Fertigation and
chemigation (the application of fertilizers and pesticides through the irrigation
systems) are relatively easy and common with drip/micro systems, as well as
automation of the operation (initiation and termination of irrigation in the different
system irrigation units).
Drip/micro systems allow uniform and efficient application of water. They adapt
very well to all kind of topographies, field sizes and shapes and crops, and are very
easy to operate. They are more and more used in large scale commercial farming
and in smallholder farming both in developed and developing countries.
19.6 Drainage
Drainage is necessary to evacuate the excess of irrigation or rainfall water and to

wash the excess of salts. This will ensure adequate aeration of the crop root zone
and prevent the harmful effect of salt accumulation. Where natural drainage is
insufficient, surface and/or subsurface drainage must be facilitated artificially.
Surface drainage to remove excess water from the soil surface is expedited by
excavating shallow open ditches. The dimensions and density of the drainage
ditches are calculated based on expected intensity and duration of storms, soil
type and crop.
Subsurface drainage is used to control the groundwater table and to remove salts
for leaching. Many irrigation projects require subsurface drains, that may be deep
ditches or buried perforated pipes. Open ditches have advantages for removing
large volumes of water. They may also serve as collectors of subsurface drains. The
main disadvantages of open drainage ditches are that they occupy land that might
otherwise be cropped, obstruct farming practices, and tend to have high mainte-
nance costs. Clay tile drains have been widely used for subsurface drainage. The
tiles are usually of 30–60 cm lengths and of 10–25 cm inside diameter. Corrugated
plastic tubing is increasingly being used. Perforated pipe lines are generally avail-
able at 8–30 cm diameters. Tile and plastic drains are normally installed with a
surrounding envelop (synthetic fabrics, sand and gravel, or other porous filter
material) that permits water to pass from the soil into the drain without significant
passage of soil particles.
The hydraulic conductivity of a soil is a measure of its drainability; it is therefore

a basic criterion for the design of drainage systems. Proper spacing of drains is
complex, thus it is best determined from field experience obtained under conditions
similar to those of the area to be drained. However several equations have been
developed for estimating the appropriate spacing of subsurface drains if available
field experience is not applicable. These methods give considerations to factors as
hydraulic conductivity, rooting depth, depth of drain, rainfall, irrigation practices,
water quality and soil salinity.
Pumped wells may serve to both lower the groundwater table and, where ground
water is of good quality for irrigation, to supply water that supplements the main
source of irrigation supply. This is called conjunctive use.
19.7 Water Reuse
When applying irrigation water to a field, some water losses (surface drainage and
deep percolation) are unavoidable in most situations. These losses return to the
hydrological system where the irrigated field is located, so they are called return
flows. In many cases, the return flows can be reused downstream, for irrigation or
other purposes. Therefore, the reuse of return flows tends to reduce the global
benefit of improving on-field efficiency. Reuse can take place within the field
(conjunctive use mentioned above), in the farm (e.g., by recycling runoff), at
irrigation district or at watershed levels. The reuse of treated urban wastewater is
one example of the later, although in this case the origin of the return flow is not
from agriculture.
Bibliography
Burt, C. M., Clemmens, A. J., Bliesner, R. D., Merriam, J. L., & Hardy, L. A. (1999). Selection of
irrigation methods for agriculture (ASCE on-farm irrigation committee report). Reston:
ASCE.
Hoffman, G. J., Evans, R. G., Jensen, M. E., Martin, D. L., & Elliot, R. L. (2007). Design and
operation of farm irrigation systems. St. Joseph: ASABE.
Chapter 20
Irrigation Scheduling Using the Water
Balance
Abstract The simplest and most robust method for irrigation scheduling,
i.e. deciding the dates and amounts of irrigation, is based on the water balance.
The soil water deficit (depth of water required to bring the soil to field capacity) is
calculated using ET and rainfall data, and rules are defined for calculating the dates
and depths of irrigation. The rules are based on the critical SWD, the amount of
water that the crop can extract in the rooting depth before water stress occurs. The
critical SWD is the product of root depth, soil available water and allowable
depletion. The later depends on evaporative demand when we want to prevent
reductions in expansive growth. Otherwise most crops can use around 70 % of
stored soil water before stomatal closure occurs. In arid areas we can calculate
mean irrigation calendars for planning purposes. Irrigation scheduling of high
frequency irrigation systems is very simple as it focuses only on irrigating with
an amount equal to actual crop ET since the last irrigation, while ignoring water
storage in the soil.
20.1 Introduction
Irrigation scheduling is a process by which one determines when to irrigate and how
much water to apply by calculating the dates and depths of irrigation.. Measure-
ments of plant (leaf water potential, canopy temperature) or soil water status (water
content, water potential) may be used for scheduling irrigations, but here we will

L. Mateos

Agriculture, DOI 10.1007/978-3-319-46116-8_20
only deal with the simplest, albeit powerful, method, the water balance approach. It
involves the determination of all the inputs and outputs of water from the field, and
it is based on maintaining adequate soil water content in terms of crop performance.
To use this method it is especially important to know exactly crop ET.
Given the technical difficulties in measuring the water balance components in
practice, this method is applied based on estimates of some of the water balance
components. It is therefore highly desirable, when possible, to take measurements
(neutron probe, gravimetric sampling) that allow us to check the accuracy of
estimated values.
This chapter first presents the foundations and applications of the water balance
method. Then the information required for applying the method and the rules for
determining the dates and amounts of irrigation are presented.
20.2 The Water Balance Equation for Irrigation

Scheduling
The water balance equation allows calculating the decrease in soil water content as
the difference between outputs and inputs of water to the field. Therefore the Soil
Water Deficit (SWD) for day t is calculated as:
SWDt ¼ SWDt1 þ ET Ie Pe ð20:1Þ
where Pe is effective rainfall and Ie is effective irrigation. In both cases the term
effective means after discounting losses due to runoff or deep percolation. The Soil
Water Deficit was defined in Chap. 8 as the amount of water required to bring the
soil to the upper limit (Field Capacity). Note that instead of considering the amount
of water in the soil we use a deficit of water that increases from zero when the soil is
at Field Capacity.
Once you have a method to estimate SWD, you must establish rules of decision
to determine the date and the depth of irrigation. The decision rule to adopt depends
on numerous factors such as the crop, the soil, the climate and the irrigation system.
The critical SWD that should not be exceeded is calculated as:
SWDc ¼ ZR PAW AD ð20:2Þ
where ZR is effective rooting depth (m), PAW is plant available water (mm/m),
which is the difference between Field Capacity and Permanent Wilting Point and
AD is the Allowable Depletion (fraction). Commonly found PAW values for sandy,
loam and clay soils are around 100, 150 and 200 mm/m, respectively. An average
PAW value of 120 mm/m has been found for a wide range of light to medium-
textured soils and may be used in the absence of local information.
20 Irrigation Scheduling Using the Water Balance 271
20.3 Effective Rooting Depth
In the context of irrigation scheduling, effective rooting depth (ZR) is the soil depth
where roots can extract most of the soil water, and is equivalent to the soil water
reservoir that is being managed by the irrigator. This depth may be considered
constant for perennials (alfalfa, fruit trees). In annual crops, ZR increases from a
minimum to a maximum value (ZRmax) that depends on the crop and the soil. The
variation of ZR for annual crops can be calculated as follows:
Z R ¼ ZRmin þ ðZ Rmax Z Rmin Þ Rf ð20:3Þ
where ZRmin is the value of root depth at planting, which is equal to the sowing
depth. The maximum value of root depth for annual crops occurs around or after
flowering. The factor Rf describes the rate of growth of the rooting depth during the
cycle and may be calculated as a function of time (t, Eq. 20.4a) or as a function of
thermal time (TT, Eq. 20.4b) from sowing:
t
Rf ¼ ð20:4aÞ
tsm
TT
Rf ¼ ð20:4bÞ
TT sm
where ts-m and TTs-m are the time and thermal time from sowing to maximum
rooting depth, respectively. The primary factor determining maximum rooting
depth is the soil, particularly the soil depth and its mechanical resistance to root
penetration. Therefore, for any given crop the maximum rooting depth depends on
the soil characteristics and that explains the wide variation in maximum effective
root depth of each species presented in Table 20.1. For instance, Table 20.1 shows
that in the case of sunflower the maximum root depth varies from 1 to 2.5 m. In
some extremely open, deep soils it has been found that water extraction by
sunflower crops occurs down to 3 m and more. Irrigation management affects the
distribution of the root system; frequent irrigation promotes root growth in the
surface layers while long irrigation intervals favor more root growth in the deeper
layers. Because mechanical resistance increases exponentially as the soil dries, root
growth occurs very slowly or does not occur at all in dry soil. This is the reason why
it is advisable to start the growing season with a fully charged soil profile (zero
water deficit in the anticipated root zone). In that case, if irrigation is delayed, root
growth will be occurring at progressively deeper moist layers because water uptake
dries progressively the soil layers above and that limits new root growth. The
distribution of roots is therefore the result of the dynamics of root water uptake
and of the application of water, which determine where the conditions are favorable
for new root expansion. This has led to the (wrong) popular belief that roots seek
out water. What roots do is to proliferate where soil environmental conditions are
good in terms of water and nutrient content and temperature. Thus, irrigation
Table 20.1 Maximum Crop Max. root depth FAD

effective root depth (m) under
Alfalfa (hay) 1.0–2.0 0.09
no soil restrictions and factor
for calculating Allowable Apple 1.0–2.0 0.10
Depletion (FAD) for important Barley 1.0–1.5 0.09
agricultural species. The wide Bean (Phaseolus) (dry seed) 0.6–0.9 0.11
interval in root depth shown Coffee 0.9–1.5 0.12
reflects the sensitivity of root
Cotton 1.0–1.7 0.07
distribution to irrigation
management. For information Grapes (wine) 1.0–2.0 0.11
on more species see Appendix Lettuce 0.3–0.5 0.14
10.3 Maize (grain) 1.0–1.7 0.09
Millet 1.0–2.0 0.09
Olives 1.2–1.7 0.07
Orange 1.0–1.5 0.10
Palm trees 0.7–1.1 0.07
Peach 1.0–2.0 0.10
Peas (dry harv.) 0.6–1.0 0.12
Potato 0.4–0.6 0.13
Rapeseed, canola 1.0–1.5 0.08
Rice 0.5–1.0 0.16
Sorghum (grain) 1.0–2.0 0.09
Soybeans 0.6–1.3 0.10
Sugar beet 0.7–1.2 0.09
Sugar cane (virgin) 1.2–2.0 0.07
Sunflower 0.8–1.5 0.11
Tea 0.9–1.5 0.12
Tomato 0.7–1.5 0.12
Winter wheat 1.5–1.8 0.09
management affects the distribution of the root system but not its depth. Frequent
irrigation favors high rooting density in the surface layers, while infrequent irriga-
tion or rainfed conditions generate root systems with less density near the surface
and much more in the deep layers. If such deep soil layers are dry very little root
growth will occur there (Table 20.1).
20.4 Allowable Depletion
Plants can extract soil water down to the Permanent Wilting Point, but crop
performance is affected before that. The Allowable Depletion is the fraction of
plant available water that can be extracted by the crop without negative effects on
yield, but could also be determined as the PAW fraction above which a certain
process such as transpiration, assimilation or growth proceeds unaffected. There-
fore the AD may depend on the process considered and on evaporative demand, as
plant water status is affected by soil water potential and transpiration rate
(Chap. 14). Growth rates may by affected after only 10–20 % of available water
has been used, while transpiration is usually affected after much higher PAW
values (60–70 %). The evapotranspiration of field crops, according to Ritchie, is
reduced after two third of available water has been extracted. This is a simple rule
valid when expansive growth is not critical for yield (full cover situations, harvest
value dependent on dry matter). On the contrary, when growing horticultural crops
we are concerned with the size of the harvestable organ (root, bulb, leaves) so water
deficits should be avoided and a low AD should be adopted. The same would
happen when we deal with field crops during vegetative growth. In those cases,
evaporative demand should be taken also into account, with AD decreasing as ET0
increases:
AD ¼ 1 FAD ET 0 ð20:5Þ
where FAD is a sensitivity factor shown in Table 20.1 for the main agricultural
species (for a more complete list see Appendix 10.3). If Eq. 20.5 yields a value of
AD below 0.2, we adopt AD ¼ 0.2.
Example 20.1 The soil in our farm is of sandy loam texture and 1 m depth.
We want to irrigate maize (after full canopy cover) and onion and need to
know the critical soil water deficit for irrigation management.
The values of maximum ET0 expected are 5 mm/day (onion) and 8 mm/
day (maize). According to the soil type we can adopt a value of potential
available water of 120 mm/m.
According to Appendix 10.1 the maximum root depth of onion is between
0.5 and 0.8 m. We take the mean value (0.65 m). The value of FAD is 0.14
(Appendix 10.1) so using Eq. 20.5 we get:
AD ¼ 1 FAD ET 0 ¼ 1 0:14 5 ¼ 0:3
Now we calculate the critical soil water deficit as:
SWDc ¼ ZR PAWAD ¼ 0:65 m 120 mm=m 0:3 ¼ 23:4 mm
The maximum root depth of maize is between 1.0 and 2 m, which in any case
is greater than the actual soil depth (1 m) so we adopt 1 m as maximum root
depth. As we have reached full canopy cover we are not concerned about
expansion and adopt AD ¼ 0.7. The critical soil water deficit will be:
SWDc ¼ ZR PAWAD ¼ 1:0 m 120 mm=m 0:70 ¼ 84 mm

20.5 Criteria for Irrigation Scheduling
The basic rule for irrigation scheduling is to irrigate just before the soil water deficit
reaches the critical SWD defined in Eq. 20.2, applying a dose equal to SWD. This
rule implies no water deficit with a minimum number of irrigations. However, the
characteristics of the irrigation system may impose restrictions to the dates and
depths of irrigation.
Irrigation scheduling by the water balance is equivalent to operating with a credit
card with a maximum credit (PAW) and a critical credit (critical SWD), above
which interest rates increase (crop yields decrease). We may deposit money when-
ever we want (irrigate) and some friends may unexpectedly deposit money as a gift
(rain). We spend money every day (ET) thus the deficit in the account increases. If
we want to avoid high interest rates and minimize the number of deposits (as they
imply a cost, the labor involved in applying irrigation) we should go to the bank just
before the deficit reaches the critical value and deposit an amount of money equal to
the deficit. This would be the basic rule. If we go sooner to the bank then the number
of deposits would increase. If we go later we will exceed the critical credit and the
interest rate would increase.
20.5.1 Restrictions to the Dates of Irrigation
Some irrigation schemes, typically those with surface irrigation, are organized
according to a rotation and farmers receive water at stated intervals (e.g. weekly)
so the possible dates of irrigation are fixed and the farmer decides the irrigation
amount. Pressurized irrigation systems may work on demand so the farmer chooses
the date and the amount. However, even in this case, dates may be restricted by
holidays or other operations in the farm that are incompatible with irrigation
(e.g. application of pesticides) or require labor or equipment necessary for
irrigation.
20.5.2 Restrictions to the Irrigation Depth
The type of irrigation system and its design may impose restrictions on irrigation
depths. Surface irrigation systems are designed for applying a rather large depth
(over 50–60 mm) and become highly inefficient when smaller depths are applied,
except in smallholder irrigation systems of very small plots. Irrigation machines
(center pivot, lateral move, siderolls) move within a given range of speeds that
determine the irrigation depths that they can apply. Hand-move sprinkler systems
have to be organized taking into account the time required for displacing the lateral,
which usually leads to an optimum irrigation depth (or duration).
20.5.3 Decisions on Irrigation Dates and Depths
The basic rule for deciding when to irrigate is to do it just before the soil water
deficit reaches the critical SWD which is calculated according to the soil and the
crop. If we irrigate sooner than that we will increase the number of irrigations, and
thus the cost, but we will be on the safe side in terms of water deficit and would be
able to cope with possible failures of the irrigation system. If we irrigate later, and
thus SWD exceeds the critical value, water stress will occur and it may have a
negative effect. This would be the case of deficit irrigation, when water available is
not enough to meet the crop water requirement. In the case of rotation of water
supply, for any date when water is available we should irrigate if we expect the
critical SWD to be exceeded before the next date with water supply.
Example 20.2 We are irrigating corn during summer with mean daily
ET ¼ 8 mm/day. Water is available every 10 days (1 July, 11 July, 21 July,
etc.) and the critical SWD is 120 mm. On 1 July the SWD is 20 mm. Should
we irrigate? The answer is no. We can wait until 11 July as SWD will be
20 þ 10 8 ¼ 100 mm. We should irrigate on that date (11 July) as SWD
would exceed the critical value soon after that, so waiting until 21 July should
be discarded.
Once the irrigation date is decided the basic rule is to refill the soil, i.e. bring the
soil to zero SWD, therefore the irrigation depth should be equal to SWD at the date
of irrigation. However, the irrigation depth could be larger than that (excess
irrigation) when salt leaching is required. On the contrary, the depth could be
smaller, thus leaving some soil water holding capacity unfilled which may store
rainfall in the days after irrigating. This strategy may improve rainfall use at the
expense of increasing the number of irrigations.
Example 20.3 Different strategies for date and depth of irrigation are
applied to grain sorghum planted on April 15 in Hinojosa (Spain). The critical
SWD is 94. Figure 20.1 shows the basic strategy (irrigated when SWD
reaches the critical value) and apply a dose equal to SWD. In this case we
apply three irrigations. In Fig. 20.2 a more conservative strategy is followed:
irrigate before SWD reaches the critical value. The number of irrigations
would be 4. If we adopt a strategy designed to take advantage of rainfall
(do not bring the soil to zero deficit) the number of irrigations is also
4 (Fig. 20.3). In Fig. 20.4 we assume a rotational water supply of 10 days.
The number of irrigations would be 5. Note that the farmer does not need to
irrigate every 10 days but only in those days when he cannot wait until the
next possible date for irrigation.
(continued)

TIME (DAP)
0 20 40 60 80 100 120
0
SOIL WATER DEFICIT (mm)
20
40
60
80 SWD
SWDc
100
Irrigation
120
Fig. 20.1 Irrigation schedule of grain sorghum planted April 15 in Hinojosa (Spain). The
curves of soil water deficit and critical water deficit are shown. The strategy followed is to
irrigate when SWD equals the critical SWD, and applying a depth equal to SWD
TIME (DAP)
0 20 40 60 80 100 120
0
20
40
60
80
SWD
100 SWDc
Irrigation
120
irrigate when SWD equals the critical SWD minus 20 mm, and applying a depth equal to
SWD
(continued)

TIME (DAP)
0 20 40 60 80 100 120
0
20
40
60
80
SWD
100 SWDc
Irrigation
120
irrigate when SWD equals the critical SWD, and applying a depth equal to SWD minus
20 mm
TIME (DAP)
0 20 40 60 80 100 120
0
20
40
60
80 SWD
SWDc
100
Irrigation
120
curves of soil water deficit and critical water deficit are shown. Here a fixed rotation water
delivery is assumed and water is available at 10-day intervals. The strategy followed is to
irrigate when the expected SWD 10 days later will exceed the critical SWD, and applying a
depth equal to SWD
The rules about irrigation depths have to be corrected near the end of the crop
cycle in order to leave the soil as dry as possible to allow rainfall storage during
fallow and to prevent excessive deep percolation and thus, nitrate leaching. To
achieve such an objective, the soil water deficit at harvest should approach 80–90 %
of available water in the profile. This can be achieved by solving the water balance
equation since the date of the last irrigation until harvest, which allows calculating
the depth for the last irrigation as:
X
tH
I ¼ SWDtL SWDtH þ ET i ð20:6Þ
tL
where tL and tH refer to the dates of the last irrigation and harvest, respectively.
Example 20.4 We are irrigating a crop whenever SWD reaches a critical

value of 96 mm, which results from determining it based on a soil depth 1 m,
PAW 160 mm/m and AD 0.6. We need to irrigate on September 1 and want to
end the irrigation season on October 1, after using 90 % of the PAW.
Therefore the final SWD should be 144 mm. Average ET during September
is 3 mm/day. Applying Eq. 20.5 we calculate the irrigation depth as:
X
tH
I ¼ SWDtL SWDtH þ ET i ¼ 96 144 þ 30 3 ¼ 42 mm
tL
Note that by applying the correction we could save 54 mm of irrigation water.
20.6 Irrigation Schedules
The dates and depths of irrigation have to be decided on real time according to the
specific weather of each year by updating the soil water deficit day after day
according to crop ET and rainfall. In arid areas where rainfall is negligible during
the irrigation season an average irrigation calendar may be defined a priori using
mean ET values. This approach was originally developed in California and allows a
better planning of the irrigation season as the dates and depths of irrigation for each
crop in the farm are calculated at the start of the season so the allocation of labor
and irrigation equipment can be optimized.
When irrigation systems are permanent (microirrigation or solid-set sprinkler
systems), and there is no labor cost associated with applying an irrigation, timing is
unimportant and irrigation is applied as frequently as desired. Thus, when using the
water balance under high frequency irrigation (e.g. drip), we may ignore, in
principle, the soil water storage, so that the scheduling strategy is to simply replace
the ET accumulated since last irrigation. As explained in Example 20.4, some
irrigation water may also be saved at the end of the season under high frequency
irrigation by safely using some of the stored soil water.
Example 20.5 We are scheduling the irrigation of a tomato crop with crop
coefficient 1.2, and a drip irrigation system with emitters of 2 L/h spaced
1 0.75 m. We will calculate the operation time of the system if yesterday’s
ET0 was 7 mm/day.
The previous day’s ET was:
Kc ET0 ¼ 1:2 7 mm d1 ¼ 8:4 mm d1
Each hour of operation of the irrigation system is equivalent to an applied

depth:

2 L=h= 0:75 m2 ¼ 2:67 mm h1
Therefore the operation time should be:
8:4=2:67 ¼ 3:15 hours d1
Bibliography
Doorenbos, J., & Pruitt, W. O. (1977). Guidelines for predicting crop water requirements (FAO
irrigation and drainage paper n 24). Rome: FAO.
Martin, D., Stegman, E., & Fereres, E. (1991). Irrigation scheduling principles. In T. Howell &
G. Hoffman (Eds.), Irrigation management monograph (pp. 153–203). Saint Joseph: American
Society of Agricultural Engineers.
Villalobos, F. J., & Fereres, E. (1989). A simulation model for irrigation scheduling under variable
rainfall. Transactions of ASAE, 32(1), 181–188.
Chapter 21
Deficit Irrigation
Abstract When the level of irrigation supply is less than crop ET, deficit irrigation
(DI) programs are needed to optimize the use of the limited water. In annual crops
where yield and transpiration are linearly related, DI aims at achieving maximum
profits by minimizing application losses and maximizing use of stored soil water.
Crops that respond positively to mild water deficits are good candidates for DI
programs that decrease irrigation water use while maintaining yield. DI programs
for fruit trees and vines aim at inducing water deficits at periods where they are least
harmful to yields, including high evaporative periods. Achieving optimal use of
limited water is accomplished by solving an optimization problem using knowledge
of water costs and crop prices.
21.1 Introduction
For many decades, the paradigm of irrigation development was to supply crops with
sufficient water to meet their full water requirements. This approach was taken
because the investments associated with the large costs of irrigation networks, from
dams to on-farm equipment, were best justified if farmers would achieve maximum
yields, normally associated with maximum transpiration. However, there were
irrigation development cases where, in order to reach a maximum number of
farmers, the irrigation supply was less than that needed to meet the full crop
demands. In other cases, the capacity of the system was inadvertently less than
needed or the crop rotation scenarios upon which the irrigation design was based,
had changed over time and different crops or most intensive rotations were intro-
duced that required more water than what the original system could supply.
More recently, water allocation to irrigation has been challenged by other sectors
of society with the result that irrigation in many areas receives less supply now than


Agriculture, DOI 10.1007/978-3-319-46116-8_21
what was originally assigned. Furthermore, water scarcity due to periodic droughts
in many areas is increasing thus reducing the original irrigation supply or making it
more unreliable. Under all these conditions, farmers do not have sufficient irriga-
tion water to meet the full crop water requirements and the consequence is that crop
transpiration is reduced below its maximum potential. Normally, this results in a
reduction in crop production of variable magnitude depending on many factors
(Chap. 14). Deficit irrigation (DI) is thus defined as an irrigation management
practice that by applying insufficient water causes crop transpiration to be below
its maximum unstressed value. This chapter discusses how to manage DI to
minimize yield reductions and to maximize farmers’ productivity and profits in
situations of water scarcity.
21.2 The Yield Response Function to Water
The use of DI forces farmers to solve an optimization problem: what is the optimum
level of irrigation water that a farmer should use to maximize its goals? Normally,
farmers aim to obtain the maximum revenue or net income from their operations
which is not necessarily the same as the maximum production. The initial infor-
mation needed to optimize the use of a limited water supply through DI is the
relation between water and yield. If the Y-I relationship is known for a specific
situation at the field scale, the manager can determine the optimum amount of I to
reach maximum net profits. This Iopt will be the level of I above which, the value of
the additional crop produced would be less than the cost of one additional water
unit. Given the close, linear relationship between transpiration and biomass pro-
duction discussed in Chap. 14, and because of the conservative nature of the harvest
index of many crops, the relationship between transpiration and crop yield is linear
for the major crops over a wide range of transpiration, starting with the maximum
yield at the maximum, unstressed transpiration value. However, the relation
between yield and irrigation water (Y-I) is not linear, and is affected by the
uniformity of distribution of irrigation water over the field.
To better understand the Y-I relationships at the field scale, Fig. 21.1 shows the
variation in irrigation depth in different parts of the field, going from the areas that
receive more water than the required depth to the areas that receive less water than
required. Such spatial differences are caused by variations within the irrigation
system that delivers different amounts in different parts of the field due to manu-
facturer and pressure variations, wind effects, and other factors that cause lack of
uniformity in the distribution of water. The shape and slope of the line that
describes the actual distribution of water in the field (Fig. 21.1) is indicative of
the distribution uniformity. If the slope is mild, water distribution is similar in
different areas of the field while if the slope is steep, the uniformity is low with
some areas receiving much more water than others. In this case, much excess water
21 Deficit Irrigation 283
0
NORMALIZED IRRIGATION WATER (X)
0.5
Deficit irrigation
(XR) 1.0
Full irrigation
Deficit
1.5 Overirrigation
2.0
0 0.25 0.50 0.75 1.00
FRACTION OF IRRIGATED AREA (%)
Fig. 21.1 Relations between normalized irrigation water and the fraction of the irrigated area that
receives such water. XR indicates the level of water required as shown by the horizontal dotted line
at 1.0, and another dotted line at 0.7 indicates a level of DI that is 30 % less than XR. The actual
water distribution across the field is shown by the solid lines for full and deficit irrigation, going
from the areas which receive the most water to the areas that receive the least
will be required to supply the required depth to the areas that received the least
water. In the case of DI where the amount of water applied is less than what is
required (see Fig. 21.1), given the same level of uniformity there will be areas
within the field that will receive much less water than what is required and where
yields could be seriously diminished by insufficient irrigation.
As it is not physically possible to achieve perfect uniformity of water application
(100 %), additional water must be applied to arrive at the required depth in the areas
that receive the least water. Figure 21.2 represents a generic yield response function
with two levels of irrigation uniformity, high and low. Also represented is the linear
function between yield and ET. As irrigation increases, more of the irrigation water
in not used in ET and is lost to runoff or deep percolation. If irrigation uniformity is
high, losses would be small and maximum yields will be achieved with little excess
water (Fig. 21.2). However, if uniformity is low, irrigation application has to
increase to meet the needs of the areas that receive the least water while others
will be getting excessive. In the case of Fig. 21.2, water application must double the
required amount to achieve maximum yields under low irrigation uniformity.
Constructing the Y-I function for a given crop and field needed to quantify Iopt,
requires knowledge of the relation between Y and ET and of the actual uniformity
of distribution of irrigation water over the field. Section 21.6 presents an advanced
example where the Y-I function is developed and an expression for Iopt is derived
based on some assumptions.
1.0
RELATIVE YIELD 0.8
0.6
0.4
0.2
0.0
0.0 0.5 1.0 1.5 2.0
RELATIVE IRRIGATION
Fig. 21.2 Generic relations between relative irrigation and relative yield under two levels of
uniformity. Also shown is the relation between relative yield and relative ET (solid line). High
uniformity is represented by the dotted line and low uniformity is represented by the dashed line
21.3 Deficit Irrigation in Annual Crops
To define the Y-I function, the relations between yield and T must be known. While
the B-T relationship is unique for a given crop and is only affected by the
environment in which the crop is grown, the Y-T relationship of a crop can vary
depending on the timing and intensity of the T deficits. This is because of the
differential sensitivity to water stress of the different crop developmental stages.
Water deficits imposed at sensitive stages affect disproportionally more the harvest
index (HI) than biomass production, while the HI does not vary if water deficits
during the sensitive stages are avoided or are not severe. As a general rule, when T
deficits are imposed progressively and are moderate, HI stays constant and the
reduction in Y is proportional to the reduction in B. Figure 21.3 shows the typical Y
responses of an annual crop as ET is decreased by water deficits. The thick line
represents the response to uniform, progressive water deficits and the two steeper
lines represent the responses when water deficits are more severe and/or occur at the
most sensitive stages of crop development. In most annual crops, the most sensitive
stages are those at which yield determining processes (flowering, fruit set and fruit
growth) occur. For instance, in the case of grain crops, the reproductive stages are
more sensitive than the vegetative phase, as discussed below.
Fig. 21.3 Relations 100

between relative crop yield
and relative ET influenced
by the severity of water
RELATIVE YIELD
deficits during sensitive
developmental stages. The
arrow indicates the
response trend as water
deficits are applied at the
most sensitive stages and
with increased severity
0 100
RELATIVE ET
21.3.1 Maize
The most sensitive period to water deficits is flowering, the period that goes from
tasseling to silking. In fact, severe water deficits during that period delay or prevent
silk emergence, impairing pollination and resulting in a very low grain number.
Following in sensitivity is the early grain filling period where severe stress results in
grain abortion. Late grain filling is less sensitive and the least sensitive is the
vegetative phase, as growth recovers after the release of water deficits. Note that
during the vegetative phase, the goal is to develop a canopy that intercepts maxi-
mum radiation, thus severe irreversible reductions in canopy expansion also limit
yield substantially. In general, maize is not a crop conducive to DI, as any
measurable reduction in T has a negative impact on Y. Mild T deficits, around
10 % or less, during the mid to late vegetative phase have the least impact on
production and may be a viable DI strategy in case of water scarcity.
21.3.2 Wheat
As in other winter cereals, flowering is the most sensitive period, in particular the
pollination to grain set period. However, the stages of development which take
place before and after pollination are also quite sensitive. In the pre-flowering
phase, when the number of florets in the spikes is being determined, water deficits
may cause a reduction in grain numbers. After pollination, abortion of grains also
reduces grain numbers under water stress. Grain filling in its early stages is also
quite sensitive to water deficits which affect negatively grain weight. During the
early vegetative phase, wheat is quite tolerant to water deficits but at the tillering
stage, water deficits may reduce tiller numbers. Winter cereals have compensatory
mechanisms, and a reduction in grain numbers may be partially compensated by
increased grain weight if conditions during grain filling are favorable. Thus, deficit
irrigation of wheat is a viable strategy if deficits are restricted to the early vegetative
and late reproductive stages, with mild reductions in transpiration. There are
situations, however, where farmers have limited access to irrigation water, suffi-
cient to apply one irrigation only. In that case, depending on rainfall probability and
soil water storage capacity, the most profitable time to supplement rainfall would be
around flowering. In shallow soils and very low post-flowering rainfall probabili-
ties, it should be delayed as much as possible to avoid severe stress during grain
filling, while in the opposite case, the single irrigation would be most profitable at
the pre-flowering stage to maximize the number of grains which could then be filled
based on stored soil water.
21.3.3 Rice
This crop is generally grown under flooded conditions thus it is not subjected to soil
water deficits. However, there are newer methods of growing rice that are based on
wetting and drying of the soil. Rice is an extremely sensitive plant to water deficits,
has a small, shallow root system and does not tolerate water stress. Research has
shown that the few days (2–3) around pollination time are the most sensitive to
water deficits and can cause almost complete crop failure. Rice is not a crop that can
be subjected to deficit irrigation and the water supply must be concentrated on an
area which permits supplying the crop with its full requirements.
21.3.4 Soybean
Although it is an indeterminate-type plant, flowering of most modern varieties is

determinate, normally occurring during a short period of 2–3 weeks. Water deficits
during flowering and fruit set are most detrimental to yield. The next, most sensitive
period is seed filling, where water deficits cause premature leaf senescence and
reduce the plant’s capacity to fill the seeds. Water stress during the vegetative
period affects canopy cover and hence radiation interception, while if it occurs
during the late part of such period, it is less detrimental, provided it can recover
before it becomes severe enough to cause premature, irreversible leaf senescence.
Soybean is mostly produced under rainfed conditions, thus the likelihood of using
deficit irrigation is small; nevertheless, in terms of relative sensitivity, soybean is
quite sensitive to water deficits, about the same as maize, thus the DI strategy
should be quite conservative, aiming at producing at least 80–90 % of full yields.
21.3.5 Potato
It is very sensitive to water deficits and thus a candidate for DI strategies that only
induce mild water stress. The most sensitive stage is the time during stolon
formation and tuber initiation where the impact on yield is most important. Early
canopy development is also sensitive and where water deficits that have irreversible
effects on canopy expansion should be avoided. The period of tuber growth is less
sensitive but yields are also affected, in particular if canopy senescence is hastened
by water deficits. The relative sensitivity to water stress varies somewhat among the
wide range of varieties that exist.
21.3.6 Sorghum
It is tolerant to water deficits and is a good substitute for maize when water
shortages force the use of DI. Mild to moderate water deficits have little or no
impact on yields due to an increase in harvest index when subjected to water deficits
relative to full irrigation supply. The most favorable DI strategy is to sustain the
level of water deficit throughout the season, for example applying 75 % of full ET
needs, aiming to reach harvest time with the root zone profile exhausted of available
water. It has been shown that severe stress that reduces ET below 50–60 % of
maximum causes a decrease in HI in this crop and is detrimental to yield. Thus DI
strategies in sorghum should aim at reducing ET not more than 60 % of maximum,
and the optimal economic level would be between 70 % and 80 % of maximum
depending on water availability and irrigation costs.
21.3.7 Tomato
As many other vegetable crops it is not amenable to deficit irrigation as yields are
reduced with reductions in ET. However, experimental work in processing toma-
toes showed that mild deficits at ripening reduce slightly ET (by about 10 % or less)
without impacting yield. Thus, the least sensitive stage for imposing water deficits
is the late ripening stage. Stress imposed during other developmental stages reduces
canopy growth, fruiting and yields.
21.3.8 Sugarcane
It is very sensitive to water deficits during the period of stalk growth early in the
crop cycle. The least sensitive period is maturation where mild to moderate deficits
lead to increase sucrose accumulation in the stalk and reduction in harvest fresh
weight (and costs). Therefore, DI should aim at imposing some ET deficits during
the maturation period with reductions of ET of no more than 10–15 % of seasonal
value. On the other hand, water supply should not limit the canopy development
processes until the crop reaches maximum canopy cover.
21.3.9 Cotton
It is a good candidate for DI in areas where the season length is limited by low
temperatures at the beginning and end of season. Water deficits can be used to
adjust the crop to those environments by imposing deficits during the late vegetative
growth period to hasten flowering and boll formation, and during the maturation
phase to enhance boll opening and synchrony for effective mechanical harvest.
Moderate water deficits applied during those two phases result in an increase of HI
relative to that of fully irrigated crops and a reduction in season length that better
fits the environment. Again, as in sorghum, severe water stress is detrimental and
has a negative effect on HI, thus DI should aim at ET deficits that do not reduce the
crop ET below 60–70 % of maximum. The most sensitive period where water
deficits should be avoided is during fruit set to prevent fruit abortion.
21.3.10 Sunflower
It is a drought avoiding species that has very high growth rates during the vegetative
phase but, during reproductive development accumulates dry matter much more
slowly as it concentrates it in seeds that have high-energy fats and protein content.
The period that goes from end of flowering to early fruit set is the most sensitive to
water deficits, followed by early seed filling, late seed filling, early vegetative, and
late vegetative, as the least sensitive. This is because sunflower has a strong
compensatory growth capacity following water deficits during the vegetative
phase. DI strategies should aim at imposing moderate deficits during the vegetative
phase, well after crop establishment but ending it before leaf growth is completed.
Such a period is not very long relative to the crop cycle; thus, ET reductions should
not exceed 10–15 % of seasonal ET for optimum production.
21.4 Deficit Irrigation of Fruit Trees and Vines
The most powerful measure to respond to water scarcity in annual crops is to adjust
the planted area to the available supply. This is not possible in permanent crops
where the investment costs are substantial and maintenance of the plantations is the
primary objective. Occasionally, farmers take advantage of drought years to replace

old orchards with new ones of different species or cultivars. With that exception,
the rest of the time there would be the need to use some form of DI to cope with the
limited supplies and to minimize the impact of water deficits. This is particularly
important because of the impact of water deficits on subsequent years’ production
in permanent crops, and furthermore to avoid the possibility of tree death with the
economic consequences due to the multi-year investment losses if the plantation
is lost.
The responses of fruit trees and vines to water deficits are much more complex
than those of annual crops. The yield determining processes are less well under-
stood than those of annual crops, and commonly, yields oscillate with time even
under optimal growing conditions, a feature known as alternate bearing that varies
in importance among the different fruit tree species and cultivars. Water deficits in
the current season affect yields of the subsequent season/s, depending on its timing
and severity. Knowledge of the differential sensitivity of the different processes to
water deficits thus become important to allocate the application water in such a way
that avoids impacting negatively on current year and/or subsequent year’s yield.
Experiments with peach and pear trees demonstrated that irrigation may be
restricted in periods when yields are insensitive to water stress and water can be
saved with minimal or no yield loss, and sometimes producing more net profits than
under full irrigation. This approach of purposely stressing tree crops and vines at
certain developmental stages considered the least sensitive to water deficits has
been called regulated deficit irrigation (RDI), a term now used for any DI approach
followed in trees. However, there is another DI approach which consists in applying
a fraction of the ET throughout the irrigation season. This sustained or continuous
DI strategy (CDI), generates a stress pattern which is different than that of RDI, at
least in temperate environments, where ET0 varies widely during the growing
season. The usual stress pattern in CDI starts early in the season when a fraction
of ET is applied, but trees extract additional water from the wet soil with the
consequence that water deficits seldom develop. Depending on the magnitude of
the deficit and on the water storage capacity of the root zone, water deficits will
increase in magnitude in CDI as the season progresses, reaching the highest level
when ET0 is highest close or near the end of the season, in the absence of rainfall.
21.4.1 Regulated Deficit Irrigation
In RDI, the physiological basis for imposing water stress at specific developmental
stages resides in the differential sensitivity of vegetative growth and of photosyn-
thesis to water stress (Chap. 14). Expansive growth is more sensitive to water
deficits than is carbon assimilation, thus mild to moderate water stress slows
down or stops leaf and stem growth before it impacts photosynthesis. During fruit
growth of some tree species, there are periods when vegetative growth takes place
and fruit growth slows down almost to a stop. In those periods, water stress reduces
growth in general but fruit growth can fully recover upon release of stress while
vegetative growth has been curtailed. The result is less vegetative growth (and less
pruning costs) while fruit size is unaffected. There are some species where moder-
ate water deficits change the partitioning of assimilates in favor of the fruits and this
also results in beneficial effects in terms of yield and fruit quality. In deciduous
trees, the period after fruit harvest is another insensitive period where water deficits
may be applied without impacting yield, provided that next year’s bud development
processes have been completed.
One important benefit of DI in fruit trees and vines is related to the enhancement
of fruit quality features caused by moderate water deficits. The most important case
is that of wine grapes. Plants grown under moderate water deficits applied both
early during berry growth and late after the berries changed color, promote the
synthesis of a number of biochemical compounds that are essential wine quality
components. In fact, high-quality wines are produced mostly in rainfed environ-
ments or under substantial DI programs, while wines produced from grapes grown
under unlimited supply are seldom of good quality. Other fruits benefit from an
increase in the content of sugars or other positive changes in composition induced
by DI, although, contrary to wine production, the quality features of most fruits are
not determinants of the price that farmers receive for their produce. In fact, fruit size
is a factor that is appreciated in many markets and DI tends to reduce fruit size
relative to that of full irrigation. In this particular case, the economic loss of DI
makes its use not viable, such as in the case of apple.
The yield response to ET deficits generated by RDI strategies may be general-
ized in Fig. 21.4. Three different response regions may be defined; Region A, where
small reductions in ET applied through RDI during insensitive periods do not
impact yields. As ET deficits increase in magnitude, yield decreases (Region B)
albeit at a low rate. If water stress becomes more severe, yield decline decreases
Fig. 21.4 Generalized

yield responses to ET 100
deficits induced by RDI
strategies in fruit trees
expressed in relative terms
RELATIVE YIELD
C B A
0
0 RELATIVE ET 100
steeply (Region C) and water deficits of that magnitude can impact yields very
severely if they occur partly in more sensitive periods (Dashed line in Fig. 21.4).
The magnitude of the response in Region A varies depending on the species;
from the very sensitive ones such as avocado, walnut and apple where it has not
been found, to about 5–15 % ET reductions in citrus (depending on species and
cultivars), pears, almonds, among others, raising to 10–20 % ET reductions in
peach, plum, apricot, pistachio, up to 20–30 % ET reductions without impacting
yields as in the olive.
Region B also varies in magnitude with species and cultivars but follow the
differential sensitivities indicated for the different species in Region A. The mild
slope indicates that the water savings must be compared against yield losses to find
an optimum net income under DI. On the contrary, the steep slopes of Region C
suggest that ET deficits of such magnitude will have detrimental effects on net
income and should be avoided at all costs.
21.5 Deficit Irrigation at Farm Level
Water restrictions that occur at farm level force managers to make strategic
decisions before planting. They must decide what crops to grow, and how to
allocate different water amounts to each of the crops to make best use of the limited
supply available. Different crops need different irrigation amounts and their irriga-
tion timing is also different. Economic issues (markets and subsidies) are critical in
determining the farm net profits once the total supply available is known. The goal
would be to optimize the use of land and irrigation water given the supply
restrictions, and this is achieved through the use of economic optimization models.
These models provide optimal cropping patterns and optimal irrigation amounts for
each of the crops that maximize the objective function which normally is the total
farm income. Once the model is built, it can be used to explore different scenarios
of varying crop and water prices and of other factors, such as the impact of changing
subsidies on the strategic decisions by farmers, for example. One complicating
factor at the farm level is the uncertainty of the level of water restrictions. Too
often, water authorities delay too much the communication of the restriction to
farmers after some of the best options for using the limited water (for example,
early planting) are no longer viable. In the event of a drought, farmers must be
proactive in securing their water supplies and should maintain open communication
with the water authority to be able to make the best decisions under uncertainty.
21.6 Quantitative Optimization of Deficit Irrigation
We start with the assumption that biomass production is proportional to transpira-

tion and that Harvest Index does not change with the level of water stress. This is
valid for mild to moderate stress. Therefore yield will proportional to transpiration
in any location of the irrigated field. We also assume that evaporation from the soil
surface (Es) is equal in all points of the field and that excess irrigation has no penalty
on yield.
Then we need to consider the variation of irrigation applied. We assume that
applied irrigation follows a uniform (i.e. rectangular) distribution. Therefore
applied irrigation (I) will lie between a minimum Imin ¼ Imean (2 Ucc-1) and a
maximum Imax ¼ Imean (3–2 Ucc), where Imean is the mean irrigation applied and Ucc
is the uniformity coefficient (Chap. 19). The irrigation crop water requirement was
defined in Chap. 10 as:
IWRn ¼ ET Pe SWE ð21:1Þ
where ET is unstressed (i.e. maximum) seasonal crop ET, Pe is effective precipi-

tation and SWE is water uptake from the soil between planting and harvest.
We have three possible situations:
(a) Imax < IWRn the whole field is under-irrigated as even where the maximum
application occurs it does not satisfy the net irrigation requirement. In this case
yield will be proportional to transpiration, and thus to the irrigation applied in
each point of the field.
(b) Imin > IWRn the whole field is over-irrigated as even where the minimum
application occurs it exceeds the net irrigation requirement. In this case yield
will be the same in the whole field and equal to Yx (maximum yield).
(c) Imin < IWRn < Imax parts of the field are over-irrigated and parts are under-
irrigated, thus parts of the field will reach Yx while others will yield in
proportion to irrigation received in that location.
To calculate average yield in the whole field we calculate the integral of yield
using applied irrigation (I) as integration variable. For instance for case c we
calculate:
Z Z
1 IWRn
ET Es IWRn þ I I max
Y¼ Yx dI þ Y x dI ð21:2Þ
I max I min I min ET Es IWRn
And after integration we get the relationship between actual yield (kg/ha) and mean
applied irrigation (Imean, mm):

B
Y ¼ Yx A C I mean ð21:3Þ
I mean
where
3 2 U cc ð2 Ucc 1ÞðSWE þ Pe Es Þ
A¼ ð21:4Þ
4 ð1 U cc Þ 4 ð1 U cc ÞðET Es Þ
IWRn 2
B¼ ð21:5Þ
8 ð1 U cc ÞðET Es Þ
ð2 Ucc 1Þ2
C¼ ð21:6Þ
8 ð1 Ucc ÞðET Es Þ
For case a (Imax < IWRn):
Yx
Y¼ ðSWE þ Pe Es þ I mean Þ ð21:7Þ
ET Es
And for case b ðImin > IWRn Þ : Y ¼ Y x

The model illustrates several interesting points:
(a) To achieve maximum yield it is necessary to apply IWRn/(2 Ucc1). For
instance, with Ucc ¼ 0.75 the gross irrigation amount should double the irriga-
tion requirement.
(b) It is clear that the best strategy to achieve maximum profits will likely be to
apply less water than that required for maximum yield. The optimum value will
depend on the price of the crop at harvest (PH, €/kg) and the cost of irrigation
water (QI, €/m3):
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
B
I opt ¼ 10 QI
ð21:8Þ
PH Y x þC
From that we may calculate an apparent optimum efficiency (AOE) of irrigation

as:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
IWRn 10 QI ðET Es Þ 8 ð1 U cc Þ
¼ þ ð2 Ucc 1Þ2 ð21:9Þ
I opt PH Y x
This equation indicates that AOE is related to the irrigation system (Ucc), to the crop
species and location (ratio (ET-Es)/Yx) and to external economic factors (ratio QI/
PH). As the difference ET-Es is equal to seasonal transpiration, the ratio (ET-Es)/Yx
is proportional to the inverse of Water Use Efficiency (Chap. 14) so it should be
proportional to VPD and differ between C3 and C4 species.
Example 21.1 We are growing maize with maximum yield of 12,000 kg/ha
and seasonal ET of 750 mm with Es¼150 mm. The values of Pe and SWE are
150 and 100 mm, respectively. The irrigation system is sprinkler with
Ucc ¼ 0.75. The price of maize is 0.3 €/kg. If the water price is 0.05 €/m3
the apparent optimum efficiency will be 0.645 so we should apply IWRn/
AOE ¼ 500/0.645 ¼ 775 mm. In this case yield would be 11,836 kg/ha (from
Eq. 21.3). With water price of 0.15 €/m3 we should apply 577 mm
(AOE ¼ 0.866) to get a final yield of 11,226 kg/ha.
Bibliography
Fereres, E., & Soriano, M. A. (2007). Deficit irrigation for reducing agricultural water use. Journal
of Experimental Botany, 58, 147–159.
Steduto, P., Hsiao, T. C., Raes, D., & Fereres, E. (2012). Crop yield response to water (Irrigation
and drainage paper 66). Rome: FAO.
Chapter 22
Control of Salinity
Francisco J. Villalobos, Luciano Mateos, Miguel Quemada,

Antonio Delgado, and Elias Fereres
Abstract Salinity is a threat to the sustainability of many agricultural systems and

especially for irrigated areas in arid and semi-arid zones. Besides the possible
specific toxicity the main effect of salts is the reduction of soil osmotic potential
causing an effect similar to that of water deficit. The expected yield under saline
conditions can be calculated with a simple model whose parameters are the
threshold ECe and the yield loss per unit increase in ECe. The salt balance equation
allows us to deduce soil salinity based on the EC of the irrigation water and the
leaching fraction. The Leaching Requirement is calculated taking into account the
irrigation frequency, the EC of the irrigation water and the desired ECe. The
presence of Sodium deteriorates soil structure but the effect depends on other
factors, in particular the salinity of irrigation water. The reclamation of saline
soils is performed by salt leaching while sodic soils are usually reclaimed adding
gypsum and leaching Na.

L. Mateos
M. Quemada
ETS de Ingenierı́a Agronómica, Alimentaria y de Biosistemas, Universidad Politécnica de
Madrid, Av. Puerta de Hierro, nº 2, 28040 Madrid, Spain
A. Delgado
Sevilla, Spain

Agriculture, DOI 10.1007/978-3-319-46116-8_22
22.1 Introduction
According to FAO around 20 % of irrigated land occupies salt-affected soils while

salts affect less than 3 % of dryland agricultural area. Salt-affected land is mainly
located in irrigated arid and semi-arid areas. Preventing the buildup of salts in soils
and waters with adequate management is crucial for avoiding land degradation and
thus for ensuring the sustainability of agriculture. In the case of irrigated areas,
sustainability relies on maintaining an appropriate salt balance in the soil which
depends on the salt concentration of irrigation water and the soil water balance.
The primary source of salts is the weathering of minerals, but this source hardly
causes soil salinization of agricultural soils. Soil salinization is the consequence of
an external supply of salts which in irrigated agriculture is associated with the
application of waters that contain salts, while in rainfed agriculture, it is related to
secondary salinization phenomena, usually caused by upward movement of water
and salts from the subsoil effected by changes in hydrology. Therefore, the salt and
water balances are tightly coupled. The salt concentration in the soil solution is
greater than that of the water that infiltrates. This difference is a result of soil
evaporation not carrying salts while root water uptake only does so very selectively.
Therefore, salinity is a real threat in agricultural areas of arid or semi-arid regions
due to the high ET rates and by the scarcity of water needed for leaching the salts
out of the root zone.
The contribution of salts from rain water is insignificant but can be important in
coastal areas when winds from the sea carry salts in aerosols. Irrigation water itself
is often the cause of salinization, as all irrigation waters contain salts in varying
concentrations. On the other hand, excess of irrigation causes percolation that
leaches salts which may eventually reach the sea. The sea can also be the direct
source of salinization in some cases: the soils originating from marine sediments
can be saline or marine intrusions can cause salinization of groundwater origin,
i.e. by upward movement of water from a saline water table. In addition to the
natural sources of salt, excessive application of fertilizers can also contribute to
salinization.
22.2 Effects of Salinity on Crops
Salinity is quantified as the electrical conductivity of the soil solution (EC, dS/m) or
by the salt concentration (CS, g L1) which are approximately related by EC ¼ 0.64
CS. The concentration of cations (CC, or anions) in meq/L is also related to the EC
by CC ¼ 10 EC. The osmotic potential (Ψo, kPa) may be estimated from the salt
concentration as Ψo ¼ 56.25 CS. The electrical conductivity is commonly measured
in a saturated soil extract (ECe) which is approximately half the value of the EC at
field capacity (ECFC).
22 Control of Salinity 297
The effects of a high salt concentration in the soil on crop performance are due
to: (i) the reduction of osmotic potential in the soil solution, which in many aspects
mimics that of water deficits, and (ii) the toxicity and nutrition disorders (antago-
nistic effects) caused by major ions present in salts (Cl, Na, B). The negative effects
of salinity on crops depend on the particular species and the developmental stage.
An additional problem occurs with sodium accumulation in soil, a problem defined
as soil sodicity, that causes degradation of soil structure.
The main difference between soil water deficit and the osmotic effects due to
salts is that a low soil water content has two effects. The first one is to lower the soil
water potential, so roots will in turn need a lower water potential for water uptake to
occur. The second is the decrease in hydraulic conductivity of dry soil, so that even
with a lower potential in the roots, water movement to the root surfaces is limited by
the low hydraulic conductivity. The effect of salinity corresponds to the first effect
of low soil water content, but the hydraulic conductivity is not affected in the case
of salinity.
Example 22.1 A non-saline soil (called A) at wilting point has a matric

potential of 1500 kPa and soil water potential is 1500 kPa. We have
another soil (called B), with the same texture, but saline. Soil B is at field
capacity (matric potential 20 kPa) and has ECe ¼ 8.4 dS/m
(so ECFC ¼ 16.8 dS/m). Therefore soil B has osmotic potential 1480 kPa
and soil water potential 1500 kPa. Despite having the same soil water
potential transpiration is zero for soil A, while in the case of soil B it can
be significant for several crop species (e.g. rye).
22.2.1 Impacts of Salinity on Crop Productivity
A soil is classified as saline when the electrical conductivity of the saturation extract
is greater than 4 dS/m. This level represents less than 10 % of the salinity of
seawater (50–60 dS/m), but is sufficient to affect negatively the growth of many
crop plants. Increasing the concentration of salts in the soil involves a decrease of
osmotic potential, and therefore a decrease in the soil water potential and thus, in
shoot water potential, which ultimately leads to reduced expansive growth.
There is a wide variability in the sensitivity of crops species to salinity, and also
among cultivars within a species. A simple response model has been proposed
where the sensitivity or tolerance is quantified by two parameters. The first is a
threshold salinity value, ECeu (measured as EC of the soil saturation extract) above
which yield decreases. The second parameter (Bs) is the percentage of yield loss per
unit increase in ECe. The relative yield (Y/Yx) expressed as a fraction of the
maximum yield can then be calculated as:
Y Bs
¼1 ðECe ECeu Þ ð22:1Þ
Yx 100
Maas and Hoffman published in 1977 a review on the crop tolerance to salinity,
which results are summarized in Table 22.1. The crops are classified as sensitive,
moderately sensitive, moderately tolerant and tolerant according to the values of the
parameters ECeu and Bs (Table 22.2). Most fruit trees are sensitive, most horticul-
tural crops are moderately sensitive and most field crops lie between moderately
sensitive to moderately tolerant.
Table 22.1 Sensitivity of different crop species to salinity

Na or Cl in Na in soil
ECeu Bs Type water B in soil (ESP)
%/ meq/L Na or mg/L saturated
Common name dS/m dS/m Cl extract %
Alfalfa (hay) 2 7.3 MS 10.0–20 4.0–6.0 >40
Apple 1 18 S <15
Barley 6 7.1 T 10.0–20
Bean (Phaseolus) 1 18.9 S <15
(dry seed)
Cotton 7.7 5.2 T >20 6.0–15.0 >40
Grapes (wine) 1.5 9.5 MS 5–10.0 0.5–0.75
Lettuce 1.3 13 MS – 2.0–4.0 15–40
Maize (grain) 1.7 12 MS 10.0–20 2.0–4.0 <15
Millet MS
Olive 5.00 7 MT
Orange 1.3 16 S <5 0.5–0.75 <15
Palm trees 4 3.6 T
Peach 1.7 21 S – 0.5–0.75 <15
Peas (dry harv.) 2.5 MS 1.0–2.0 <15
Potato 1.7 12 MS 5.0–10 1.0–2.0
Rapeseed, Canola 10.5 13.5 T
Rice 11.4 10.8 MT 15–40
Sorghum (grain) 6.8 16 MT 10.0–20 4.0–6.0 15–40
Soybeans 5 20 MT
Sugar beet 7 5.9 T >20 4.0–6.0 >40
Sugar cane (virgin) 1.7 5.9 MS 15–40
Sunflower 5.5 25 MS >20 0.75–1.0
Tomato 2.5 9.9 MS 5.0–10 4.0–6.0 15–40
Winter wheat 6 7.1 MT – 0.75–1.0 15–40
For the response to total salt concentration two parameters are shown: ECeu (dS/m) is the value of
ECe below which yield is not affected. Bs (%/(dS/m)) is the slope of the linear relationship between
yield (% of maximum) and ECe. Crops are classified as sensitive (S), moderately sensitive (MS),
moderately tolerant, (MT) and tolerant (T). Concerning the foliar damage by sprinkler irrigation,
the threshold concentration of Na or Cl is shown (meq/L). The maximum concentration of B (mg/L
in soil saturated extract) and Na (ESP, %) in the soil above which toxicity may occur are also
shown
Adapted from Ayers and Westcott (1989)
Table 22.2 Examples of species in the four groups of response to salinity and typical parameters
of the response
Moderately
Tolerant tolerant Moderately sensitive Sensitive
ECeu (dS/m) 6.0–10 3.0–6.0 1.3–3.0 0–1.3
Slope Bs 12 13 17 30
(typical)
(%/(dS/m))
ECe for zero <32 <24 <16 <8
yield (dS/m)
Fibre, seed Barley Winter cereals Sugarcane Bean
and sugar (wheat, oat, rye,
crops triticale)
Cotton Legumes (cow- Legumes (faba bean, Guayule
pea, soybean) peanut)
Sugarbeet Sorghum Summer cereals Sesame
(maize, rice, millet)
Safflower Oilcrops (castorbean,
flax, sunflower)
Forage Wildrye Wildrye (some Alfalfa
crops (some spp.) spp.)
Wheatgrass Wheatgrass Common vetch
(some spp.) (some spp.)
Bermuda Barley, wheat Oats, rye, maize
grass
Ryegrass, fes- Brome
cue, sudangrass
Clover Clover (Trifolium spp.)
(Melilotus spp.)
Vegetable Asparagus Red beet Solanaceae (potato, Carrot
crops tomato, pepper,
eggplant)
Zucchini squash Cucurbitaceae (cucum- Onion
ber, melon, water-
melon, squash)
Brassicaceae (cabbage, Parsnip
cauliflower, kale, tur-
nip, broccoli)
Other (lettuce, spinach,
celery, radish, sweet
potato)
Fruits and Date palm Olive, fig, Grape Citrus spp. (orange,
nuts pomegranate lemon, tangerine,
etc.)
Pineapple Prunus spp. (peach,
almond, cherry,
etc.)
(continued)

Moderately
Tolerant tolerant Moderately sensitive Sensitive
Papaya Berries (Rubus
spp., Ribes spp.)
Pome fruits (apple,
pear)
Other (strawberry,
avocado
Tropical fruits
(cherimoya,
mango, passion
fruit)
The information in Table 22.1 has to be taken with caution since the response of
plants varies with development stages and growing conditions (climate, soil man-
agement and irrigation, cultivar, etc.). Furthermore, it should be noted that the data
in Table 22.1 were determined with surface irrigation following conventional
management, including the application of excess water to obtain a steady state
and uniform distribution of salts throughout the root zone. However, soil ECe may
vary substantially in space and during the growing season, so the application of
Eq. 22.1 is not straightforward, but would represent the response of a given crop to
the long term, uniform application of an irrigation water of a given EC.
22.2.2 Specific Toxicity
Toxicity occurs when certain ions are absorbed by the plant and accumulate to
concentrations high enough to cause crop damage. Toxicity related to saline
conditions is ascribed to ions usually present in soluble salts, mainly Na and Cl,
and to B, frequently found in saline irrigation waters. The first symptoms are
usually marginal leaf burn and interveinal chlorosis. The sensitivity of crops to
ion toxicity is quite variable with tree crops being the most sensitive, which may be
affected by very low ion concentrations (see Table 22.1). Under high evaporative
demand ion accumulation is faster, thus enhancing the toxic effect.
Toxicity can also be due to direct absorption of Cl or Na through the leaves
(e.g. sprinkler irrigation). This may be an important problem in certain sensitive
crops such as citrus.
22.3 Irrigation Systems and Distribution of Salts
In general, salts accumulate in lands where the leaching is limited, i.e. they tend to
concentrate close to the soil surface, and the depth at which salts accumulate
depends on the amount of drainage water. Lack of drainage causes the build up
of a water table near the surface from where water evaporates, bringing salts to the
uppermost soil layers causing salinization. This is shown in Fig. 22.1 where a saline
water table at a depth of 100 cm and EC¼10 dS/m is the source of water that
evaporates, resulting in a high concentration of salts on the surface. A situation of
this type has also occurred under rainfed conditions, for example in the Murray
Basin and other areas in Australia. The substitution of evergreen forests that have
high rates of ET for crops that use less water, has changed the hydrology of basins,
leading to the accumulation of excess water in the low lands where the water table
rose with the consequent soil salinization. The problem is of such magnitude that it
is estimated that in 50 years it could affect 25 % of the cultivated area of Australia
unless specific measures are adopted to alleviate the problem.
Figure 22.2 shows a different profile of salt concentration. In this case the salts
were accumulated in the lower layers due to leaching. A period of heavy rainfall
during winter helped in salt leaching and the EC decreased across the soil profile.
This would be the typical profile of salts in irrigated semiarid areas without the
presence of a water table.
Since salt movement is related to water movement in soil, the irrigation method
affects the distribution of salts in the soil. Figure 22.3 distinguishes areas where
salts concentrate more or less depending on the method of irrigation. Furrow
irrigation provides leaching in almost the entire surface of the ground, except in
the ridges. Drip irrigation wets only part of the ground and therefore leaching
ECe (dS/m)
0 5 10 15 20 25 30
0.0
0.2
DEPTH (m)
0.4
0.6
0.8
1.0
Fig. 22.1 Soil salinity profile above a saline water table (ECw ¼ 10 dS/m) at the end of a growing
season of a sorghum crop
ECe (dS/m)
0 1 2 3 4 5 6 7
0.0
Initial
After 220 mm
0.4
Afer 420 mm
DEPTH (m)
0.8
1.2
1.6
2.0
Fig. 22.2 Soil salinity profiles in a sandy loam soil before (initial) and after 220 and 420 mm of
rainfall
FURROW DRIP SPRINKLER
Fig. 22.3 Accumulation of salts from irrigation water for different methods of irrigation (Adapted
from Rhoades and Loveday and Stewart and Nielsen 1990)
occurs only under the emitters while the remainder of the surface tends to accumu-
late salts, particularly at the edges of the wetted zones. Overhead sprinkler or flood
irrigation leaches salts evenly throughout the field depending on the uniformity of
irrigation.
The EC of the soil solution increases as the soil water is depleted so the osmotic
effect of salts depends on the actual soil water content. This is illustrated in
Fig. 22.4 where changes in EC of the soil solution are parallel to changes in soil
water potential. Irrigation frequency also affects the concentration of the salts. High
frequency irrigation keeps high soil water content and is therefore able to maintain a
lower salt concentration in soil solution. In addition, a relatively constant high soil
water content contribute to counteract the effect of salinity on water potential by
avoiding low matric potentials in soil. On the contrary, low frequency irrigation
involves drying cycles as those shown in Fig. 22.4, which may lead to very low soil
water potential. This is one reason why drip irrigation in cotton with saline water is
advantageous as compared to traditional furrow irrigation.
The potentially harmful salts in the soil are those in areas where root absorption
takes place. This must be considered if we conduct a survey of soil salinity on a plot.
–0.6
–0.7
low frequency
–0.8
high frequency
Soil water potential (MPa)
–0.9
–1.0
–1.1
–1.2
–1.3
–1.4
–1.5
–1.6
0 10 20 30 40 50 60 70 80 90 100 110
Time (days after planting)
Fig. 22.4 Time course of soil water potential of a sorghum crop irrigated with water of
ECw ¼ 4 dS/m with leaching fraction 0.2 and low or high irrigation frequency
Fig. 22.5 Effects of bed shape and plant distribution on salt accumulation and emergence of
furrow irrigated crops (Adapted from Rhoades and Loveday and Stewart and Nielsen 1990)
A special problem is that of seeds and seedlings. Germinating seeds and seed-
lings are especially sensitive to salinity so crop establishment is the most critical
stage for some species, in particular under furrow irrigation. The seedbed config-
uration, the design of the irrigation system and the distribution of plants can have a
huge impact on crop establishment in saline conditions. Figure 22.5 illustrates
various situations for furrow irrigated crops. Planting on flat ridges with lateral
furrow irrigation leads to an accumulation of salts in the center of the ridge. This
can be avoided by widening the bed and creating a small furrow in the middle to
separate the two central rows. The situation can be further improved if planting is
performed on sloping beds with a central furrow which is watered until crop
establishment.
In sodic soils (see Sect. 2.4.4), surface crusts develop and prevent seedling
emergence. This may be alleviated by sprinkler irrigation (small doses, high
frequency) which also would reduce salt concentration close to the seeds.
Alternatively, mechanical removal of the crust in the rows could be beneficial

although it is impractical except for small plots.
22.4 Salt Balance and Leaching Fraction
To quantify the risk of soil salinization it is necessary to evaluate its salt balance.
This balance implies that the quantity of salts entering the system minus the amount
going out must equal the increase in the salt content of the soil (ΔS):

Vw Cw þ Vr Cr þ Ss þ Sf Vd Cd þ Sp þ Sc ¼ ΔS ð22:2Þ
where V and C refer to volume and concentration, respectively, corresponding to

the irrigation water (w), rain (r) and drainage (d). Ss, Sf, Sp and Sc refer to the
amounts of salts released from the soil (by dissolution or weathering), provided as
fertilizer, precipitated and absorbed by the crop, respectively, during the period.
To prevent soil salinization the amount of salts in the soil should be constant
(ΔS ¼ 0). In order to simplify Eq. 22.2, assuming Ss+Sf¼Sp+ Sc and neglecting the
term related to rainfall, we may write:
Vw Cw ¼ Vd Cd ð22:3Þ
On the other hand, the Leaching Fraction (LF) is defined as the ratio of the drainage
volume to the irrigation volume. Using Eq. 22.3, and replacing concentration by
electrical conductivity, LF can be expressed as:
LF ¼ Vd =Vw ¼ ECw =ECd ð22:4Þ
Example 22.2 The electrical conductivity of irrigation water is 1 dS/m.

Drainage is 25 % of applied irrigation. The leaching fraction is thus 0.25
and the EC of the drainage water is obtained by dividing ECw by 0.25,
i.e. ECd ¼ 4 dS/m.
Therefore, by changing the LF, it is possible to control the concentration of salts

in the soil water within values above the salt concentration in the irrigation water.
This is shown in Fig. 22.6 for different Leaching Fractions. Note that the soil EC
increases with increasing depth, taking values similar to that of the irrigation water
at the top. As LF decreases this variation with depth is more pronounced.
Fig. 22.6 Generic profiles

of soil salinity for high and
low leaching
fractions. Salinity is shown
as the ratio of EC of the
saturation extract and
DEPTH
irrigation water EC
HIGH LOW
0.5 1.0 1.5

ECd/ECw
22.5 Salinity Profiles and Crop Yields
The profiles of EC in the soil are affected by the EC of irrigation water and the LF as
we have seen above. This EC refers to the soil with its water content at Field
Capacity. To convert to EC of saturation extract (which is required to apply
Eq. 22.1) we use a factor of 0.5, as ECe is roughly half of EC at Field Capacity
(ECFC).
We will now show how EC varies with depth in the soil with the following
example.
Example 22.3 A bean crop (ECeu ¼ 1.0 dS/m, Bs ¼ 19 %/dS/m) is irrigated

with water of 1.5 dS/m with a leaching fraction of 0.2. We assume a water
uptake distribution of the type 40–30–20–10 (the soil is divided into four
parts of equal depth from where 40 %, 30 %, 20 % and 10 % of the total water
extraction occurs, respectively).
For every 100 mm applied as irrigation, 80 mm are absorbed by the root
system and 20 mm are drained. Applying Eq. 22.4 to each of the four layers in
which we have divided the profile we may deduce the average EC (at Field
Capacity) for each layer.
In the first layer water uptake is 40 % of total (0.4 80 mm) so the amount
of water draining from this layer is the difference between the water applied
(100 mm) and that absorbed, i.e. drainage is 10032 ¼ 68 mm. Therefore for
the first layer, LF and the calculated EC are:
LF01 ¼ ð100 0:4 80Þ=100 ¼ 68=100 ¼ 1:5=ECFC1 ; ECFC1 ¼ 2:2
(continued)

And for the other layers:
LF12 ¼ ð68 0:3 80Þ=68 ¼ 44=68 ¼ 2:2=ECFC2 ; ECFC2 ¼ 3:4

LF23 ¼ ð44 0:2 80Þ=44 ¼ 28=44 ¼ 3:4=ECFC3 ; ECFC3 ¼ 5:35
LF34 ¼ ð28 0:1 80Þ=28 ¼ 20=28 ¼ 5:35=ECFC4 ; ECFC4 ¼ 7:5
The values of EC of the saturation extract will be half of those values: 1.1,
1.7, 2.7 and 3.75 dS/m.
The arithmetic mean of ECe will be 1.93 dS/m.
The weighted EC average considering water uptake will be:
0:4ð0:75 þ 1:1Þ=2 þ 0:3ð1:1 þ 1:7Þ=2 þ 0:2ð1:7 þ 2:7Þ=2

þ 0:1ð2:7 þ 3:75Þ=2
¼ 1:56 dS=m
If we use the weighted average which is recommended for high frequency

irrigation we would predict a relative yield of 89 % of the maximum. Using
the arithmetic mean (recommended for low frequency irrigation) we would
predict a relative yield of 82 %.
It is possible to deduce analytically the profile of soil EC and then the mean
values of EC may be computed as a function of EC of the irrigation water and the
LF (Fig. 22.7):
(a) The weighted average does not depend on the distribution of root water uptake
and may be calculated as:
lnðLFÞ
ECem1 ¼ 0:5 ECw ð22:5Þ
1 LF
(b) If water uptake decreases linearly with depth like in Example 22.3 then the
arithmetic mean will be:
pffiffiffiffiffiffi
arc cos LF
ECem2 ¼ ECw pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð22:6Þ
LFð1 LFÞ
where the arccos function is given in rad.

A simpler equation may be deduced from the Leaching Requirement (see below)
equation commonly found in the salinity literature:
1 þ LF
ECem2 ¼ ECw ð22:7Þ
5LF
3.5
3.0 WEIGHED MEAN

ARITM MEAN
2.5
ECe/ECw
2.0
1.5
1.0
0.5
0.0
0.0 0.2 0.4 0.6 0.8 1.0
LEACHING FRACTION
Fig. 22.7 Mean soil salinity (relative to irrigation water EC) as a function of the Leaching
Fraction computed as a weighed (of root water uptake) or arithmetic average
Example 22.4 Following example 22.3 of the bean crop (ECeu ¼ 1.0 dS/m,
Bs ¼ 19 %/dS/m) irrigated with water of ECw ¼ 1.5 dS/m and LF ¼ 0.2 we
can evaluate the average ECe using Eqs. 22.5 and 22.7.
Weighted average:
ECem ¼ 0:5 ECw lnðLFÞ=ð1 LFÞ ¼ 0:51:5 lnð0:2Þ=0:8 ¼ 1:51 dS=m
Relative yield would be:
Y Bs 19
¼1 ðECe ECeu Þ ¼ 1 ð1:51 1Þ ¼ 0:90
Yx 100 100
The result would be very close to that obtained using a numerical approach
(Example 22.3).
Arithmetic mean:
1 þ LF 1 þ 0:2
ECem2 ¼ ECw ¼ 1:5 ¼ 1:8 dS=m
5LF 5 0:2
And the expected yield (as fraction of the maximum) is:
Y Bs 19
¼1 ðECe ECeu Þ ¼ 1 ð1:8 1Þ ¼ 0:85
Yx 100 100
22.6 Leaching Requirement
The Leaching Requirement (LR) is the LF desired to keep a given ECe (ECem) and
will depend on the crop sensitivity (Eq. 22.1) and the EC of irrigation water. Then
the amount of irrigation to be applied would be:
ET
AW ¼ ð22:8Þ
1 LR
The LR could be obtained by solving Eqs. 22.5 and 22.6 for LF but no analytical
solution is possible. Then approximate equations may be used:
For low irrigation frequency:
1:7
ECw
LR ¼ 0:31 ð22:9Þ
ECem
Or, alternatively, the following equation is commonly found in the literature:
ECw
LR ¼ ð22:10Þ
5ECem ECw
For high irrigation frequency:

3
ECw
LR ¼ 0:18 ð22:11Þ
ECem
Example 22.5 A barley crop is irrigated with water of ECw 5 dS/m. To reach
a relative yield of 90 % of the maximum we should keep an average ECem that
satisfies the following equation:
90 ¼ 100 5:0ðECem 8:0Þ
Thus ECem ¼ 10 dS/m.

Therefore the LR for high frequency irrigation will be:
3
5
LR ¼ 0:18 ¼ 0:022
10
And for low frequency:

1:7
5
LR ¼ 0:31 ¼ 0:095
10
(continued)

or
5
LR ¼ ¼ 0:11
5 10 5
Example 22.6 A farm has a limited amount of irrigation water from a canal
(EC ¼ 0.4 dS/m). It also has a well of unlimited water supply with 2.5 dS/m.
We want to grow peppers (ECeu ¼ 1.5, B ¼ 14) with a low frequency irriga-
tion system that has a leaching fraction LF ¼ 0.2. How can the irrigated area
be expanded by using water from the well?
Using only water from the canal:
1 þ LF 1 þ 0:2
ECem ¼ ECw ¼ 0:4 ¼ 0:48 dS=m
5LF 5 0:2
This value is lower than the threshold EC (1.5 dS/m) thus yield would be
maximum.
With water from the well:
1 þ 0:2
ECem ¼ 2:5 ¼ 3:0 dS=m
5 0:2
And the relative yield would be:
Y Bs 14
¼1 ðECe ECeu Þ ¼ 1 ð3 1:5Þ ¼ 0:79
Yx 100 100
The two waters could be blended to obtain a given ECem. For instance if
we want to obtain the maximum yield then the average ECe should be 1.5 dS/
m, which implies the following equation:
1 þ LF
½f ECwell þ ð1 f ÞECcanal ¼ 1:5
5 LF
Where f is the fraction of water taken from the well. In our case we may
deduce f ¼ 0.4, i.e. by mixing 60 % of water from the canal and 40 % from the
well we could achieve maximum yield while increasing the irrigated area by
67 %.
We could also change the LF. For instance, by increasing LF up to 0.3 we
could use 63 % of water from the well thus further increasing the
irrigated area.
22.7 Optimum Leaching Requirement
From the sections above it is clear that to maintain crop productivity we may
increase the LF, and thus water applied. In other words irrigation water of poor
quality (high ECw) is equivalent to having a smaller amount of fresh water in terms
of crop productivity.
In the previous section we determined the Leaching Requirement when the
objective yield is known. This may lead to unreasonably high values of LR when
ECw and irrigation costs are high. In this case we may be interested in maximizing
the Crop Water Productivity (yield per unit of irrigation applied). To do that we
may apply Eq. 22.1 for different values of LR until a maximum CWP is found. In
the case of low irrigation frequency it is also possible to deduce an analytical
solution for LRopt by maximizing the function:

1 þ LR
f ðLRÞ ¼ ð1 LRÞ 1 B0 ECw ECeu ð22:12Þ
5LR
Where B0 ¼Bs/100. The function is maximized when:

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
0:2 B0 ECw 1
LRopt ¼ 0 < ð22:13Þ
1 þ B ðECeu 0:2ECw Þ 5 EC EC 1
eu
w
As the inequality indicates, the solution is valid below the value of LR at which
maximum yield is achieved.
Example 22.7 We want to irrigate peach (ECeu ¼ 1.7, Bs ¼ 21) with irriga-
tion water of ECw ¼ 3 dS/m.
Using Eq. 22.13 we deduce LRopt ¼ 0.35. The solution is valid as it is
lower than the limit LR to get maximum yield:
1 1
¼ 1:7 ¼ 0:54
5 EC 1 5 3 1
eu
ECw
If we had water with ECw ¼ 1 dS/m, a value of LRopt ¼ 0.18 would be

deduced but now it would exceed the limit value (0.13). Therefore in this
case the optimum LR would be 0.13.
The previous analyses are based on a steady state salt concentration in the soil
solution which would be achieved after continuous use of a given irrigation water.
However, ECe may change during the season and from year to year. For instance, in
Mediterranean areas rainfall is concentrated in winter which would provide for salt
leaching (at least partial, depending on amount) and thus to ECe below the steady
state value. Therefore the calculation of Leaching Requirement based on the steady
state solution leads to an upper limit for LR. The desired LR would vary depending
on the amount of winter leaching, and thus, on the actual ECe at the start of the
irrigation season, which should be measured routinely to keep track of soil salinity
trends.
22.8 Sodicity and Soil Structure
Sodium-affected soils, referred to as sodic soils, are those with high levels of
adsorbed (exchangeable) Na in soil. High exchangeable Na promotes the dispersion
of soil colloids and consequently the degradation of soil structure. This degradation
involves deterioration in aggregate stability, soil water transmission capacity due to
soil surface sealing, and increased susceptibility to crust formation. Crusting is
particularly relevant since, in addition to limit crop establishment, it affects the
water balance by decreasing infiltration and enhancing runoff, and subsequently
increasing erosion risk.
Crusting is the consequence of the breakdown of aggregates and the closing of
the pores in soil surface resulting from rapid soil wetting accompanied by a
dispersive effect of the impact of raindrops. In sodic soils the problem is exacer-
bated because of the dispersing effect of sodium, which leads to a lack of aggregate
stability and large reduction in the infiltration rate, with the attendant problems of
waterlogging and increased runoff.
A soil is classified as sodic when the Exchangeable Sodium Percentage (ESP)
exceeds 15 % of cation exchange capacity. In practice, an indicator used is the
Sodium Adsorption Ratio (SAR, meq0.5/L0.5) in the soil saturation extract which is
approximately equal to ESP:
Na
SAR ¼ qffiffiffiffiffiffiffiffiffiffiffi ð22:14Þ
CaþMg
2
where all the concentrations are expressed in meq/L.

ESP can be more precisely estimated from SAR in the saturation extract follow-
ing the empirical equation:
ð100 0:0126 þ 0:01475 SARÞ

ESP ¼ ð22:15Þ
1 0:0126 þ 0:01475 SAR
SAR is an index applied to solutions, and also for irrigation water. It is also used to
assess the effect of the sodium content of the irrigation water on soil structure and
hydraulic conductivity due to an excess of sodium in relation to calcium and
magnesium. However, this negative effect depends on the EC of irrigation water
as the dispersing effect of Na is counteracted by the aggregating effect of a high salt
Table 22.3 Quality criteria for irrigation water

Degree of restriction on use
Potential irrigation problem Units None Slight to moderate Severe
Salinity reduces crop growth and transpiration
ECw dS/m <0.7 0.7–3.0 >3.0
Specific ion toxicity (affects sensitive crops)
Sodium (surface irrigation) SAR <3 3–9 >9
Sodium (sprinkler irrigation) meq/L <3 >3
Chloride (surface irrigation) meq/L <4 4–10 >10
Chloride (sprinkler irrigation meq/L <3 >3
Boron mg/l <0.7 0.7–3.0 >3.0
Infiltration is reduced
SAR ¼0–3 And ECw dS/m >0.7 0.7–0.2 <0.2
¼3–6 dS/m >1.2 1.2–0.3 <0.3
¼6–12 dS/m >1.9 1.9–0.5 <0.5
¼12–20 dS/m >2.9 2.9–1.3 <1.3
¼20–40 dS/m >5.0 5.0–2.9 <2.9
Miscellaneous effects
Bicarbonate (whitewash, sprinkler) meq/L <1.5 1.5–8.5 >8.5
pH Normal range 6.5–8.4
concentration in solution (Table 22.3). In addition, it is necessary to take into

account changes in Ca in the soil water that take place resulting from precipitation
(as calcium carbonate) or dissolution during or following irrigation. Sodium
remains soluble and in equilibrium with exchangeable soil sodium at all times,
whereas calcium concentration, however, varies until the equilibrium is established.
Dissolution is promoted by dilution and by carbon dioxide dissolved in the soil
solution. Precipitation may take place when the presence of calcium is accompa-
nied by enough carbonate, bicarbonate or sulfate to exceed the solubility of calcium
carbonate (limestone) or calcium sulfate (gypsum). This is why different correc-
tions of the SAR value in irrigation water have to be considered for a more realistic
estimation of the potential effects of irrigation water on soil structure and hydraulic
conductivity. The most common correction is done by estimating a corrected Ca
concentration (Cax):
pffiffiffiffiffiffiffiffiffi
HCO3 þ CO3
Cax ¼ exp 0:552 þ 0:1637 ECw 0:668 ln ð22:16Þ
Ca
which is used to calculate an adjusted SAR (now called Adjusted Sodium Ratio,
SARx):
Na
SARx ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffi ð22:17Þ
Cax þMg
2
Example 22.8 The composition of irrigation water is:

Ca: 1.3 meq/l, Mg: 1.6 meq/l, Na: 6.6 meq/l, CO3: 0.4 meq/l, HCO3:
4.00 meq/l, ECw: 0.90 dS/m
The ratio (CO3+HCO3)/Ca is (4.0 + 0.4)/1.3 ¼ 3.38
The adjusted Ca concentration is:
pffiffiffiffiffiffiffi
Cax ¼ exp½0:552 þ 0:1637 0:9 0:668 lnð3:38Þ ¼ 0:90
And the adjusted Sodium Ratio:
6:6
SARx ¼ qffiffiffiffiffiffiffiffiffiffiffiffi ¼ 5:9
0:9þ1:6
2
According to Table 22.3 which summarizes the possible problems related to

the quality of irrigation water, this water would cause almost no problem in
terms of water availability, slight to moderate problems of infiltration, toxic
effects in sensitive crops and slight to moderate problems of “whitewash” due
to bicarbonate when applied by sprinkler irrigation.
22.9 Reclamation of Saline Soils
In general a soil is classified as saline when its ECe is greater than 4 dS/m although
the Soil Science Society of America uses a threshold of 2 dS/m. Soluble salts most
commonly present are the chlorides and sulfates of Na, Ca and Mg. Sodium and
chloride are the most dominant ions, especially in highly saline soils, but Ca and Mg
concentrations are usually enough to meet the crop needs. Many saline soils contain
appreciable quantities of gypsum (CaSO4 ∙ 2H2O) while soluble carbonates are
never present. The pH of the saturated soil paste is always lower than 8.2 and is
usually close to 7.
Saline soils usually have good physical properties as excess salts keep the clay in
a flocculated state. Some saline soils, in special heavy clays, tend to disperse when
leached with low salt water.
Saline soils can be recognized by the spotty growth of crops (irregular plant size,
barren spots) and often by the presence of white salt crusts on the surface. This is
because there is always substantial spatial variability in the soil water properties
that leads to wide spatial variations in soil salinization. If salinity is moderate and in
the few cases where it is more uniform across the field it may go undetected as it
may not cause visible symptoms other than reduced growth rate, with the exception
of a blue-green tinge in some cases. Symptoms of salinity stress may resemble those
of water deficit without wilting due to gradual osmotic adjustment. Symptoms of
specific toxicities (marginal or tip burn of leaves) are typical in woody plants.
The reclamation of saline soils is performed by salt leaching. The important

question is the amount of water required and the best application strategy. Evacu-
ation of drained water must be guaranteed, and if required to this end, a drainage
network should be installed.
Theoretically, if the soil behaves as a perfect porous medium and assuming that
there is no precipitation or dissolution of salts, the change in salts stored in the soil
after applying an amount of water (AW) with ECw are given by:

AW
ECefinal ¼ ECeinit exp
Zðθsat θinit Þ

AW
þ 0:5 ECw 1 exp ð22:18Þ
Zðθsat θinit Þ
where Z is the soil depth being considered and subscripts sat and init refer to
saturation and initial conditions, respectively.
Example 22.9 The soil has ECe ¼ 10 dS/m with drains at 1 m depth. We
apply 1000 mm with ECw ¼ 0.1 dS/m when the soil is at PWP (0.10 m3 m3).
Soil water content at saturation is 0.40 m3 m3.

1 1
ECe f inal ¼ 10 exp þ 0:5ECw 1 exp
1ð0:4 0:1Þ 1 ð0:4 0:1Þ
¼ 0:357 þ 0:048 ¼ 0:40 dS=m
Several empirical models have been proposed to estimate the volume of water
required. Each soil differs in behavior and the same amount of water has different
leaching efficiencies in different soils, so field experiments are usually performed to
determine the amount of water required. We have seen before (Example 22.9) that
an amount of water equal to the depth of the soil, would leach theoretically 96 % of
the salts. However in practice such an amount which is equivalent to 1.5–2 times the
pore volume removes only around 70 % of soluble salts.
Sprinkler irrigation is more efficient in salt leaching than surface irrigation. This
is mainly due to preferential flow occurring under saturated conditions, leaving part
of the soil without leaching. For the same reason, intermittent application of surface
irrigation is more efficient (although slower) than continuous application. To
calculate the depth of water required (Iw, mm) to reclaim a given soil depth Z
(mm) to go from an initial EC (ECinit) to a final desired EC (ECfinal) we may use the
following formula:
Iw ECeinit 0:5 ECw

¼ kleach ð22:19Þ
Z ECefinal 0:5 ECw
where the parameter kleach depends on soil properties (soil water content at satura-
tion and texture) and irrigation method. For continuous ponding kleach ¼ 0.45 in
peat soils, 0.30 in clay loams and 0.1 in sandy loams. For intermittent ponding or
sprinkler irrigation kleach ¼ 0.10.
Example 22.10 The soil has ECe ¼ 10 dS/m with drains at 1 m depth. We
plan to reclaim the whole soil depth (Z ¼ 1000 mm) to reach a final
ECe ¼ 2 dS/m using sprinkler irrigation (so kleach ¼ 0.1) and water with
ECw ¼ 1 dS/m. Using Eq. 22.19:
Iw ECeinit 0:5 ECw 10 0:5 1

¼ kleach ¼ 0:1 ¼ 0:633
Z ECefinal 0:5 ECw 2 0:5 1
Therefore the amount of irrigation to apply will be 633 mm.
In the case of saline-sodic soils the addition of gypsum may help in improving
water infiltration and thus accelerate both desalinization and desodification of the
soil, in special in heavy textured soils, or when low electrolyte water is applied. The
application of amendments should be tested by trials on an experimental scale for
large scale reclamation projects.
22.10 Reclamation of Sodic Soils
The main characteristic of sodic soils is the high content of exchangeable Na that
adversely affects soil properties and the growth of most crop plants. By definition,
sodic soils are those with Exchangeable Sodium Percentage greater than 15. In
many cases the ECe is not too high (less that 4 dS/m). The pH is 8.2 or higher and in
extreme cases it may be above 10.5. Dispersed and dissolved organic matter in the
soil solution may be deposited on the soil surface by evaporation generating a dark
surface which is why these soils have also been termed black sodic soils.
The objective of reclamation of a sodic soil will be decreasing the amount of Na
in the exchange complex and/or increasing the amount of Ca. Leaching of Na alone
may be difficult because of the low permeability of sodic soils. This may be
improved by adding electrolytes (chemical amendments added to the irrigation
water) and tillage.
Calcium is usually added as calcium chloride, calcium carbonate or gypsum, the
latter being the most frequently used. The availability of gypsum has increased in
recent years as it is the by-product of scrubbing sulfur dioxide gases from the
emissions of coal-fired power plants. Although gypsum is not a liming agent, it
reduces the Aluminum toxicity that often accompanies soil acidity.
Gypsum amendments are normally applied broadcast and then incorporated with
the soil by disking or ploughing. Fine ground gypsum is more quickly solubilized.
When the problem is a surface crust, gypsum needs are reduced. If the problem is in
deeper layers, gypsum contributions should be much larger. Gypsum may also be
applied dissolved in the irrigation water which increases the efficiency.
In general the reclamation process with gypsum is performed in several stages.
The common practice is to make a first application of approximately 10 t/ha of
gypsum in the first year with 1.5 m of water. In subsequent years (2 or 3)
applications of gypsum of 4 t/ha accompanied by some leaching may be performed.
According to the USDA the amount of gypsum to apply (kg/ha) may be
calculated as:
Gypsum amount ¼ 860 Fg ρb Z CEC ðSARi SARf Þ ð22:20Þ
where ρb is soil bulk density (t/m3), Z is soil depth (m) to be restored, CEC is the
Cation Exchange Capacity (mmol/kg), SARi and SARf are the initial and final
values of SAR in the saturation extract of soil, respectively, and Fg is an efficiency
factor that varies from 1.1 (SARf ¼ 0.15) to 1.3 (SARf ¼ 0.05).
Example 22.11 Soil bulk density is 1.3 t/m3, CEC is 400 mmol/kg. The
initial SAR is 0.20 and we want a final value of 0.10. The amount of gypsum
to recover the top 0.3 m layer of soil will be:
Gypsum amount ¼ 860 Fg ρb Z CEC ðSARi SARf Þ

¼ 860 1:2 1:3 0:3 400 ð0:2 0:10Þ ¼ 16099 kg=ha
Appendix 22.1: Sensitivity of Different Crop Species

to Salinity
For the response to total salt concentration two parameters are shown: ECeu (dS/m)
is the value of ECe below which yield is not affected. Bs (%/(dS/m)) is the slope of
the linear relationship between yield (% of maximum) and ECe. Crops are classified
as sensitive (S), moderately sensitive (MS), moderately tolerant, (MT) and tolerant
(T). Concerning the foliar damage by sprinkler irrigation, the threshold concentra-
tion of Na or Cl is shown (meq/L). The maximum concentration of B (mg/L in soil
saturated extract) and Na (ESP, %) in the soil above which toxicity may occur are
also shown.
Response to salinity Toxicity

mg/L
meq/L Na saturated
Species dS/m %/dS/m or Cl extract %
Cereals and Na or Cl in Na in soil
pseudocereals ECeu Bs Type water B in soil (ESP)
Barley, grain 8 5 T 10.0–20 0.75–1.0 >40
Maize (grain, 1.7 12 MS 10.0–20 2.0–4.0 <15
sweet)
Millet MS
Oats 5 20 MT 2.0–4.0 15–40
Rice, paddy 3 12.2 S 15–40
Rye 11.4 10.8 MT 15–40
Sorghum 6.8 16 MT 10.0–20 4.0–6.0 15–40
Wheat 6 7.1 MT – 0.75–1.0 15–40
Wheat, durum 5.9 3.8 T – 0.75–1.0 15–40
Forages ECeu Bs Type water B in soil (ESP)
Alfalfa 2 7.3 MS 10.0–20 4.0–6.0 >40
Barley, forage 6 7 T 10.0–20
Barley, hay 6 7.1 T 10.0–20
Bermuda grass 6.9 6.4 T >40
Clover (red) 1.5 12 MS
Clover, berseem 2 10.3 MS 15–40
Clover, white 1.5 12 MS – 2.0–4.0 15–40
Cowpea 2.5 11 MS
(vegetative)
Fescue 3.9 5.3 MT 15–40
Lovegrass 2 8.5 MS
Maize (forage) 1.8 7.4 S 10.0–20
Meadow foxtail 1.5 9.7 MS
Orchard grass 1.5 6.2 MS
Paspalum 1.8 9 MS 15–40
Phalaris 4.2 MT
Ryegrass 5.6 7.6 MT 15–40
Sesbania 2.3 7 MS
Setaria 2.4 12.2 MS
Siratro 2 7.9 MS
Sudan grass 2.8 4.3 MS
Townsville stylo 2.4 20.4 MS
Trefoil, big 3 11.1 MS
Trefoil, birdsfoot 5 10 MT
Vetch 3 11 MS 4.0–6.0 15–40
Wheatgrass, 3.5 4 MT >40
crested
(continued)

mg/L
meq/L Na saturated
Wheatgrass, 7.5 6.9 T >40
fairway
Wheatgrass, tall 7.5 4.2 T >40
Fruit trees, vines Na or Cl in Na in soil
and shrubs ECeu Bs Type water B in soil (ESP)
Almond 1.5 18 S <5 <15
Apple 1 18 S <15
Apricot 1.6 24 S <5 0.5–0.75 <15
Avocado 1.3 21 S – 0.5–0.75 <15
Banana MS
Blackberry 1.5 22 S <0.5 <15
Boysenberry 1.5 22 S <15
Cherry – – S <5 0.5–0.75 <15
Coconut MT
Date palm 4 3.6 T
Fig 4.2 MT 0.5–0.75 <15
Grape 1.5 9.5 MS 5–10.0 0.5–0.75
Grapefruit 1.8 16 S <5 0.5–0.75 <15
Lemon 1 – S <5 <0.5 <15
Orange 1.3 16 S <5 0.5–0.75 <15
Peach 1.7 21 S – 0.5–0.75 <15
Pear 1 S <15
Pineapple MT
Plum 1.5 18.2 S 0.5–0.75 <15
Pomegranate 4 MT
Prune 1.5 18 S <5 <15
Raspberry 1 S <15
Rosemary 4.5 MT
Walnut S 0.5–0.75 <15
Horticultural Na or Cl in Na in soil
crops ECeu Bs Type water B in soil (ESP)
Artichokes 6.1 11.5 MT 2.0–4.0
Asparagus 4.1 2 T 6.0–15.0
Bean (green) 1 18.9 S <15
Beet (table) 4 9 MT >20 4.0–6.0 >40
Broadbean 1.6 9.6 MS
Broccoli 2.8 9.1 MS
Brussels sprouts 1.8 9.7 MS
Cabbage 1.8 9.7 MS – 2.0–4.0
Cauliflower 1.8 6.2 MS >20
(continued)

mg/L
meq/L Na saturated
Celery 1.8 6.2 MS 2.0–4.0
Cucumber 2.5 13 MS 10.0–20 1.0–2.0
Eggplant 1.1 6.9 MS
Kale 6.5 T
Lettuce 1.3 13 MS – 2.0–4.0 15–40
Melons 2.2 7.3 MS – 2.0–4.0
Pea 2.5 ms 1.0–2.0 <15
Pepper 1.5 14 MS 5.0–10 1.0–2.0
Pumpkin, winter 1.2 13 MS
squash
Radish 1.2–2.0 7.6–13.0 MS 1.0–2.0 15–40
Spinach 2.0–3.2 7.7–16.0 MS 15–40
Squash 2.5 MT 2.0–4.0
Squash, scallop 3.2 16 MS
Squash, Zucchini 4.7 10 MT
Strawberry 1 33 S – 0.75–1.0
Tomato 2.5 9.9 MS 5.0–10 4.0–6.0 15–40
Watermelon – – MS
Legumes ECeu Bs Type water B in soil (ESP)
Bean (dry) 1 18.9 S <15
Chickpea MS <15
Cowpea (seed) 4.9 12 MT 0.5–0.75 <15
Faba bean 1.6 9.6 MS
Pea 1.5 14 S 1.0–2.0 <15
Peanut 3.2 29.4 MS – 0.75–1.0 <15
Soybean 5 20 MT
Roots, tubers and Na or Cl in Na in soil
bulbs ECeu Bs Type water B in soil (ESP)
Carrot 1 14 S – 1.0–2.0 15–40
Onion 1.2 16.1 S – 0.5–0.75 15–40
Parsnip – – S
Potato 1.7 12 MS 5.0–10 1.0–2.0
Sweet potato 1.5 11.1 MS 0.75–1.0
Turnip 0.9 9 MS 2.0–4.0
Cassava MS
Garlic 3.9 14.3 MS 0.75–1.0
Sugar, oil and Na or Cl in Na in soil
fiber crops ECeu Bs Type water B in soil (ESP)
Cotton 7.7 5.2 T >20 6.0–15.0 >40
(continued)

mg/L
meq/L Na saturated
Castorbean – – MS
Flax/Linseed 1.7 12 MS
Kenaf 8.1 11.6 T
Olive 5.00 7 MT
Rapeseed 10.5 13.5 T
Safflower 6.5 MS 10.0–20
Sesame S 10.0–20 0.75–1.0
Sugar beet 7 5.9 T >20 4.0–6.0 >40
Sugarcane 1.7 5.9 MS 15–40
Sunflower 5.5 25 MS >20 0.75–1.0
Adapted from Ayers and Westcott (1989)
Bibliography
Ayers, R. S., & Westcott, D. W. (1989). Water quality for agriculture, FAO irrigation and drainage
paper 29 rev 1. FAO, UN, Rome.
Mass, E. V. (1996). Crop salt tolerance. In K. K. Tanji (Ed.), Agricultural salinity assessment and
management. New York: American Society of Civil Engineers.
Suarez, D. L. (2001). Sodic soil reclamation: Modelling and field study. Australian Journal of Soil
Research, 39, 1225–1246.
Chapter 23
Fertilizers
Antonio Delgado, Miguel Quemada, and Francisco J. Villalobos
Abstract In this chapter we review the classification and main features of fertil-
izers. Among nitrogen fertilizers, nitric and ammonium are the most appropriate for
topdressing and basal applications, respectively, while urea may be applied in both
modes. Phosphorus fertilizers are distinguished mainly by their solubility which
determines the application form. Potassium fertilizers are highly soluble. Deficien-
cies of micronutrients are often due to their conversion to insoluble forms in
calcareous soils or the inability of plants to mobilize and transport them; their
deficiency is usually solved with forms bound to organic compounds (complexes or
chelates).
23.1 Introduction
Fertilizers are inorganic or organic products that are used to provide nutrient for
plants. In general they have to comply with official regulations. For instance, the
European Regulation 2003/2003, sets the rules that all EU must follow in their
A. Delgado (*)
Sevilla, Spain
M. Quemada
Madrid, Av. Puerta de Hierro, n 2, 28040 Madrid, Spain
F.J. Villalobos

Agriculture, DOI 10.1007/978-3-319-46116-8_23
322 A. Delgado et al.
national directives to regulate properties, quality, and traffic of commercial

fertilizers. According to the European regulation, different types of nutrients can
be contained in fertilizers:
(a) Primary nutrients: nitrogen, phosphorus and potassium, which are usually the
nutrients that have to be supplied in large amounts (tens or hundreds of kilo-
grams per hectare).
(b) Secondary nutrients: calcium, magnesium, sodium and sulfur, also taken up by
plants in high amounts, but not always have to be applied. Sodium is not an
“essential nutrient” from a physiological point of view, but is required by some
species such as C4 plants, and it is defined as “beneficial nutrient”.
(c) Micronutrients: boron, cobalt, copper, iron, manganese, molybdenum and zinc,
are required in small amounts compared with primary and secondary nutrients;
all are “essential nutrients” except cobalt (e.g. beneficial for legumes). The
European regulation does not include nickel and chloride, which are also
essential but seldom required as fertilizer.
The fertilizer product and the application technique should be chosen to achieve
the maximum efficiency in the use of applied nutrients by plants, which implies that
the maximum fraction of applied nutrient should be taken up by plants. Fertilizers
can be applied before planting (basal or preplant) or after it (topdressing or
sidedressing). The application of fertilizer to soil can be done manually, using
machines (fertilizer spreader), or through the irrigation system (fertigation). Fertil-
izers can also be applied to vegetative organs (foliar spray), particularly when soil
conditions are not favorable for nutrient absorption (e.g. very dry) or to achieve a
fast response under deficiency conditions. Fertilizer may be applied on the entire
field or just on part of it (localized, preferable close to plants), the latter being
preferred for nutrients that can be fixed in the soil, such as phosphorus or potassium,
particularly in poor-nutrient soils or soils with a high fixing capacity.
Increased global use of fertilizers is partly responsible for the increase in food
production (Fig. 23.1). Table 23.1 shows the variation in fertilizer use for the
different continents. Fertilizer use has a decreasing trend in Europe except for N,
while it shows a clear increasing trend in Asia and America (only for N). The values
are very low and do not show any clear trend in Africa.
23 Fertilizers 323
120
FERTILIZER CONSUMPTION (M t) 100
80 N
P
60 K
40
20
0
2000 2002 2004 2006 2008 2010 2012
YEAR
Fig. 23.1 Global use of N, P and K fertilizers (2001–2011)
Table 23.1 Fertilizer consumption by continents 2003–2011 (Source: FAO stats)

America and
Europe Asia Africa Pacific
Year N P2O5 K2O N P2O5 K2O N P2O5 K2O N P2O5 K2O
M ton
2002 12.9 4.0 5.2 52.0 18.6 9.3 2.6 1.0 0.5 18.8 10.2 9.1
2003 13.5 4.2 4.7 52.6 18.9 9.6 3.0 1.0 0.5 20.3 11.7 10.3
2004 13.1 4.2 4.8 52.5 22.2 7.1 3.1 1.1 0.5 20.2 12.2 11.0
2005 12.7 4.1 4.4 54.8 23.5 12.9 3.1 1.0 0.5 19.9 11.4 9.6
2006 12.8 4.0 4.6 59.8 24.2 12.6 2.8 1.0 0.6 20.2 11.4 9.6
2007 13.7 4.3 4.9 61.9 22.0 15.5 2.8 1.1 0.6 22.0 12.7 10.5
2008 13.1 3.2 3.7 71.1 20.9 13.6 3.3 1.0 0.5 20.5 10.2 9.7
2009 12.7 3.1 3.4 77.6 25.6 11.3 2.8 1.1 0.4 19.9 8.8 7.0
2010 13.3 3.4 4.1 73.6 27.2 12.7 3.2 1.2 0.4 21.8 10.7 9.8
2011 13.6 3.5 4.1 71.4 23.2 14.9 3.3 2.2 0.4 24.1 12.2 10.9
Agricultural 470 1634 1170 1626
land 2011
(M ha)
Average 28.9 7.4 8.8 43.7 14.2 9.1 2.8 1.9 0.3 14.8 7.5 6.7
fertilizer use
(kg/ha) 2011
23.2 Classification
Fertilizers can be classified according to different criteria:

(a) Depending on their nature: Organic and inorganic (also referred to as mineral or
chemical). Inorganic fertilizers are those in which nutrients are mineral form
obtained by extraction (mining) or by industrial processes. Calcium cyanamide,
urea and its condensation and association products, and fertilizers containing
chelated or complexed micro-nutrients can be classified as inorganic fertilizers
by convention. Chelated and complexed micronutrients refer to a product in
which the micronutrient is held by complexation reaction with organic mole-
cules; depending on the type of organic molecule defined in the European
regulation the product is considered “chelated” (synthetic organic molecules
such as EDDHA) or “complexed” (natural organic molecules or present in
by-products; not always a single compound, such as lignosulphonates). In
organic farms only those from natural sources are allowed (e.g. farm organic
residues or Chilean nitrate).
(b) According to their composition:
– Straight fertilizers, which are those containing only one primary nutrient, so
they can be nitrogenous, phosphatic, or potassic.
– Compound fertilizers, which are fertilizers with a declarable content
according to law of at least two primary nutrients obtained chemically or by
blending or by a combination of both processes.
– Complex fertilizers, which are those obtained by chemical reaction, by
solution, or in its solid state by granulation with a declarable content of a
least two of the primary nutrients.
Compound and complex fertilizers can be binary, when they have a declarable
content of two primary nutrients, or ternary or complete, when they have a
declarable content of three primary nutrients.
Fertilizers with secondary nutrients or micronutrients, are those products of
one of the three previous types with declarable amounts of these nutrients.
(c) According to its physical presentation, which often determines the conditions of
its use and its effectiveness, fertilizers can be classified as:
– Solid fertilizers, with different types of presentation depending on its produc-
tion, solubility, and method of application:
Powder or non-granular, when the product is presented as fine particles
usually up to 3 mm diameter. Very few materials are sold now in this
form as they present problems in handling (they tend to “cake”) and cannot
23 Fertilizers 325
be applied with spreaders. Powder is a usual presentation for sparingly

soluble products since low particle size enhances its solubilization.
Crystalline which are usually very soluble fertilizers for preparing fertigation
or foliar spray solutions. They are not suitable for application with
mechanical spreaders.
Granules which are designed to improve the uniformity of mechanical distri-
bution. More than 90 % of the particles have to present diameters between
1 and 4 mm. The spherical shape is desirable. The distinction between
granular and prilled refers to the industrial production method.
Pelletized or pelleted: They are granular fertilizers with very uniform size of
spherical granules which improves the uniformity of distribution.
Macrogranules: granules of 1–3 cm to produce a slower release of the
nutrients.
– Fluid fertilizers, which can be fertilizers in suspension or solution or both;
fertilizers presented only in suspension as dispersed particles are called
“suspension fertilizers”, while solutions free of solid particles are “solution
fertilizers”. Pressure solutions are those including anhydrous ammonia in a
concentration greater than that that can be maintained in equilibrium with the
atmosphere.
– Gaseous fertilizers, with only one fertilizer in this category, anhydrous
ammonia which must be injected to the soil.
23.3 Fertilizer Properties
23.3.1 Physical Properties
The physical properties of fertilizers are not regulated by law. However, these
properties are critical for an accurate handling, storage, conservation, and correct
and homogeneous application to crops. For solid fertilizers the following properties
are most relevant:
(a) Hardness, i.e. the resistance to be broken, which is important to prevent the
breaking of granules during handling and to avoid powder formation due to
abrasion.
(b) Fluidity which means a low risk of caking after storage
(c) Particle size, which must be homogeneous to guarantee a correct application by
mechanical spreading
(d) Humidity, must be low to avoid caking

(e) Density, which is relevant for storage and for segregation during application of
blended fertilizers with different density; the distribution of a blend of several
fertilizers could result heterogeneous if compounds with very different density
are blended.
23.3.2 Chemical Properties
Chemical properties are important in the potential speed of action of the fertilizer
and in the potential collateral effects on crops and soil properties. The main
chemical properties to be considered in fertilizers are the following:
(a) Solubility determines the speed at which nutrients can pass to soil solution and
thus be potentially available to plants. It is usually measured in water for
nitrogenous and potassic fertilizers. For phosphate, usually less soluble, besides
water, ammonium citrate or citric acid has been traditionally used to charac-
terize its solubility trying to mimic the effect of plant roots in soil (exudation of
low molecular weight acids). Solubility in water is critical for fertilizers used in
fertigation to avoid clogging of drippers. Solubility increases with increasing
temperature and acidity of solutions. Care should be taken with mixtures which
can promote precipitation of compounds, such as fertilizers with Ca which can
promote the precipitation of Ca phosphates.
(b) Reaction of fertilizer in the soil, acid or basic, depending on what’s the fertilizer
effect on soil pH. Traditionally it has been measured by the “acidity index”
which is the equivalent amount of CaO which neutralizes the effect of fertilizer
with acid reaction or to promote the same pH rising in soil in fertilizers with
basic reaction. Fertilizer reaction can be the result of: (i) its chemical compo-
sition, e.g. anhydrous ammonia is a base, or base (e.g. Ca) which is the
counterion in the nitric fertilizers; (ii) of its reactions in soil, e.g. nitrification
of ammonium in the soil produces acidity, or the decomposition of calcic
cyanamide forms Ca(OH)2 which increases the soil pH, or (iii) presence of
impurities such as sulfuric acid in ammonium sulfate.
(c) Salt index, which measures the effect of the fertilizer on the osmotic pressure; it
is a relative value compared with sodium nitrate which receives an arbitrary
value of 100.
(d) Hygroscopicity: It is the ability to absorb atmospheric moisture and is measured
as the relative humidity value at which the fertilizer starts to absorb water. In
many cases, hygroscopicity is proportional to the solubility of the fertilizer.
Water absorption causes the dissolution of the particles, which melts the
23 Fertilizers 327
physical structure of the fertilizer and converts it to clumps instead of the initial
granules which worsens the mechanical distribution. Deliquescence is the
property of being dissolved in the water retained by hygroscopicity. This
extreme situation occurs with hygroscopic fertilizers very soluble in water,
such as many of the nitrogenous fertilizers.
23.3.3 Nutrient Concentration in Fertilizers
The nutrient concentration in a fertilizer is the amount of nutrient per unit mass of
product. After estimating crop requirement of nutrients, this information is basic to
calculate the amount of fertilizers to be applied. The concentration of nutrients in
fertilizers must be expressed in the following form, according to the legislation of
many countries (e.g. Spain):
(a) Nitrogen, as elemental nitrogen (N)
(b) Phosphorus, potassium, calcium, magnesium, sodium, and sulfur as oxides
(K2O,P2O5, CaO, MgO, Na2O, and SO3)
(c) Other nutrients, such as micronutrients, in elemental form.
However, European regulation allows European countries to choose between
oxides and elemental expression in the nutrients considered in the case (b). The
current trend is to express the concentration of all nutrients in its elemental form.
The concentration of primary nutrients of a compound or complex fertilizer or
fertilizer grade is usually indicated by three numbers separated by hyphens that
correspond to the percentages of N, P2O5 and K2O.
Example 23.1 A ternary 15-15-15 fertilizer has concentrations of 15, 15 and

15 % of N, P2O5 and K2O, respectively. If we express the concentration in
elemental form we have:
N 15 %
P 15% 62 kg P=142 kg P2 O5 ¼ 6:5%
K 15% 78 kg K=94 kg K2 O ¼ 12:45%
The factor of conversion is calculated as the ratio of the element mass to the
molecule mass.
The content of secondary nutrients or micronutrients in a complex or compound
fertilizer is expressed by another number with the percentage of the nutrient and
indication of the nutrient. For example if the ternary mentioned above has 2 % of
MgO, this should be indicated in the following way: 15-15-15-2 MgO.
The nutrient concentrations of different fertilizers are presented in Table 23.2.
Table 23.2 Most relevant mineral fertilizers and their macronutrients content
N (%) P2O5 (%) P (%) K2O (%) K (%)
Straight nitrogen fertilizers
Sodium nitrate 15.5
Calcium nitrate 16
Magnesium nitrate 10.5
Ammonium sulphate 21
Urea 46
Calcium cyanamide 16–20
Anhydrous ammonia 82
Pressured ammonia solutions 41 %
Ammonium sulfate 21
Ammonium nitrate 32
Calcium ammonium nitrate 20.5–30
Ammonium nitrosulphate 26
Nitrogen solutions 20–32
Slow release fertilizers
Urea formaldehyde (UF) 38
Isobutylidene diurea (IBDU) 32
Crotonylidene diurea (CDU) 31
Straight phosphorus fertilizers
Superphosphate 18–21 8–9
Triple superphosphate (TSP) 45 20
Phosphoric acid 54 24
Sperphosphoric acid 76 33
Dycalcium phosphate 40 17
Calcium metaphosphate 64 28
Calcined phosphate 18–28 8–12
Basic slags 15 7
Ground phosphate rock 25–40 11–17
Straight potassium fertilizers
Potassium chloride 60 50
Potassium sulphate 50 41.5
Complex fertilizers
Binary N-P
Mono-ammonium phosphate (MAP) 10–12 48–60 21–26
Di-ammonium phosphate (DAP) 18 46 20
Ammonium polyphosphates (APP) 10–11 34–37 15–16
Nitrophosphates 20 20 9
Binary P-K
Potassium phosphates 52 23 34 28
Binary N-K
Potassium nitrate 13 44 36.5
23 Fertilizers 329
23.4 Straight N Fertilizers
23.4.1 Fertilizers with Nitrate
The nitrate fertilizers are very soluble in water and the nitrate ion, a N form in which
plants readily absorbs this nutrient, is not fixed by soil particles when applied to the
soil so it remains in the soil solution. Therefore, N applied in this form is easily
absorbed by plants, but may be leached and lost from the soil. It can be considered a
fast-action N fertilizer but should therefore be applied when it can be used by the
crop to avoid losses (typically as side-dressing). These fertilizers show high hygro-
scopicity and a slightly basic reaction. Beside this reaction of nitrate fertilizers,
nitrate can have an effect of increasing pH of plant apoplast or rhizosphere due to its
absorption into cells which is coupled with Hþ which decreases the acidity of these
media.
The main fertilizers in this group are calcium nitrate (16 % N) and sodium nitrate
(Chilean nitrate) (15.5 % N). This group also includes magnesium nitrate (10.5 %
N), very soluble and used in fertigation, sometimes as solution fertilizer.
23.4.2 Fertilizers with Ammonium
Ammonium supplied with these fertilizers is a cation which is readily adsorbed by

the soil exchange complex and is therefore not leached when percolation occurs.
Thus, this type of fertilizers is recommended for basal applications in winter crops
when high risk of leaching and low extraction by crops occurs at the beginning of
the growing season. Although ammonium can be absorbed by plants, the progres-
sive nitrification of ammonium (microbial transformation to nitrate) enhances its
use by plants. At high rates, ammonium can be toxic for crops. At basic pH there is
an increased risk of losses to the atmosphere by volatilization. This risk is increased
if fertilizer is not mixed with the soil, thus making it less suitable for topdressing
applications. In hydroponics, a portion of N should be applied in ammonium form
(10–20 %) to avoid pH rising in the solution due to nitrate absorption which can
result in decreased availability of other nutrients such as Fe.
This group includes ammonium sulfate (21 % N, 24 % S), anhydrous ammonia
(gas, 82 % N) and pressured solutions of ammonia (41 %) which have to be injected
into the soil at 15–20 cm depth, with moderately wet conditions. Anhydrous
ammonia is the cheapest N fertilizer but requires special machinery for its appli-
cation and is difficult to store and handle. Another problem in the US has been the
theft of anhydrous ammonia fertilizer from storage tanks on farms for production of
the illegal drug methamphetamine.
23.4.3 Fertilizers with Nitrate and Ammonium
These fertilizers combine the advantages of both forms of N: rapid availability of

nitrate and longer availability of ammonium. They do not depend entirely on
nitrification to provide nitrate so they can be used in low temperature periods
when nitrification rate is low. They are used primarily in winter and in spring for
topdressing.
This group includes ammonium nitrate (32 % N) and ammonium nitrosulfate
(26 % N). One of the main concerns in the use of ammonium nitrate is its applica-
tion for producing explosives like ANFO which is used in mining and may be home
made by mixing ammonium nitrate (AN) with fuel in the right proportions. It has
been thus the choice for terrorists which has led to strict regulations in many
countries regarding the purchase of AN or its commercial formulation. For instance
in Ireland and Northern Ireland AN fertilizer is marketed as a mixture of ammo-
nium nitrate and calcium carbonate.
23.4.4 Urea and Related Products
Urea (46 % N) and calcium cyanamide (around 20 % N in commercial products) are

included in this group. Urea has N in ureic form and cyanamide is transformed to
urea in the soil. It has acidic reaction, and it is very soluble and hygroscopic. Urea is
a white crystalline solid that can be purchased as prills or as a granulated material.
The importance of granules is increasing as they are larger, harder, and more
resistant to moisture. Urea is rapidly hydrolyzed to ammonium through the activity
of urease enzyme, present in soils. This requires a certain temperature and humid-
ity. Urea is highly soluble so it may be leached before hydrolysis. Application to the
soil surface may involve volatilization losses of ammonia formed during hydrolysis
so it is advisable to incorporate urea by tillage or irrigation. Urea is widely used
both for basal applications and topdressing because of its low cost per unit of N
applied. It can also be applied as solutions which may also contain ammonium
nitrate. Its high solubility makes feasible its use in fertigation and foliar sprays,
which are recommended in tree orchards when soil conditions are not appropriate
for N absorption by roots (e.g. dry soil). The content of biuret (condensation
product) in urea fertilizers must be controlled since it is phytotoxic, particularly
for foliar sprays with maximum recommended contents of 0.25 %. To reduce
leaching, slow release fertilizers based on urea have been developed. These are
based on either reducing the solubility (larger granules, special coatings as paraffin
or sulfur) or by adding nitrification inhibitors.
Calcium cyanamide reactions in soil release urea and calcium hydroxide, which
explain its strong basic reaction. It is expensive, can be phytotoxic, and release
slowly available N, thus it is only recommended as basal fertilizer.
23 Fertilizers 331
23.5 Straight P Fertilizers
Phosphate fertilizers are produced by physical (grounding, calcination) or chemical

(acid attack) of phosphate rock which is a natural non-renewable resource. The
most important feature of straight phosphate fertilizers is the reduced solubility in
water of many of them. This low solubility in water does not necessarily imply that
plants cannot use them as P source. Organic acids exuded by roots, such as citrate,
contribute to the mobilization of P by solubilizing many of the precipitates of this
nutrient in the soil. This is why the solubility of P fertilizers is also measured with
ammonium citrate. The sum of the water-soluble and citrate-soluble phosphorus in
the fertilizer is considered to be the amount available to plants so it is given on the
fertilizer label. Usually, the citrate-soluble component is less than the water-soluble
component. According to solubility, three main groups of phosphate fertilizers can
be distinguished:
(a) Mostly soluble in water, including single superphosphates (18–21 % P2O5,
8–9 % P, 85 % soluble in water), triple superphosphate (45 % P2O5, 20 % P,
85 % soluble in water), phosphoric acid (54 % P2O5, 24 % P) and
superphosphoric acid (76 % P2O5, 33 % P). The two acids are only used in
fertigation (see Chap. 26)
(b) Mostly soluble in ammonium citrate, such as dicalcium phosphate (40 % P2O5,
17 % P) and calcium metaphosphate (64 % P2O5, 28 % P).
(c) Insoluble, including calcined phosphate (18–28 % P2O5, 8–12 % P), basic slags
(or Thomas slags, byproduct of iron and steel industry) (15 % P2O5, 7 % P) and
ground phosphate rock (25–40 % P2O5, 11–17 % P, from mining without
chemical treatment in the industry).
The more soluble fertilizers are to be incorporated in granular form and localized
when possible to enhance its efficiency, particularly in P-poor soils or in soils with a
high P-fixing capacity. The less soluble forms are available as powder or fine
granules which should be mixed with the soil to enhance its dissolution.
23.6 Straight K Fertilizers
Potassic fertilizers also come from mining resources, which are not so limited as
phosphate rock. Although potassic fertilizers are very soluble in water, potassium
ions are usually adsorbed to the soil exchange complex, which reduces the risk of
losses. As with phosphatic fertilizers, localization in bands of potassic fertilizers is
recommended in K-deficient soils with high cation exchange capacity to saturate it
and maintain a high availability of K in the soil solution.
The two fertilizers with only K as primary nutrient are potassium chloride (60 %
K2O, 50 % K) and potassium sulfate (50 % K2O, 41.5 % K). The former is cheaper
and more soluble but should be avoided under saline conditions to avoid negative
effects of chloride.
23.7 Compound and Complex Fertilizers
This group includes binary and ternary fertilizers. Binary are usually complex
forms, and ternary are compound fertilizers, usually obtained by mixture of straight
and complex fertilizers. Complex and compound fertilizers facilitate the simulta-
neous application of several nutrients, avoiding self-made mixtures of fertilizer by
farmers which can be less effective and adequate for a homogeneous distribution
and can have problems of compatibility between blended products. The selection of
compound or complex fertilizers must be based on the relative proportion of N, P,
and K needed by the crop, and on the price per nutrient unit applied. Blending of
fertilizers to produce compound fertilizers must consider basic rules of incompat-
ibility: avoiding mixtures of P fertilizer with products with Ca to avoid P precip-
itation, and the mixtures of ammonium fertilizers with basic reaction products to
avoid volatilization of ammonia.
(a) Binary NP fertilizers include ammonium phosphates, mono- (MAP) or
di-ammonium phosphate (DAP), ammonium poliphosphates and
nitrophosphates. DAP (18-46-0) is the most widely P fertilizer used in the
World. It is highly soluble and promotes a basic reaction around the granule
in the soil. MAP (10/12-48/60-0) is less soluble than DAP but its reaction is
acidic, which makes it a better choice in fertigation. Ammonium
polyphosphates (APP, 10/11-34/37-0) have part of the P as polyphosphates
which must be hydrolyzed by the action of enzymes in the soil to pass to the
available orthophosphate form which takes a few weeks with adequate temper-
ature and water content. APP is frequently used in fluid fertilizers among other
reasons by its acidic reaction. Nitrophosphates (20-20-0) have only a portion of
water soluble P and N in nitric and ammonium form. Ammonium phosphates
are typical fertilizers used in basal applications, particularly if no K is neces-
sary. Combination of ammonium and phosphate seems to enhance P uptake by
plants compared with other P sources.
(b) Binary PK fertilizers are mixtures of phosphates and potassium chloride or
potassium sulfate, or potassium phosphates and polyphosphates. Potassium
phosphate (0-52-34) is soluble and has a slight acid reaction. It can be used in
fertigation and foliar sprays. It is more expensive than other binary fertilizers
with P.
(c) Binary NK fertilizers include blends of straights N and K products and only one
complex fertilizer, potassium nitrate (13-0-44). This is a very soluble product
23 Fertilizers 333
recommended for fertigation and foliar sprays. It can be applied as topdressing

if additional K (beside the basal application) is required.
(d) Ternary NPK fertilizers are solid or liquid mixtures of straight and compound
fertilizers with a wide range in grades and presentations. They are used for basal
broadcast applications. Although the amount of N applied as ternary products is
really modest, they can represent a large fraction of the total consumption of P
and K applied to the soil. Depending on each particular product, the nutrients
can be present in different chemical forms.
23.8 Fertilizers and Products with Secondary Nutrients
Although crops can take up large amounts of secondary nutrients, its application is
not frequent because available pools in the soil can cover plants extractions. Its
application usually follows a “sufficiency” strategy, which means that nutrient is
applied only if an increased yield can be expected from its application.
Calcium can be extracted in high amounts by crops, its concentration in leaves
being sometimes higher than that of N (e.g. in citrus). The need to apply Ca as a
fertilizer is rare; its deficiency is typical in acidic soils with low base saturation of
the exchange complex. Seldom, antagonistic problems with Mg make its applica-
tion advisable. Fertilizers with significant amounts of Ca are:
(a) Nitrogenous fertilizers: calcium ammonium nitrate (10–20 % CaO, 7–14 % Ca),
calcium cyanamide (54 % CaO, 39 % Ca), and calcium nitrate (28 % CaO, 20 %
Ca)
(b) Phosphate fertilizers: superphosphate (17–28 % CaO, 12–20 % Ca, mostly
present as gypsum), slags (45–50 % CaO, 32–36 % Ca), and dicalcium phos-
phate (32 % CaO, 23 % Ca)
Ca is usually added in the amendments used for the reclamation of sodic or acid
soils, which implies a nutrient supply that can overcome Ca deficiency in crops. As
amendments to correct soil acidity, products containing Ca and/or Mg carbonate or
Ca oxide or hydroxide are used. The most common products are: limestone
(45–55 % CaO, 32–39 % Ca), lime (100 % CaO, 71 % Ca), and dolomite (30 %
CaO, 21 % Ca). These products are efficient in increasing the base (mainly Ca)
saturation of the soil and soil pH.
For the correction of sodic soils gypsum (33 % CaO, 24 % Ca) or phospho-
gypsum (a byproduct of the P fertilizer industry which is mostly gypsum) are the
best choice in efficiency and price.
Magnesium is required in lower amounts than Ca. Its deficiency is frequently
due to an antagonism with Ca, and sometimes with K when potassic fertilizers are
applied in high amounts, particularly in K-rich soils. Its concentration is low in
most fertilizers. When needed it may be added as dolomite (20 % MgO, 12 % Mg),
magnesium oxide (90 % MgO, 54 % Mg), magnesium chelates (foliar application),
and magnesium sulfate (16 % MgO, 10 % Mg). The latter can be applied by foliar
sprays.
Sulphur extraction by crops can be as high as that of P, being particularly high in
legumes (e.g. more than 45 kg/ha in alfalfa) and cruciferous crops. It is present in
many fertilizers, such as ammonium sulfate (24 % S), ammonium nitrosulfate (12 %
S), and superphosphates (12 % S in single superphosphate), which has been a
traditional source of S for crops. However, the decreasing trend in the use of
ammonium sulfate, superphosphates, and elemental S as fungicide could lead to S
deficiencies in the next future. The need for adding S is not common, but if
necessary it can be applied as sulfuric acid (30 % S), elemental sulfur (30–99 %
S), potassium sulfate (17 % S) and urea-sulfur (19 % S).
23.9 Fertilizers and Products with Micronutrients
Micronutrients are usually applied following a “sufficiency strategy”, which means

that their application is done if a deficiency is expected. Frequently, micronutrients
deficiency is the consequence of soil conditions promoting a failure in the mobili-
zation, absorption or transport mechanisms of plants, not the result of a lack of
nutrient in soil. The paradigmatic case is the iron deficiency chlorosis, related to
alkaline and calcareous conditions, not to the lack of iron in soil. The most common
fertilizers in the market do not contain significant amounts of trace elements with
the exception of Chilean nitrate and slags. Micronutrients are not added to other
fertilizers because of the risk of toxicity. The deficiencies, when detected, are
treated with specific products.
Iron deficiency is called “iron deficiency chlorosis” and is related typically to
calcareous soil and sensitive plants. The application of inorganic Fe salts (sulfates
and carbonates) to the soil is usually not effective due to the rapid oxidation of Fe in
the soil which results in the precipitation of insoluble Fe(III) oxides. These products
are more effective by foliar sprays or injections to the trunk. The only inorganic salt
effective in overcoming the problem (during several years) is vivianite (ferrous
phosphate). Siderite (Fe carbonate) with colloidal size can also be effective. The
easiest to use and most effective products are Fe-chelates, although they have the
problems of high price and low residual effect (3-4 applications per growing season
are usually needed). Fe-chelates are Fe complexed by organic compounds (usually
synthetic amino carboxylic acids) which provide a supply of Fe that is maintained
available in the soil, and also a positive effect on Fe transport mechanism through
plasmamembranes. Accurate selection of Fe-chelate is necessary depending on soil
conditions, e.g. chelates applied to calcareous soils may be stable in condition of
high Ca concentration and pH in the soil solution. In calcareous soils, the most used
Fe-chelate is EDDHA-Fe. Fe-chelates can be applied directly to the soil or by
fertigation; care should be taken with foliar sprays since the chelates are not always
photo-stable. Other type of Fe-complexes can be obtained using natural organic
23 Fertilizers 335
matter as complexing substances (e.g. lignosulphonates); in this case, the definition

of the product is “Fe-complexes”.
The deficiency of zinc is, beside that of Fe, mostly contributing to decrease
agricultural yields in calcareous soils. The stability of Zn-chelates is high but in a
very narrow pH range (typically 6.5–7.5 for many commercial products). Further-
more, Zn is strongly bound to complexing agents when low free Zn is present in the
soil, limiting its use by plants. This is the reason why joint applications of chelates
and inorganic salts (sulfates, nitrates, chlorides, oxides) are usually done. The
application of inorganic sources alone could be more effective that in the case of
Fe. Lignosulphonates of Zn are also commercially available as Zn source for crops.
Manganese and copper can be applied if necessary as inorganic products (chlo-
rides, sulfates, nitrates, or oxides, which however are less soluble) or chelates.
There are not specific chelating compounds for Cu or Mn, which explains why its
application as chelates is not always successful.
The deficiency of molybdenum is usual in strongly acid soils, where its solubility
is significantly decreased. If needed it is applied as inorganic salts such as ammo-
nium or sodium molybdate. Amendments to increase soil pH in acidic soils can
contribute to an increased availability of this nutrient.
The deficiency of boron occurs often in the most demanding crops (alfalfa, beet,
cauliflower, sunflower, olive, etc.). Sandy soils poor in organic matter and soils with
very high pH can promote B deficiency. The main products are Borax (11 % B) and
sodium borate in foliar or soil application. B bound to etanolamine or
trietanolamine can be also used.
23.10 Slow Release Fertilizers
Slow release fertilizers were firstly focused on N supply since the progressive
solubilization of N decreases the risk of leaching, particularly in sandy soils.
Recently, there are available slow release products which supply different combi-
nations of nutrients, including also those considered non-mobile in the soil such as
P and K. Nowadays, a new line of products, particularly phosphatic fertilizers based
on organometallic compounds, whose solubilisation is enhanced under rhyzosphere
conditions (e.g. increased organic acid concentration) is being developed thinking
in a progressive and highly efficient use of nutrients.
Slow N release fertilizers can be classified into different categories:
(a) Natural organic sources such as manures, which contain part of the N in organic
form which must be mineralized to be used by plants. This category will be
studied in Sect. 23.11.
(b) Products formulated from urea that have been chemically reacted, slowing
down the urea release into the soil solution since these compounds must be
hydrolyzed; the release speed depends on environmental conditions such as
temperature and humidity and also on the microbial activity in the soil because
this activity breaks the link between urea and other organic molecules. The
main groups in this category are: urea-formaldehyde reaction products (UF,
commercial products can have 38 % N and maximum decomposition speed at
pH 6.1–6.5), isobutylidene diurea (IBDU, typically 31 % N, maximum decom-
position speed at pH 4), crotonylidene diurea (CDU, usually 31 % N, maximum
decomposition speed at basic pH), and triazone (cyclic compounds with
ammonium).
(c) Products with slow release of N achieved by physical coating of urea prills.
Coatings are usually composed of sulfur, wax or resins, which form a semiper-
meable membrane which allows a slow dissolution of covered fertilizers.
Release is enhanced with the progressive decomposition of membranes. Some
commercial products have different contents of primary and secondary ele-
ments and micronutrients covered plastic polymers with pores which allow a
slow release. In some cases, this type of products is commercialized with a wide
range of release times.
(d) Some authors also consider slow release N fertilizers those based on urea or
ammonia mixed with inhibitors of the nitrification such as nitrapyrin. This
allows ammonium to remain retained in the exchange complex during several
weeks, decreasing the risk of losses through leaching or denitrification.
23.11 Organic Fertilizers
Organic fertilizers are products consisting of animal or plant material or obtained

by transformation (e.g. composting) of these type of material which contains
enough plant nutrients to be useful as fertilizer. Nutrients in the organic fertilizer,
such as N, P and S, can be at least partly in organic form. To be available for plants
they should be previously mineralized by the action of microorganisms. Other
nutrients, such as K, are in cationic form and can be considered readily available
in these products.
If these products contain significant amounts of organic matter, they can be used
as organic matter source for soils, thus contributing to improve soil structure, or can
increase biological activity in soils. Then, they can be also considered as organic
amendments.
The most commonly used organic fertilizers can be classified into the following
three groups:
– Animal wastes: slurry, dung, farmyard manure and poultry litter.
– Within farm: crop residues, pruning residues, green manure and garbage.
– External: peat, compost from different origin (e.g. organic urban residues),
byproducts or wastes of food industry (blood, bones, etc.).
It should be noted here that only plant residues generate humus. Thus, only this
type of residue contributes to increasing the soil organic matter content and is
advisable as organic amendment for soils. On the other side, animal residues only
23 Fertilizers 337
contribute with nutrients that can be used by the crop or by microorganisms.

Manures are composed of animal excreta mixed with straw “beds” of livestock,
which contribute to generate soil humus. The yield in humus increases with
increasing C to N ratio in the residue; e.g. fresh manures (mixed with livestock
beds) or plants residues rich in lignin such as cereal straw or sawdust. On the other
hand, fresh plant residues with low C to N ratio do not provide much humus but are
efficient in increasing the microbiological activity in the soil which is important in
many relevant processes affecting the good functioning of soil such as those
involved in the N cycle.
The direct application of animal waste or its direct deposition in the field by the
animals has several drawbacks:
– High water content and low nutrient concentration which increases the applica-
tion cost per unit of nutrient applied when compared with mineral fertilizers
(with high nutrient concentration)
– Uneven distribution
– High losses of N by volatilization of ammonia, particularly if they are not
incorporated.
– Bad smells and potential chemical and microbial contamination of water courses
if the product is eroded, e.g. after an intense rainfall following application
– Addition of weed seeds, pathogenic microorganisms and insect larvae
(e.g. flies).
– Fermentation in the field of fresh organic residues can reduce seed germination
and seedling growth due to the production of phytotoxic compounds or to
decreased oxygen partial pressure around the seedlings.
– Some residues with high C to N ratio, such as fresh manure or cereal straw, can
promote an initial N immobilization which may decrease N availability to plants.
Later on, after humification of residues, N remaining in them increases the total
N content of the soil.
Many of these problems are greatly reduced if the residues are subjected to
composting (aerobic decomposition in the temperature range of 40–65 C).
Composting implies a decrease in C to N ratio of the residue (typically, in the
case of manure from >50 to <20) and an increase in the density. To avoid problems
it is also advisable to apply some time before sowing (2–3 months at least in the
case of fresh manures or residues). It is also advisable to incorporate the residues or
compost into the soil and avoid their application along with basic reaction products
such as lime as that could promote ammonia volatilization.
The main limitations in the use of organic fertilizers are the following:
– Nutrient concentration in organic fertilizers is highly variable depending on the
nature and processing of the product. In the case of manures the composition is
affected by animal species (Table 23.3), age, proportion of bed, diet and
composting time. Usually the ranges in N, P and K concentrations in manure
are 0.3–0.8, 0.07–0.13 and 0.33–0.58 kg/t, respectively.
Table 23.3 Total manure produced per year and per animal for several species and average
macronutrient concentration on fresh weight basis (compiled from various sources)
Manure (fresh) N P K Total N Total P Total K
kg/animal/year % kg/animal/year
Dairy Cow 17,883 0.48 0.11 0.40 85.8 19.5 71.2
Heifer 10,367 0.53 0.17 0.51 54.9 17.7 53.3
Beef Cow 9,925 0.6 0.16 0.33 59.6 16.0 32.9
Feeder 8,275 0.55 0.19 0.45 45.5 15.5 37.1
Stocker 2,867 0.51 0.14 0.41 14.6 3.9 11.9
Swine Finishing 2,350 0.76 0.31 0.46 17.9 7.4 10.7
Growing 3,317 0.55 0.18 0.35 18.2 5.9 11.6
Nursery 495 0.55 0.20 0.32 2.7 1.0 1.6
Gestating sow 2,110 0.78 0.37 0.43 16.5 7.7 9.1
Sow and litter 4,963 0.45 0.20 0.32 22.3 9.8 15.6
Poultry Layer 46.2 1.32 0.47 0.53 0.6 0.2 0.2
Broiler 24.4 1.98 0.66 0.87 0.5 0.2 0.2
Turkey 112 1.89 0.78 0.91 2.1 0.9 1.0
Duck 50 1.01 0.45 0.56 0.5 0.2 0.3
Goose 100 1.1 0.26 0.41 1.1 0.3 0.4
Other Horse 8,600 0.57 0.11 0.45 49.0 9.4 38.5
Sheep 610 0.94 0.17 0.65 5.7 1.0 3.9
Goat 1,100 0.99 0.24 0.89 10.9 2.6 9.8
Rabbit 56 1.56 0.53 0.71 0.9 0.3 0.4
In poultry litter, nutrient concentrations are usually larger, with N and P con-
centrations above 2 %. Thus, it is difficult to know the amounts of nutrient
applied with a given rate of organic fertilizer if specific analysis of each batch
applied is not performed.
– Organic fertilizers have low nutrients concentration thus making difficult to
meet crop demand. This forces to apply high rates of organic fertilizers or to
an additional supply of mineral fertilizers. For example, the application of 20 t/
ha of cow manure may provide 60 kg N/ha, 30 kg P/ha and 80 kg K/ha.
– Part of the nutrients in these products are in organic form, particularly N and
P. Thus their release is not immediate, since it requires the mineralization of
organic matter which may take several years to be completed. It is assumed that
N in manures is released in 3–5 years; in slurries, a greater portion of N is readily
available in the first season after application (60–70 %). This slow release of
nutrients presents the advantage that the mobile elements as nitrogen are
retained by the soil, so leaching losses are reduced. Beside this, the application
of P as organic forms or with organic matrix is more efficient in increasing
available pool in the soil than mineral fertilizers. On the other hand, the slow
release of nutrients through mineralization implies that the whole amount of
applied nutrient is not readily available to plants, thus they should be
complemented with mineral fertilizers.
23 Fertilizers 339
– The nutrient equilibrium in organic fertilizers does not match the equilibrium
required by crops. For instance, cereals usually have a N:P:K requirement
around 7:1:4.8, so meeting the crop N requirement with the cow manure
described above implies an excess application of P. Thus, for an adequate
nutrient supply to the crop a combination of organic and mineral fertilizers is
required.
Animal wastes and plant residues or green manure with legumes produced
within farms must be integrated in a nutrient balance at the farm scale for a
sustainable fertilizer management which must consider that external mineral or
organic fertilizers must complete the availability of nutrient in farm soils plus the
potential internal supply with residues or green manure. In organic or ecological
farming systems only organic or natural fertilizers are allowed. This often leads to
reduced yields which can partially be explained by the limited supply of nutrients,
although this may be compensated by the higher prices of organic crops.
Bibliography
European Commission. (2003). Regulation 2003/2003 relating to fertilizers.

Guertal, E. A. (2009). Slow nitrogen release fertilizers: A review. Hortechnology, 19, 16–18.
International Plant Nutrition Institute. Nutrient source specific. http://www.ipni.net/specifics-en.
Accessed 20 Nov 2013.
Lampkin, N. (1990). Organic farming. Ipswich: Farming Press.
Lucena, J. J. (2006). Synthetic iron chelates to correct iron deficiency in plants. In L. L. Barton &
J. Abadı́a (Eds.), Iron nutrition in plants and rhizospheric microorganisms (pp. 103–128).
Dordrecht: Academic Publishers, Springer.
Mengel, D. B., & Rehm, G. W. (2012). Fundamentals of fertilizer application. In P. M. Huang,
Y. Li, & M. E. Sumner (Eds.), Handbook of soil sciences, resource management and environ-
mental impacts (2nd ed., pp. 14-1–14-15). Boca Ratón: CRC Press.
Chapter 24
Nitrogen Fertilization I: The Nitrogen
Balance

Abstract Nitrogen is the most important nutrient in agricultural production. The

natural input of N is due to N fixation, especially by Rhizobium bacteria that infect
the roots of legumes. Organic N becomes inorganic through mineralization, and
then inorganic N is absorbed by plants. Often soil microorganisms “capture”
temporarily inorganic N when residue with high C/N ratio decompose (immobili-
zation process). The ammonium in the soil is converted to NO3 through nitrifica-
tion which is greatly reduced in waterlogged soils. In the latter denitrification
generates gaseous N forms that are lost (oxides of N and N2). Major losses of N
may occur by nitrate leaching which is proportional to deep percolation and to
nitrate concentration in the soil solution. A Leaching Index may be calculated as a
function of rainfall and soil type to quantify the risk of leaching.
M. Quemada (*)
A. Delgado
Sevilla, Spain
L. Mateos
Instituto de Agricultura Sostenible (CSIC) & Universidad de Cordoba, Alameda del Obispo
s/n, 14004 Cordoba, Spain
F.J. Villalobos

Agriculture, DOI 10.1007/978-3-319-46116-8_24
342 M. Quemada et al.
24.1 Introduction
The original source of nitrogen for terrestrial plants is the N2 gas which constitutes
78 % of the atmosphere. As plants cannot convert N2 to protein, first it has to be
transformed following one of the following paths:
1. Fixation by microorganisms living in symbiosis with the roots of legumes.
2. Fixation by free living soil microorganisms.
3. Fixation as oxides by electrical discharges in the atmosphere.
4. Fixation as NH3, NO3 or CN22 by N fertilizer manufacturers.
The contribution of atmospheric N2 is in dynamic equilibrium with the forms
fixed in the soil. While N2 is fixed according to various processes, other chemical
and microbiological processes release N2 to the atmosphere (Fig. 24.1). Except for
industrial fixation or combustion, all other processes are natural, but can be altered
by soil and crop management.
Understanding the N cycle in the soil-crop system is the key to optimizing
nitrogen fertilizer management, maximizing yields and minimizing negative envi-
ronmental impacts on water (nitrate pollution) and atmosphere (emission of green-
house effect gases). The sources of N for crops are inorganic and organic N
fertilizers and symbiotic N2 fixation.
Although N2 in the atmosphere can be considered an infinite source of N for
fertilizer production, this industrial process requires huge amounts of energy. Thus,
N for industrial production of mineral fertilizers cannot be considered a renewable
resource since it depends on a very high consumption of energy which is mostly
obtained from non-renewable resources.
N2 atmosphere
N2O, N2
N fertilizers N fixation
N plant
NH4+ N – O.M.
Mineralization Denitrification
Nitrification NO2–
NO3
Leaching
Fig. 24.1 Nitrogen cycle

24 Nitrogen Fertilization I: The Nitrogen Balance 343
24.2 N Forms in the Soil
The soil N concentration ranges from 0.02 % (subsoil) to 2.5 % (peat) with a typical
range 0.03–0.4 %. This N can be inorganic or organic, with the latter being the
predominant form.
Organic N appears as proteins, amino acids, amino sugars and other N com-
pounds. Inorganic forms include ammonium (NH4+), nitrite (NO2), nitrate (NO3),
nitrous oxide (N2O), nitric oxide (NO) and N2. The first three are important from the
fertility point of view and are derived from fertilizers or come from the organic matter
mineralization. The other three forms are gases that are lost as a result of the
denitrification.
24.3 N Forms Absorbed by Plants
Plants absorb NH4+ and NO3 although often the presence of both improves plant
nutrition. The nitrate concentration is generally higher than that of ammonium and
NO3 in the soil solution, so it reaches the roots with the water flow (mass transport
flow). Plant preference for one or another form of inorganic N depends on the
species, plant age, environmental conditions and other factors. For instance, cereals
and beets absorb either NO3 or NH4+. The Solanaceae (potato, tobacco, tomato),
benefit from a high NO3/ NH4+ ratio in the soil solution. Species adapted to acid
soils are used to low NO3/NH4+ ratio, as NH4+ tends to accumulate due to
nitrification slowdown in acid environments.
In terms of energy, NO3 uptake is less efficient than that of NH4+, as the nitrate
has to be reduced to ammonium before the N becomes part of the organic com-
pounds. However, NH4+ absorption leads to acidification of the rhizosphere and
decreases the absorption of Ca2+, Mg2+ and K+ while it increases the absorption of
H2PO4, SO42 and Cl. On the other hand, NO3 uptake is co-transported with H+
thus contributing to rizhosphere and apoplast alkalinization which can decrease Fe
uptake by plants. Small amounts of organic N are absorbed by plants mainly in the
form of amino acids, however, evidence that organic N contributes significantly to
plant N nutrition is still lacking.
24.4 Symbiotic N2 Fixation
Symbiotic N fixation involves the reduction of atmospheric N2 to NH3 by an

enzyme (nitrogenase) in aerobic microorganisms (mainly Rhizobium bacteria)
that form nodules on the legumes roots. For centuries this was the major N source
in agriculture, but the increasing availability and better prices of synthetic N
fertilizer reduced the importance of this source.
The Rhizobium-legume symbiosis is characterized by its specificity, i.e. specific

Rhizobium species will only infect a specific type of legume. Therefore, it is often
necessary to inoculate the seeds with the adequate species or strain of Rhizobium.
The mere presence of nodules on the roots does not imply fixation activity. For
example, in alfalfa active nodules are enlarged (2–4 4–8 mm) and are grouped in
the primary roots. The red color inside the nodules denotes the presence of
leghemoglobin, a N and O carrier required for the activity of the Rhizobium.
The factors that affect the rate of N fixation by Rhizobium are pH, concentration
of nutrients in the soil, photosynthetic activity, climate and overall crop manage-
ment. Soil acidity restricts the presence and activity of Rhizobium, although major
differences in the sensitivity to acidity of different species and even races, of
Rhizobium exist. For example, a pH below 6 drastically reduces the number of
nodules of Rhizobium meliloti in alfalfa roots while pH between 5 and 7 hardly
affects nodulation of R. trifoli in clover.
An excess of NO3 in the soil reduces the nitrogenase activity and thus N
fixation. The maximum fixation occurs when there is little inorganic N available
in the soil. However, small doses of N fertilizer are often recommended to ensure
good seedling establishment of legumes while the Rhizobium nodulation is com-
pleted. Applications of N may also be necessary at the beginning of spring, when
the demand for N by the plant exceeds the supply by Rhizobium due, for example, to
low temperatures. In some legumes (i.e. beans), fixation is so poor that it is
necessary to apply N fertilizer systematically.
In general, a high photosynthetic activity is related to high N2 fixation, and
therefore water stress, low temperature or any other stress that reduces photosyn-
thetic activity will also decrease N2 symbiotic fixation.
24.5 Quantifying N2 Fixation
Perennial crops fix between 110 and 225 kg N/ha/year, although the values may be
above or below that range depending on environmental conditions. Annual legumes
fix between 50 and 110 kg N/ha/year.
As a first approximation, the amount of N fixed by a legume crop, can be
estimated as:

1 HI
N fixed ¼ ð1 þ f NR Þ Y NCh þ NCr FNBF ð24:1Þ
HI
where fNR is the ratio of N in roots and N in shoots, Y is yield, NCh and NCr are the
N concentrations in the harvested product and the residues, respectively, HI is
harvest index and FNBF is the fraction of N resulting from biological fixation. The
value of fNR lies between 0.05 and 0.25. FNBF depends greatly on the availability of
soil N, which in turn is related to fertilizer application rate and the type of legume.
Table 24.1 Nitrogen concentration in different crop species

DM N N N DM N N N
Crop species % min max typical % min max typical
Alfalfa (hay) 85.0 2.80 3.80 3.30
Apple 18 0.25 0.45 0.35
Barley 88.5 1.50 1.80 1.60 Straw 90 0.58 0.88 0.70
Bean (Phaseolus) 89 3.50 4.50 4.00 Straw 89 1.10 1.40 1.20
(dry seed)
Cotton 91 2.32 2.75 2.53 Residues 92.5 0.90 1.00 0.98
Grapes (wine) 19 0.50 0.60 0.57
Lettuce 6 4.00 4.40 4.27
Maize (grain) 30 1.10 1.45 1.25
Millet 89.5 2.00 Stover 91.5 0.80
Olives (60 % can- 50 0.20 0.40 0.30 Vegetative 70 1.00 2.00 1.50
opy cover)
Orange 18 1.00 1.40 1.20
Palm trees 79 1.25
Peach 12 0.80 1.20 1.00
Peas (dry harv.) 90 4.00 4.30 4.20 Straw 88.5 1.20 1.40 1.30
Potato 23.5 1.20 1.90 1.60 Residues 51 2.00 2.40 2.20
Rapeseed, Canola 91 3.40 4.30 3.90 Residues 82.5 0.55 0.90 0.80
Rice 94 1.33
Sorghum (grain) 87.5 1.45 2.00 1.90 Stover 92 0.60 0.80 0.70
Soybeans 87.5 6.10 6.90 6.50 Residues 89 1.00 1.00 0.85
Sugar beet 21 0.90 1.10 1.05 Residues 18 1.80 2.80 2.30
Sugar cane (virgin) 25 0.13 26 0.41
Sunflower 91.5 2.20 3.20 2.95 Residues 87 0.40 1.10 0.80
Tomato 6 2.30 3.10 2.60 20 1.80
Winter wheat 87.5 1.85 2.30 2.10 Straw 90.5 0.40 0.85 0.65
Maximum and minimum values are shown when available. Also the dry matter content (% over
fresh mass) is indicated
When soil N availability is low, most of the crop N comes from fixation
(Table 24.2). If the organic matter content is high the lower values of the proposed
intervals should be used. On the other hand the N concentrations in the harvested
product and the residues may be measured or taken from Table 24.1.
Example 24.1 The expected yield of an alfalfa crop is 8 t/ha (15 % moisture)
on a soil with 1 % organic matter. Initial inorganic N in the soil is 40 kg/ha
and expected N mineralization is 35 kg N/ha. We assume that HI is 0.9 and
that residues have the same N concentration as the harvested part.
(continued)

In Table 24.1 we find that N concentration of alfalfa is 3.3 kg N/100 kg dry
matter. As water content is 15 %, harvested dry matter biomass is 6800 kg dry
matter/ha.
Available N in the soil will be the sum of initial inorganic N and expected
mineralized N:
40 kg N=ha þ 35 kg N=ha ¼ 75 kg N=ha
So we are in the 55–110 kg/ha interval of Table 24.2, implying that 60–90 %
of N comes from fixation. As organic matter content is low (1 %) we use the
upper limit (0.90). We also take fNR ¼ 0.2, which means that total fixed N is:

1 HI
N fixed ¼ ð1 þ f NR ÞY NCh þ NCr FNBF
HI
0:1
¼ ð1 þ 0:2Þ6800 0:033 þ 0:033 0:9 ¼ 270 kgN=ha
0:9
Table 24.2 Percent of crop N obtained from symbiotic fixation in legumes as a function of
legume type, % organic matter and inorganic N in the soil
Available inorganic N (kg/ha)
% OM Type 55 55–110 110–225 >225
>3 Annuals 70 50 30 5
Perennials 80 60 50 10
<3 Annuals 95 80 60 40
Perennials 95 90 80 50
Adapted from Meisinger and Randall (1991)
24.6 Transformations of N in the Soil
24.6.1 Mineralization and Immobilizacion
Dead plant materials (senesced leaves, residues left after harvest) suffer a process
called decomposition, which is the breaking down of the structure into
unrecognizable organic matter. This process is performed by bacteria and fungi
which get energy from the respiration of carbon compounds of the residue. Decom-
position rate increases with temperature up to 32–35 C and with water content up
to Field Capacity, so faster decomposition is expected if the residue is buried into
the soil. In general decomposition rate is proportional to N concentration in the
N fixation
Mineralization
Nitrification
RELATIVE VALUE 1
Denitrification
0
SOIL WATER CONTENT
Fig. 24.2 Response of N fixation, N mineralization and nitrification to soil water content
Fig. 24.3 Response of N

mineralization and Mineralization
nitrification to soil 1
Nitrification
temperature
RELATIVE VALUE
0
0 20 40
TEMPERATURE
residue, which explains the faster breakdown of legume residues as compared to

those of cereals.
Nitrogen mineralization is the conversion of organic N to NH4+. After decom-
position of plant residues, N mineralization occurs in two stages, aminization
(breaking up of proteins to amino acids, amines and urea, with release of CO2)
and ammonization (conversion of amines and amino acids to NH4+). This transfor-
mation is performed by heterotrophic microorganisms (fungi and bacteria), and is
based on aerobic and to a lesser extent, anaerobic respiration.
Mineralization is favored by high soil water content, without reaching saturation
to ensure oxygen supply (Fig. 24.2). The decomposition does occur in waterlogged
conditions but at a lower rate. As most biological processes, mineralization is
affected by temperature. The temperature coefficient, Q10 for mineralization is
2 in the range of 5–35 C, i.e. mineralization rate is doubled by raising the
temperature 10 C. The optimum temperature is around 35 C (Fig. 24.3).
Immobilization is the conversion of inorganic N (NH4+ and NO3) to organic N
being basically the reverse of mineralization. If decaying organic matter contains
little N relative to C, the microorganisms use (immobilize) soil mineral

N. Microorganisms require a C/N ratio of about 8:1, therefore the soil inorganic
N may decrease rapidly during waste decomposition and the crop may experience N
deficiency. When the residue with low N content is finally decomposed, C avail-
ability as energy source for microbes is decreased and microbial activity decreases,
thus finishing N immobilization.
The predominant process (mineralization or immobilization) depends on the
C/N ratio of decomposing organic matter. At the start of the decomposition of
organic residues, there is a rapidly growing population of heterotrophic microor-
ganisms which is detected in the increased release of CO2. If the C/N is greater than
30/1, immobilization occurs. As decomposition proceeds, the C source decreases,
and so does the C/N ratio, until the microorganisms begin to die. Finally a new
equilibrium is reached that starts with mineralization of N and ends with a higher
inorganic N level and C/N ratio of around 10/1. The time required depends on the
amount of added organic residue, the availability of inorganic N, the resistance of
the residue to be decomposed (i.e. its lignin content), the temperature and soil water
content.
Example 24.2 After harvest of a cereal we incorporate 3000 kg/ha of straw

with 45 % C and 0.75 % N (C/N ¼ 60/1) to the soil. The total amounts of C
and N are:
3000 kg=ha 0:45 ¼ 1350kg C=ha

3000kg=ha 0:0075 ¼ 22:5kg N=ha
We assume that 35 % of C will be used in growth of microorganisms while

65 % of C is lost as respired CO2. The amount of C accumulated in the
microbial biomass will be:
1350 kg C=ha 0:35 ¼ 472:5kg C=ha
The C/N ratio of the microorganisms is 8/1 so N accumulated will be:
472:5kg=ha=8 ¼ 59 kg N=ha
And the amount of immobilized N is:
59kg N=ha 22:5 kg N=ha ¼ 36:5 kg N=ha
The C/N in the surface layer of a natural soil is between 8 and 12, being 10 the
most common value. These soils have a relatively stable microorganism population
and deposition of organic residues (and thus mineralization) is also constant. If this
soil is cultivated a rapid increase in decomposition and mineralization will occur
which will decrease the organic matter content.
Example 24.3 A soil has 2.0 % organic matter in its surface layer (0.20 m)
and bulk density of 1.3 t/m3. The N concentration in organic matter is 5 %.
Therefore the total amount of organic N in this layer is:
1:3103 kg soil=m3 soil 0:20m 104 m2 =ha 0:02kg O:M:=kg soil

0:05kg N=kg O:M:
¼ 2600kg organicN=ha
If mineralization rate is 1 %/year, the amount of inorganic N released will be:
2600kg organic N=ha 0:01 kgN=kgorganic N=year ¼ 26kg N=ha=year
24.6.2 Nitrification
The NH4+ transformation to NO3, called nitrification, is performed by bacteria

(Nitrosomonas and Nitrobacter) in two stages:

NHþ
4 þ O2 > NO2 þ H2 O þ H
þ
ðNitrosomonasÞ

NO2 þ O2 > NO3 ðNitrobacter Þ
Both Nitrosomonas and Nitrobacter are autotrophic bacteria, although in both
processes some heterotrophic organisms are also involved. The second stage is
faster than the first, which prevents the accumulation of NO2, which is toxic to
plants.
Among the factors affecting the nitrification rate, the first is the substrate
concentration (NH4+) that depends on fertilization and mineralization. The need
for oxygen in the reactions indicated above, means that a good aeration is required
(optimum oxygen concentration is 20 %), so waterlogging is undesirable. However,
nitrification is high with relatively high water content, and is maximized when
80–90 % of the soil pores are filled with water. The optimum temperature for
nitrification is between 25 and 35 C. It is generally accepted that nitrification is
optimal at neutral to slightly alkaline soil pH, but nitrification can occur in the range
of 4.5–10 (Fig. 24.4). A secondary effect of lime application to acid soils is
enhancing nitrification and mineralization contributing to an increase in N supply
to crops.
The product of nitrification (NO3) is very soluble in water and is hardly
adsorbed by soil colloids so it may be lost by leaching.
Nitrogen fertilizer management has to take into account the facts stated above. In
regions with low soil temperatures and/or low winter rainfall, NH4+ applications in
the fall, before planting, save time and money. If air temperatures are below 4–5 C
or mean soil temperature is below 10 C, the preplant applications of ammonium in
the autumn are efficient since nitrification rates are low.
Fig. 24.4 Response of N

fixation, nitrification and N fixation
1
denitrification to soil pH Denitrification
RELATIVE VALUE
Nitrification
0
3 5 7 9 11
pH
Nitrosomonas activity is very sensitive to the presence of a large number of

compounds. Because of that, the fertilizer industry has developed substances called
‘nitrification inhibitors’ that are blended to the fertilizer granules or added to
manures and slurries. The inhibitors slow down the nitrification process, controlling
nitrate accumulation in the soil and thus nitrate losses. These compounds need to be
biodegradable and in many countries they can be only commercialized when
blended to the fertilizers.
24.6.3 Denitrification
Denitrification is not the opposite of nitrification but the reduction of nitrate into
volatile N compounds. When soil oxygen availability is reduced because of high
water content, soil compaction or the application of easily decomposable organic
matter, the rate of denitrification increases. Anaerobic micro-zones containing still
a source of labile C appear and a broad number of microorganisms (mainly bacteria
as Pseudomonas, Bacillus and Paracoccus, but also some fungi) are able to use
NO3 or NO2 as oxidizing agents releasing gaseous N forms to the atmosphere:
NO3 > NO2 > N2 O > NO > N2
The incomplete reduction promotes the emission of N2O, a very reactive gas that
enhance the ozone destruction and the atmosphere warming capacity. The reaction
is very fast and peaks of N2O emission are observed after application of organic or
synthetic fertilizers. It was previously thought that denitrification required water
logging conditions, but nowadays it is recognized as a main pathway of N losses to
the atmosphere under a broad range of environmental conditions.
Among the factors affecting denitrification, the soil water content is one of the
most important. Water logging prevents the diffusion of oxygen and thus enhances
denitrification (Table 24.3). Because of that, the highest denitrification N losses
from agricultural fields have being reported in rice paddy fields, were potential
losses of up to 16 kg N/ ha in the day after soil saturation have been found. A
strategy to control losses in paddy fields is application of urea or ammonium based
fertilizers. The N will remain in the soil as NH4+ and only small amounts
transformed to NO3 in the proximity of roots were oxygen is available, therefore,
denitrification will be slow due to a lack of substrate. In well-aerated soils, however,
nitrification rate is high and denitrification will only occur in anaerobic micro-zones
of the soil (e.g. cattle dung). Care should be taken when combining inorganic N
fertilizers with manure application as denitrification may be greatly enhanced and
fertilizer efficiency drastically reduced. Recently, a leak on the first stage of the
nitrification has been identified as a source of N2O, adding uncertainty to gaseous
emissions in well-aerated soil. Quantification of denitrification is complicated as it
is hard to tell apart from the atmospheric N2, so we use Table 24.3 that shows the
effect of organic matter on denitrification rate for different soil types.
Many of the bacteria responsible for denitrification are very sensitive to acidity
(Fig. 24.4). Thus, in soils with pH below 5, denitrification is negligible while it can
be high in basic soils. However, in soils with pH>7 the N tend to be reduced to N2
whereas in acid soils most of the emission gases are as N2O. Moreover, the
denitrification is very sensitive to temperature and increases rapidly as soil temper-
ature goes from 2 to 60 C, above which it is inhibited.
24.6.4 Ammonia Volatilization
The ammonium ion in solution is in equilibrium with ammonia (NH3), which is

volatile. Ammonia volatilization occurs naturally in soils but at slow rate. However,
volatilization losses of N fertilizer can be very important, depending on the type of
fertilizer, application form, the cation exchange capacity and climatic factors
(Table 24.4). The set of conditions with higher ammonia losses would be the
surface application of urea on a soil with basic pH and low CEC under dry
conditions. Volatilization risk can also be high with surface application of manures
since a relevant fraction of N can be ammonium. To reduce ammonia volatilization,
incorporation of ammonium fertilizers/urea/manure is recommended by tillage or
irrigation within 2 days of fertilization.
The general ranges of loss by volatilization of ammonia fertilizer are 2–50 %
(pH> 7) and 0–25 % (pH <7). The NH3/NH4+ equilibrium is pH dependent. In
acidic and neutral conditions the equilibrium is shifted to NH4+ which explains the
lower losses.
352
Table 24.3 Denitrification losses (% of inorganic N) for different cases and soil types as a function of organic matter content
Arid and semiarid rainfed crops Irrigated crops or humid areas
Rate of drainage
Case % O.M Very high High Medium Low Very low Very high High Medium Low Very low
N from fertilizer. Tilled <2 2 3 4 6 10 4 9 14 20 30
2–5 3 4 6 10 15 9 16 20 25 45
>5 4 6 10 15 25 12 20 25 35 55
N from fertilizer. No tillage <2 3 4 6 10 10 9 14 20 30 30
2–5 4 6 10 15 15 16 20 25 45 45
>5 6 10 15 25 25 20 25 35 55 55
N from manure. Tilleda <2 4 6 8 12 20 8 18 28 40 60
2–5 6 8 12 20 30 18 32 40 50 90
>5 8 12 20 30 50 24 40 50 70 95
a
The same values apply to N from fertilizer in tilled soils when a compacted impervious layer is present below the plow depth (Adapted from Meisinger and
Randall (1991))
M. Quemada et al.
Table 24.4 Ammonia volatilization loss (% of N applied) for soils as a function of soil pH, CEC and climatic conditions
24
High CEC>250 meq/kg Low CEC <100 meq/kga

Rainfall after application and climate type
>12 mm in <6 mm in No rain in >12 mm in <6 mm in No rain in
Application 2 days 7 days 7 days 2 days 7 days 7 days
N source method Humid Sub-humid Dry Humid Sub-humid Dry
pH>7 Urea Surface 0 2 2 20 30 40
broadcast
Surface 0 2 2 15 20 30
localized
Incorporated 0 0 0 10 10 10
Ammonium Surface 0 2 5 40 50 60
sulphate broadcast
Nitrogen Fertilization I: The Nitrogen Balance
nitrate broadcast
Anhydrous Injected 0 0 0 2 3 5
ammonia
ph <7 Urea Surface 0 5 5 5 30 40
broadcast
Surface 0 2 2 5 20 30
localized
(continued)
353
354
High CEC>250 meq/kg Low CEC <100 meq/kga

Rainfall after application and climate type
>12 mm in <6 mm in No rain in >12 mm in <6 mm in No rain in
Application 2 days 7 days 7 days 2 days 7 days 7 days
N source method Humid Sub-humid Dry Humid Sub-humid Dry
sulphate broadcast
nitrate broadcast
Anhydrous Injected 0 0 0 0 2 2
ammonia
a
The same values applied in no tilled soils covered with more than 50 % surface residue cover (Adapted from Meisinger and Randall (1991))
M. Quemada et al.
24.7 Crop N Uptake
Nitrogen is an essential nutrient for crops. It is a constituent of proteins, nucleic

acids, and other intermediate metabolites. If the N supply is limiting, crop growth is
reduced and so is intercepted radiation. A more severe N deficiency leads to lower
Radiation-Use Efficiency. Therefore N availability will limit biomass accumulation
and yield. There is an optimum N concentration range, above which excess N can
cause decreased yield. For instance, in indeterminate crops, high N concentration
promotes vegetative growth at the expense of reproductive growth which results in
lower harvest index. At a global level, N is the second limiting factor (after water)
in crop production.
Nitrogen uptake is parallel to biomass accumulation so it shows a typical
sigmoid curve with an initial exponential increase followed by a fast linear accu-
mulation phase. In this rapid phase, accumulation may be up to 3–5 kg N/ha/day.
The concentrations of N in the different organs are high when the plants are young
and decrease with age. Therefore, the crop response to N depends not only on the
amount absorbed but also on the translocation capacity to the growing organs (and
finally to the grain or harvestable part).
In most crops, N concentration decreases with increasing aboveground biomass
and the decline is described by a negative power function called the nitrogen
dilution curve. The critical nitrogen dilution curve has been developed for many
species (NCcrit ¼ a Biomassb) based on datasets of N concentration and biomass
under different fertilization regimes. The critical N concentration of a crop (NCcrit)
is defined as the minimum crop N concentration (%) allowing maximum biomass
production. The coefficient a is crop N concentration when biomass equals 1 t/ha,
and b is a dimensionless parameter governing the slope of the relationship. The
critical nitrogen dilution curve (Fig. 24.5) can be used to determine the crop N
status: if crop N concentration is close to the NCcrit corresponding to the current
biomass, it indicates that N is not limiting crop growth, while when it is below it
indicates a N deficiency. The ratio between the actual crop N concentration and the
NCcrit for a given biomass is known as the N nutrition index (NNI):
NCactual NCactual
NNI ¼ ¼ ð24:2Þ
NCcrit a Bb
where B is crop biomass (t/ha). In general, C4 species have a lower NCcrit for a
given biomass than C3 species, presumably related to a lower content of photosyn-
thetic proteins. The dilution curves are generally accepted because of its simplicity
for modeling crop growth during vegetative stages, however, when other factor
different from N limits growth (i.e. severe drought, disease) the curve may depart
greatly from the model. The requirement of destructive samples for measuring
biomass and the need to fit the dilution curve to local conditions or cultivar specific
characteristics, make difficult the adoption of dilution curves as a management tool.
Fig. 24.5 Critical nitrogen 6

concentration curve for
N CONCENTRATION (% dm)
various crops (Adapted 5
from Gastal F. and Lemaire C3
G. (2002). J Exper Bot 53: 4
789–799) C4
3
0
0 2 4 6 8 10 12 14
BIOMASS (t dry matter/ha)
YIELD
IRRIGATED
IRRIGATED
RAINFED
RAINFED
N APPLIED N UPTAKE
IRRIGATED
RAINFED
N APPLIED
Fig. 24.6 Response of sunflower yield to N applied and N uptake
Other techniques based on crop N status have been developed for fertilizer recom-
mendation and will be discussed in Chap. 25 (Fig. 24.5).
The relationship between yield and N uptake is generally linear until the
maximum yield is reached. From that point, if there is N available in the soil,
absorption continues, but it does not result in higher yield. This limit depends on
environmental conditions and crop management. Figure 24.6 shows a linear rela-
tionship between yield and N uptake for an experiment conducted with sunflower in
Cordoba with different levels of irrigation and N fertilizer. The maximum yields,
where the yield-N uptake relation saturates, increased with irrigation levels. There-
fore, despite the linear relationship between yield and N uptake when no other
factor is limiting, it is important to set the objective yield to define the maximum
level of N uptake.
For a set of environmental conditions, the relationship between yield and applied
N is curvilinear (Fig. 24.6). Therefore, the apparent efficiency of the N fertilizer
decreases with increasing the dose. When this dose reaches a certain value, an
increase of fertilizer does not result in a yield increase, and in some cases it may
even be detrimental. Furthermore, the amount of residual soil N will be greater,
which increases the risk of nitrate leaching.
The yield response to N applied, depends on the initial availability of soil N and
the mineralization potential during the season, besides the production potential of
the crop. Thus, in very fertile soils, the crop may not respond to the application of N
or the response may be negative. If another factor (i.e. water) is limiting, the high N
input will not bring yield increases. Figure 24.6 illustrates this behavior in the
sunflower experiment mentioned earlier. As the irrigation amount was higher, so
were yields for any dose of N. The response to applied N was also higher under
irrigation, at least for low N doses.
As a framework for understanding the responses to fertilization, de Wit proposed
to represent in different quadrants the curves of N uptake and yield in response to
the N application (Fig. 24.6). In quadrant (a) yield is plotted as a function of N
applied for different irrigation regimes of the sunflower experiment This yield
corresponds to an amount of N absorbed (quadrant b), which in turn corresponds
to a rate of N application (quadrant c). Each line of the lower quadrant is charac-
terized by its slope and its intercept. We see that both the intercept and the slope of
the N uptake-N applied lines increased with applied irrigation. This means that
irrigation increased N availability, because either NO3 was applied in irrigation
water, the mineralization rate was enhanced by irrigation or N uptake was facili-
tated in a wetter soil. Furthermore, as the level of irrigation increased, the curves of
N uptake did not show saturation, i.e. a ceiling of N absorption was not reached.
Water and N are the main limiting factors in many irrigated systems and a
combined management should be followed for a successful crop performance.
In any case the criterion for choosing the amount of N fertilizer should be
economical, i.e. the optimum amount will be that that results in maximum profit.
This amount will be lower than that required for maximum yield and may be
calculated as the point where the marginal profit is zero.
The results of the sunflower experiment mentioned above contrast with other
previous experiments on sunflower fertilization in Cordoba, which did not show
response to N fertilization. It should be noted that in the experiment mentioned the
soil had been “cleaned” of N with a previous unfertilized cereal crop. Obviously
initial fertility conditions and other environmental factors (i.e. water supply) greatly
affect crop responses to N application. This is why production functions of yield
versus N applied cannot be extrapolated to other situations. To emphasize this
concept, results from a rainfed experiment conducted in three adjacent fields in
Navarra (Spain) where the response of wheat to increasing rates of N fertilizer
application are presented in Fig. 24.7. No yield response was observed when high
initial inorganic N (>140 kg N/ha) was present in the soil (top 0.9 m) before
5.0
4.5
GRAIN YIELD (t/ha)
4.0
3.5
3.0
Low soil inorganic N
2.5 Medium soil inorganic N

High soil inorganic N
2.0
0 50 100 150 200 250
NITROGEN APPLIED AS FERTILIZER (kg N/ha)
Fig. 24.7 Wheat response to N fertilizer application based on the soil inorganic N content
determined in the upper 0.9 m of soil
planting, whereas yield response increased for the medium (90 kg N/ha) and low
(30 kg N/ha) inorganic N fields. Because of that, determination of available N in
soil samples taken before planting or before side-dress applications is a
recommended practice to avoid over-fertilization in many regions.
24.8 Nitrate Leaching
The consequences of N losses from agricultural systems to water bodies is a major

social concern in developed countries, with special attention to aquifer contamina-
tion by nitrate and excessive N availability in lakes and estuaries. Potential harmful
effects of nitrate on human health (cyanosis, risk of cancer) had led to the estab-
lishment of maximum allowable concentrations of nitrate in drinking water of 50 g
NO3 m3 (World Health Organization). The EU and the USA have identified
regions affected by excessive nitrate contamination and passed legislation to
prevent it (i.e. Directives 2000/60/EC; USA Congress, 1978). As the main contrib-
utor to nitrate pollution is agriculture, restrictions to the use of N fertilizers and to
agricultural practices that may enhance nitrate leaching have been implemented in
many countries.
Nitrate leaching occurs as soil water containing dissolved nitrate drains below
the root zone. Therefore leaching will be proportional to deep percolation and to
nitrate concentration in the soil solution. Deep percolation will depend on the
components of the water balance (rain and irrigation), on the water retention of
the soil and on its hydraulic conductivity. Soils with high water retention capacity
and low conductivity (e.g. fine-textured) will have therefore a lower percolation and
leaching potential. Apart from soil characteristics and climatic conditions, fallow
periods between successive crops in the rotation are the most dangerous for
leaching. Nitrate left in the soil at harvest plus that originated from mineralization
and nitrification during the fallow period is left available for leaching during
drainage episodes after heavy rains. The absence of a crop extracting water and
nitrate is ideal for keeping a high risk of leaching (high water content, high nitrate
concentration). This has led to the introduction of “catch” crops to fill the gap of
fallow periods as they reduce water content and absorb inorganic N which is thus
fixed in organic form. Other possible measures for reducing leaching would be
earlier plantings (to reduce fallow periods), reduce basal N applications in the
autumn or use slow-release fertilizers (Chap. 23). In irrigated systems it is
extremely important to follow irrigation schedules based on the water balance
with corrections at the end of the season to deplete soil water as much as possible.
To evaluate the risk of leaching we may use the Leaching Index (LI, mm) which
is an estimate of the amount of percolation below a soil depth of 1 m and was
proposed by the USDA (Williams and Kissel 1991). The LI is calculated as the
product of a Percolation Index (PI) and a Seasonal Index (SI):
LI ¼ PI SI ð24:3Þ
The Percolation Index is calculated as:

0
ðP 10160=CN þ 101:6Þ2 0
PI ¼ 0 if P 10160=CN þ 101:6 > 0 ð24:4Þ
P þ 15240=CN 152:4
where P is annual rainfall (mm) and CN0 is a modified curve number with values
28, 21, 17 and 15 for hydrologic groups A, B, C and D, respectively (Chap. 8). If the
condition stated in Eq. 24.4 is not met then PI¼0.
The Seasonal Index represents the concentration of rainfall during the winter
period:
1=3
2Pw
SI ¼ ð24:5Þ
P
where Pw is total rainfall (mm) during autumn and winter (1 October–31 March in
N latitudes, 1 April–30 September in S latitudes).
The Leaching Index is only indicative of potential losses by leaching but not of
actual losses. If the LI is high, adequate crop and soil management may lead to low
actual leaching. On the contrary with low LI we may expect low actual leaching
independently of actual management. In other words, measures to reduce the
concentration of nitrate will be very effective in reducing leaching in locations
with high LI.
Example 24.4 Let’s calculate the LI for two locations, Adelaide (Australia)
and Dublin (Ireland) for a soil of hydrologic class A (CN0 ¼ 28) using the
monthly rainfall shown below:
Month 1 2 3 4 5 6 7 8 9 10 11 12 Year
Adelaide, 19 20 22 38 57 50 67 51 40 37 23 24 448
Australia
Dublin, Ireland 69 50 54 51 55 56 50 71 66 70 64 76 732
Applying Eq. 24.4 to Adelaide (P ¼ 448 mm, Pw ¼ 303) we obtain

PI¼41.5 mm and SI ¼ 1.1 (Eq. 24.5) so LI¼45.9 mm.
For Dublin (P ¼ 732 mm, Pw ¼ 383 mm) the PI¼197 mm and SI ¼ 1.02 so
LI¼200 mm. The risk of leaching is much higher in Dublin than in Adelaide.
We will now illustrate how we can convert LI values to approximate
leaching values. Let’s assume that the soil has water contents at Field
Capacity and saturation, θFC ¼ 0.25 m3m3 and θSAT ¼ 0.45 m3m3. We
also assume that during percolation the soil water content is the average of
those values, i.e. 0.35 m3m3. We can compare situations of low and high
nitrate content at the start of the winter period (e.g. 25 versus 100 kg N/ha in
1 m depth) by assuming that all percolation occurs during winter. The N loss
by leaching can be calculated as:

LI
N leached ¼ N init 1 exp ð24:6Þ
Z θmean
where Ninit is the initial N content (kg N/ha), Z is soil depth (mm) and θmean is
the average water content during percolation (0.35 m3m3 in this soil).
Applying this equation we deduce that leaching could be between 3 and
12 kg N/ha in Adelaide and between 11 and 44 kg N/ha in Dublin. In the
latter the reduction in leaching by reducing soil N would be 33 kg N/ha, while
in the former the reduction would only be 9 kg N/ha.
24
Appendix 24.1: Nitrogen concentration in different crop species. Maximum and minimum values are
shown when available. Also the dry matter content (% over fresh mass) is indicated
DM N N N DM N N N
Cereals & pseudocereals % min max typical % min max typical
Barley 2 row Grain 86.5 2.20 2.40 2.30 Straw 90.5 0.58 0.88 0.70
Barley 6 row Grain 88.5 1.50 1.80 1.60 Straw 90 0.58 0.88 0.70
Buckwheat Seed 94.85 2.96
Corn Grain 86 1.35 1.75 1.60 Stover 87 0.90 1.10 0.97
Maize Silage 30 1.10 1.45 1.25
Millet (finger) Grain 90 2.20 Stover 92 0.67
Millet (pearl) Grain 89.5 2.00 Stover 91.5 0.80
Millet (proso) Grain 90.5 2.30 92 0.80
Oat Grain 91 1.50 1.80 1.60 Straw 89.5 0.60 0.80 0.70
Quinoa Seed 94.5 2.43
Rice Grain 94 1.33
Rice (milled) Grain 87.5 1.05 1.65 1.40 Straw 90 0.50 0.80 0.70
Rye Grain 87 2.00 2.40 2.20 Straw 90.5 0.35 0.65 0.50
Sorghum Grain 87.5 1.45 2.00 1.90 Stover 92 0.60 0.80 0.70
Sorghum Silage 26 0.70 1.30 1.00
Sorghum Green 20 1.30 1.40 1.37
Triticale Grain 89 2.20 2.50 2.45 Straw 90 0.60 0.90 0.70
Wheat (bread) Grain 87.5 1.85 2.30 2.10 Straw 90.5 0.40 0.85 0.65
Wheat (durum) Grain 87.5 2.05 2.70 2.40 Straw 90.5 0.40 0.85 0.65
(continued)
361
DM N N N DM N N N
362
Sugar, oil & fiber crops % min max typical % min max typical
Castor bean Seed 95 2.70
Cotton Seed 91 2.32 2.75 2.53 Residues 92.5 0.90 1.00 0.98
Flax Seed 93.5 3.30 4.30 3.80 Straw 93 1.00 1.20 1.06
Opium poppy Capsule 87.5 2.30 3.10 2.60 Straw 90 0.80 1.20 1.00
Rapeseed Grain 91 3.40 4.30 3.90 Residues 82.5 0.55 0.90 0.80
Safflower Seed 92 2.60 2.80 2.70 Residues 90 0.60
Sugar beet Root with 20 1.20 1.40 1.30 Residues 18 1.80 2.80 2.30
crown
Sugar beet Root w/o 21 0.90 1.10 1.05 Residues 18 1.80 2.80 2.30
crown
Sugarcane Tops 25 0.13 26 0.41
Sunflower (oil) Seed 91.5 2.20 3.20 2.95 Residues 87 0.40 1.10 0.80
Sunflower (seed) Seed 91.5 2.80 3.60 3.20 Residues 87 0.40 1.10 0.80
Tobacco Burley Leaf + stem 75 3.80 4.20 4.00
Tobacco Flue Leaves 8 2.00 2.30 2.10 Stalk 9 0.75 1.00 0.80
DM N N N DM N N N
Legumes % min max typical % min max typical
Bean (dry) Seed 89 3.50 4.50 4.00 Straw 89 1.10 1.40 1.20
Black eyed pea Seed 90 4.00 4.20 4.10 Straw 90 1.25
Chickpea (desi) Seeds 89.5 3.50 Straw 89.5 0.85
Chickpea (desi) Seeds 89.5 3.60 Straw 89.5 0.85
Faba bean dry Grain 90 3.00 4.90 3.70 Straw 85 0.80 2.50 1.60
Groundnut Fruits 93 4.10 4.30 4.25 Residues 90.5 1.50 1.70 1.65
Groundnut Seeds 92 4.70 4.90 4.85 Residues 90.5 1.50 1.70 1.65
Lentil Grain 89 4.20 4.40 4.30 Straw 91 1.10
M. Quemada et al.
24
Pea Seed 90 4.00 4.30 4.20 Straw 88.5 1.20 1.40 1.30
Soybean Grain 87.5 6.10 6.90 6.50 Residues 89 1.00 1.00 0.85
DM N N N DM N N N
Forages % min max typical % min max typical
Alfalfa (green, vegetative) Medicago Biomass 25.0 3.05 4.05 3.55
Alfalfa (green, flowering) Medicago Biomass 25.0 2.10 3.10 2.60
Alfalfa (hay, vegetative) Medicago Biomass 85.0 2.80 3.80 3.30
Alfalfa (hay, flowering) Medicago Biomass 85.0 2.00 3.00 2.50
Bluegrass-Kentucky (hay) Poa pratensis Biomass 89.1 1.60
Bromegrass (hay) Bromus sp. Biomass 91.1 1.53
Canarygrass-Reed (hay) Phalaris arundinacea Biomass 89.0 1.70
Clover-Alsike (hay) Trifolium hybridum Biomass 87.4 2.27
Clover-Crimson (hay) Trifolium incarnatum Biomass 88.3 2.65
Clover-Red (hay) Trifolium pratense Biomass 86.1 2.51
Clover-White (hay) Trifolium repens Biomass 90.3 3.09
Clover-White-Ladino (hay) Trofolium repens Biomass 89.2 3.32

Fescue (hay) Festuca or Lolium sp. Biomass 90.0 1.51
Fescue-Meadow (hay) Lolium pratense Biomass 88.4 1.19
Fescue-Tall (hay) Lolium arundinaceum Biomass 91.6 2.31
Grass (hay) Poaceae Biomass 89.1 1.52
Grass (silage) Poaceae Biomass 24.9 1.84
Millet-Foxtail (silage) Setaria italica Biomass 27.6 1.59
Millet-Pearl (silage) Pennisetum glaucum Biomass 21.6 1.54
Oat (hay) Avena sativa Biomass 89.8 1.37
Orchardgrass (green chop) Dactylis glomerata Biomass 25.9 2.37
Orchardgrass (hay) Dactylis glomerata Biomass 89.2 1.71
Rye (hay) Secale cereale Biomass 92.6 1.21
(continued)
363
Ryegrass-Perennial (hay) Lolium perenne ssp. Biomass 87.3 1.49
364
perenne
Sweetclover (hay) Melilotus sp. Biomass 89.6 2.65
Timothy (hay) Phleum pratense Biomass 90.5 1.20
Trefoil-Birdsfoot (hay) Lotus corniculatus Biomass 90.0 2.52
Turnip (green chop) Brassica rapa var. rapa Root 13.5 3.30
Vetch (green) flowering Biomass 25.0 2.60 3.40 3.00
Vetch (hay) Vicia sativa Biomass 87.8 2.60
Vetch (hay) flowering Biomass 85.0 2.50 3.30 2.90
Vetch-Hairy (hay) Vicia villosa Biomass 87.9 3.68
Wheatgrass (hay) Poaceae Biomass 91.6 1.34
DM N N N DM N N N
Horticultural crops % min max typical % min max typical
Artichoke Fruit 17 2.40 2.60 2.50
Asparagus (green) Stems 8 5.40 5.60 5.50
Asparagus (white) Stems 7 4.20 4.40 4.30
Bean (green) Pods 30 3.00 3.70 3.30 Straw 89 1.10 1.40 1.20
Beet Root 12 2.10 2.20 2.15 Residues 12 3.60 3.80 3.70
Broccoli Stems 45 3.30 3.40 3.35
Brussels sprouts 13 4.90 5.10 5.00
Cabbage Cabagge 9 3.00 3.50 3.30 Residues 14 2.60 2.80 2.70
Carrot Root 12 1.40 1.60 1.50 Leaves 16 2.30 2.40 2.35
Cauliflower 9 4.30 4.50 4.40
Celery Leaves 5 2.30 2.50 2.40
Chicory Leaves 6 3.10 3.30 3.20
Cucumber Fruit 4 2.40 2.60 2.50
Eggplant Fruits 7 2.40 2.60 2.50
M. Quemada et al.
24
Endive Leaves 5 4.10 4.20 4.15
Faba bean green Fruits 19 4.60 4.80 4.70 Straw 25 2.20
Garlic Heads 39 2.50 2.80 2.60 1.50
Leek Bulb 17 1.30 1.50 1.40 Leaves 20 0.90
Lettuce Iceberg Leaves 5 2.40 2.70 2.55
Lettuce Roman Leaves 6 4.00 4.40 4.27
Melon Fruit 12 0.80 1.00 0.90
Musk melon Fruit 10 1.40 1.60 1.50
Onion Bulb 10 1.90 2.50 2.20 Leaves 8 3.80 4.10 4.00
Parsley Leaves 10 3.30 3.60 3.50
Pea (green) Fruits 12.5 2.20 3.70 3.00 Straw 25 1.90 2.10 2.00
Pea (green) Seeds 21 4.30 4.50 4.40 Straw 25 1.90 2.10 2.00
Pepper green Fruit 10.5 2.10 2.40 2.30
Pepper red Fruit 12.5 1.40 2.00 1.90
Pumpkin Fruit 9 2.50 2.70 2.60
Radish Root 6 1.50 1.70 1.60

Spinach Leaves 9 5.10 5.30 5.20
Squash (immature) Fruit 5.5 2.70 3.84 3.20
Squash (mature) Fruit 14 0.90 1.00 0.91
Strawberry Fruit 9 1.10 1.60 1.35
Sweet corn dry Grain 90.5 1.90 2.10 2.00 Stover 91 0.90 1.10 1.05
Sweet corn fresh Grain 35 1.50 1.70 1.60 Stover 23 1.10 1.30 1.20
Tomato Fruit 6 2.30 3.10 2.60 20 1.80
DM N N N DM N N N
Fruit trees, vines and shrubs % min max typical % min max typical
Watermelon Fruit 9 1.00 1.50 1.25
Almond With hull 85 3.00 3.60 3.30
(continued)
365
Apple Fruit 18 0.25 0.45 0.35
366
Apricot Fruit 14 1.50 1.70 1.65

Avocado Fruit 27 1.10 1.30 1.20
Banana Fruit 26 0.60 0.70 0.65
Cherimoya Fruit 26 0.70 0.90 0.78
Cherry Fruit 19 1.00 1.20 1.07
Date palm Fruit 77 0.35 0.45 0.40
Fig Fruit 21 0.50 0.60 0.57
Grape Fruit 19 0.50 0.60 0.57
Grapefruit Fruit 11 0.70 0.80 0.75
Hazel nut Fruit 94.5 2.10 2.30 2.20
Kiwi Fruit 17 0.80
Lemon Fruit 13 1.30 1.70 1.50
Mango Fruit 18 0.50
Oil palm Kernel 79 1.25
Olive Fruit 50 0.20 0.40 0.30 Vegetative 70 1.00 2.00 1.50
Orange Fruit 18 1.00 1.40 1.20
Peach Fruit 12 0.80 1.20 1.00
Pear Fruit 18 0.30 0.50 0.40
Persimmon Fruit 20 0.55
Plum Fruit 15 0.80 0.90 0.85
Pomegranate Fruit 25 0.60
Quince Fruit 16 0.54
Walnut Fruit 93 0.20
M. Quemada et al.
24
DM N N N DM N N N
Roots, tubers & bulbs % min max typical % min max typical
Cassava Root 68.8 0.42
Chinese yam Tuber 59.3 1.39
Potato Tuber 23.5 1.20 1.90 1.60 Residues 51 2.00 2.40 2.20
Sweet potato Tuber 65 0.88
White yam Tuber 63.1 0.94
Yellow yam Tuber 58.3 0.99
367
Bibliography
Brady, N. C., & Weil, R. R. (2008). The nature and properties of soils (4th ed.). Upper Saddle
River: Pearson Education.
Meisinger, J. J., & Randall, G. W. (1991). Estimating nitrogen budgets for soil-crop systems. In
R. F. Follet, D. R. Keeney, & R. M. Cruse (Eds.), Managing nitrogen for groundwater quality
and farm profitability. Madison: Soil Science Society of America.
Scheppers, J. S., & Raun, W. R. (2008). Nitrogen in agricultural systems (Agronomy monograph,
Vol. 49). Madison: ASA, CSSA & SSSA.
Williams, J. R., & Kissel, D. E. (1991). Water percolation: An indicator of nitrogen-leaching
potential. In R. F. Follet, D. R. Keeney, & R. M. Cruse (Eds.), Managing nitrogen for
groundwater quality and farm profitability (pp. 59–83). Madison: Soil Science Society of
America.
Chapter 25
Nitrogen Fertilization II: Fertilizer
Requirements

Abstract The N balance allows the calculation of the fertilizer requirement which
depends on the amount of N absorbed by the crop, the amount of inorganic N in the
soil or produced by mineralization and N losses of the system, which are quantified
by the Recovery Efficiency. A fertilization plan should take into account the
variability of environmental factors, especially rain, to distribute the N with flex-
ibility to match the specific conditions of each year. Doing so we will avoid yield
reductions due to N deficiency and the negative environmental impacts by excess
application. Fertilization of trees should be based on the nutrient balance (mature
trees) or the growth rate expected from actual transpiration.
M. Quemada (*)
A. Delgado
Sevilla, Spain
L. Mateos
F.J. Villalobos

Agriculture, DOI 10.1007/978-3-319-46116-8_25
25.1 Calculation of N Fertilizer Doses Using the Nmin

Method
As stated in Chap. 24, crop response to N fertilization is very dependent on the

amount of available N (essentially inorganic N) present in soil before fertilization.
Thus, it can be expected low N fertilizer needs when a large amount of available N
is present in the soil. Nmin methods are based on the amount of mineral N present in
the soil at the end of winter, recommending decreased N rates with increasing levels
of available N in the soil. To implement this method, it is necessary to establish a
relationship between the economic optimum N rate for a crop and the amount of
mineral N in the soil at a given depth at the beginning of the crop cycle in a given
geographical area with similar environment (soil + climate) and crop management.
To establish this relationship it is necessary to conduct a network of trials in various
locations including different levels of mineral N in the soil. The model fitted is:
N f ¼ Af Bf N min ð25:1Þ
where Nf is the optimum economic dose of N, the parameter Af represents the total
N needed by the crop for these given environmental and management conditions,
and Bf is the amount of N provided by the soil per unit N content in the sampled soil
depth (Nmin). Thus, this model for estimating N rates represents the difference
between the required mineral N by the crop and the amount actually delivered by
the soil.
Sometimes, the method is simplified by only determining soil nitrate, since most
of inorganic nitrogen is in nitric form at the end of the winter, but care should be
taken as ammonium may be relevant after cold winters or in acid soils. Another
common simplification is taking soil samples for inorganic N analysis only in the
top layer (i.e. 30 cm depth) and use it as a surrogate of the N supply potential of the
whole soil. However, it is recommended to sample the depth of the effective
nutrient uptake by the crop. These methods have being applied with success in
Europe (Nmin method) and also in the USA, where it is known as “preplant nitrate
test” (PPNT).
Some additional factors have to be considered when this method is used to
estimate the requirement of N fertilizer:
• Winter crops: at the end of winter (before applying topdressing) the soil is
analyzed to determine Nmin and adjust the N rate. If the expected crop growth
during winter is small, basal N fertilizer application is not recommended to
avoid nitrate losses by leaching or denitrification. In warm areas where crop
growth and N uptake are important before topdressing, a small fraction of
fertilizer (15–30 %) might be applied before planting.
• Spring crops: analysis and estimation of the dose is based on Nmin determination
before planting and the fertilizer is split in a pre-planting application and
25 Nitrogen Fertilization II: Fertilizer Requirements 371
topdressings during the growth cycle, as performed for sugar beet in several
Northern European countries or for maize in South Europe.
• The method is only valid for environmental and management conditions for
which it has been developed. Changes in management, such as adding manure
amendments or incorporation of crop residues may require adapting the
recommendation.
25.2 Calculation of N Fertilizer Doses Using the N Balance
The estimation based on nutrient mass-balance can be the recommended strategy

for mobile nutrients, such as N, which are not efficiently retained in soils and can be
lost through leaching or gaseous emissions. It should be remarked that the mass-
balance approach is targeted to preplant estimates of N needs. For N, the increase in
soil inorganic N, i.e. the difference between final (Nend) and initial (Ni) soil N
content, for a non-leguminous crop cycle may be written as:
N end N i ¼ N f þ N m þ N irr þ N dep N c N l N d N v ð25:2Þ
where Nf, Nm,Nirr and Ndep are N inputs as fertilizer applied, mineralized, and N in
irrigation water and atmospheric deposition, respectively. The outputs are N
absorbed by the crop, N lost by leaching, N lost by denitrification and N lost by
ammonia volatilization. The N balance equation can be simplified to:
N end N i ¼ N f þ N m þ N other ðN biom þ N root Þ N loss ð25:3Þ
where Nloss includes all losses of N, and Nother includes other minor inputs (irriga-
tion, deposition). The equation now shows the two components of crop N: that
accumulated in aboveground biomass (Nbiom) and that accumulated in roots (Nroot).
25.2.1 Crop N Uptake (Nc)
Crop N content of the aboveground biomass (Nbiom, kg N/ha) is calculated as a

function of expected aerial biomass production (yield and residues) and N concen-
tration of biomass components:

1 HI
N biom ¼ Y NCyield þ ðB Y ÞNCres ¼ Y NCyield þ NCres ð25:4Þ
HI
where Y is yield (kg/ha), B is aerial biomass (kg/ha), HI is harvest index and NCyield
and NCres are N concentrations in the harvested organ and the residues,
respectively. To calculate total N accumulated by the crop (Nc) we need to add the
N accumulated in roots as follows:

N c ¼ N biom ð1 þ f NR Þ ¼ N yield þ N res ð1 þ f NR Þ ð25:5Þ
where fNR is the ratio of N in roots and N in shoots and Nyield and Nres are the
amounts of N in yield and residues, respectively.
If biomass is overestimated, the same will happen with N applied and N losses
will occur. On the other hand, under predicting biomass will lead to insufficient N
and thus, N will become the limiting factor for yield. Usually, N in roots accounts
for 5–25 % of N in aboveground biomass, so values of 15–20 % can be considered
acceptable for field crops, and up to 25 % for horticulture crops.
The estimated yield (also called target yield) should be based on previous years’
yields with inputs similar to those the farmer intends to use. If this species has not
been previously cultivated, the yield should be estimated based on yields from
neighboring farms, always making sure that there were no limiting production
factors unrelated to those in our crop. Additionally, for N it is always useful to
set the maximum and minimum expected yields to establish maximum and mini-
mum values of Nc and so decide the most appropriate strategy, which we will
discuss later.
The crop N concentration can be determined a posteriori by analyzing the
biomass produced. However, to design the fertilization program it is necessary to
have estimates of N concentration a priori. Table 24.1 lists the N concentrations
referred to dry matter for different crop species. In years when we plan to apply less
N than needed or when other factors are not limiting (rainy year or under irrigation),
we must choose the lower values of the ranges given in Table 24.1. In all cases it
should be noted that the values refer to dry matter, so estimated yields should be
corrected according to the expected moisture content (Table 24.1).
25.2.2 Initial and Final Soil N and Mineralization
The initial soil inorganic N content can vary greatly and values between 30 and
500 kg N/ha in the upper 1 m soil have been reported. The common strategy will
seek to deplete soil N during the crop cycle, i.e. try to make the final soil inorganic
N (Nend) as low as possible. The Nend is also called the residual N and below a
threshold value (between 10 and 70 kg N/ha depending on soil texture and depth)
it cannot be recovered by the crop. Determinations of inorganic N in the soil
should include both NO3 and NH4+. However, the concentration of NO3 is
usually high relative to that of NH4+ so we may analyze only NO3 concentration
to determine Ni.
Mineralization of soil organic matter may be an important source of N. The N
mineralized during the crop campaign is the net result of the mineralization of the
stable organic matter and the residues and roots from the previous crops. This
contribution is hard to estimate, so a common approach is assuming that the soil is
in steady state and to calculate the soil N supply as the addition of the N in the
residues and the roots of the previous crop. Management factors such as tillage,
irrigation or application of manure, can greatly modify the steady state and produce
pulses of N mineralization that should be taken into account to profit from the N
supply and avoid pollution problems. For calculation purposes we can group initial
N and N mineralized during the current season into a single component (Ni+m)
which should be proportional to the amount of N in residues from the previous crop
(N0 res in shoots and N’root in roots):

N iþm ¼ kim Fres N 0res þ N 0root ¼ kim Fres N 0res þ f NR N 0yield þ N 0res ð25:6Þ
where Fres is the fraction of residues that are left in the field, and N0 yield and N0 res
refer to N accumulated in the harvest organ and residues of the previous crop,
respectively. The coefficient kim has a maximum value of 1, if all the aboveground
residues are mineralized with no loss. Lower values are expected if the residues are
not incorporated by tillage or when the N concentration in residues is low. We
propose kim ¼ 0.9 for legumes with tillage, 0.7 for legumes left on the ground and
for non-legumes with tillage and 0.5 for non-legumes left on the ground. Note that
N in roots is assumed to be fully available and can be calculated as a function of N
in aboveground biomass of the previous crop.
25.2.3 Fertilizer Requirement Calculation According to N

Balance
From Eqs. 25.5 to 25.6 we may deduce the following formula for calculating N
fertilizer requirements:

N end þ ð1 þ f NR Þ N yield þ N res kim Fres N 0res f NR N 0yield þ N 0res N other
Nf ¼
ð 1 nÞ
ð25:7Þ
Nother is the total N received by atmospheric deposition (5–10 kg N/ha, as a

conservative value) and irrigation water. The coefficient n is the fraction of applied
N lost (leaching, volatilization, denitrification) and fNR is the ratio N in roots/N in
shoots.
This model assumes that most of soil N supply (Nm and Ni) and N in irrigation
water and atmospheric deposition are taken up by crops with no losses. There are
several ways to calculate the efficiency of fertilization. The most commonly used,
which we may call Ef, is the fraction of applied N that is finally accumulated in the
aerial part of the crop, and is equal to the slope of the relationship between N uptake
and N applied (Chap. 24, Fig. 24.6). However, this definition does not take into
account N in roots and the fact that a fraction of the N left in the soil at the end of the
growing cycle may be available for the next crop in the rotation. This explains why
measured values of Ef are typically in the range 0.4–0.75, and values between 0.5
and 0.7 are usually acceptable. A second definition of efficiency (EfR) would
include roots, so it would be the fraction of total N applied that is accumulated in
the crop shoots and roots. In mathematical terms:
Ef R ¼ Ef ð1 þ f NR Þ ¼ 1 n ð25:8Þ
With fNR ¼ 0.2, this equation implies that the normal range of EfR is 0.5–0.9. So if
we aim at Ef ¼ 0.6–0.7, then our aim is EfR ¼ 0.7–0.85.
Besides agronomic considerations in some cases we will have to take into
account other environmental constraints. In several countries, the amounts and
types of N fertilizer have been restricted by law to prevent nitrate pollution.
The N balance and other methods to estimate crop N requirement are intended to
achieve an optimal N supply for optimal yields, which depends on environmental
conditions. However, depending on the ratio of fertilization cost to crop value,
estimated requirements can be modified. When fertilizer prices rise, maximum
profit is obtained with fertilization rates lower than those estimated as optimal.
Estimated high N rates expecting high yields can lead to increased N losses or high
amounts of residual N remaining in the soil after crop harvest if conditions are not
suitable for optimal crop performance.
Example 25.1 A cereal crop is grown as monoculture on an acid soil with the
following distribution of yields:
2000–3000 kg/ha in 40 % of the years
The average yield is therefore 3500 kg/ha.
We assume that all the residues of the previous crop stay in the field
(FRES ¼ 1) and are incorporated by tillage. So kim ¼ 0.7.
The concentration of N is 1.5 % in grain and 0.5 % in straw. The Harvest
Index is 0.5. The water content of grains is 10 %. We assume fNR ¼ 0.2,
Nother ¼ 5 kg N/ha. Our aim is EfR ¼ 0.85 (losses of denitrification and
volatilization are very low in acid soils) and Nend ¼ 25 kg N/ha.
How would we modify the cereal N fertilizer strategies if faba bean (dry
yield 1500 kg/ha, HI ¼ 0.3) is introduced as a preceding cash crop?
(continued)

(a) For each interval of yield (taking the midpoint) we calculate the N
fertilizer requirement. For instance for yield 2500 kg/ha:
Nc (N uptake):
Yield ðdry matter Þ ¼ 2500ð1 0:1Þ ¼ 2250 kg=ha
Residues ¼ Y ð1 HIÞ=HI ¼ 2250 kg=ha
Nyield ¼ Y NCy ¼ 2250 0:015 ¼ 33:75 kg N=ha
Nres ¼ 2250 0:005 ¼ 11:25 kg N=ha
We asume that the previous wheat crop had an average yield (3500 kg/ha)
Yield ðdry matterÞ ¼ 3500ð1 0:1Þ ¼ 3150 kg=ha
Residues ¼ Y ð1 HIÞ=HI ¼ 3150 kg=ha
N0yield ¼ Y0 NCy ¼ 3150 0:015 ¼ 47:25 kg N=ha
N0res ¼ 3150 0:005 ¼ 15:75 kg N=ha
25 þ ð1 þ 0:2Þð33:75 þ 11:25Þ 0:7 1 15:75 0:2ð47:25 þ 15:75Þ 5

Nf ¼
0:85
¼ 59 kg N=ha
For the other yield values we will have:

For Y ¼ 3500 kg/ha: Nf ¼ 85 kg N/ha
For Y ¼ 4500 kg/ha: Nf ¼ 110 kg N/ha
For Y ¼ 5500 kg/ha: Nf ¼ 135.5 kg N/ha
(b) Possible strategies:
– Apply 59 kg N/ha every year. The crop would always be limited by N
and we would have an average yield of 2500 kg/ha so the average
exported N would be 2500 0.015 ¼ 37.5 kg N/ha. The ratio N
exported/N applied would be 0.64.
– Apply 85 kg N/ha every year. In the best case yield would be 3500 kg/
ha. The average yield is then:
0:4 2500 þ 0:6 3500 ¼ 3100 kg=ha
So the average exported N would be 46.5 kg N/ha and the ratio N

– Apply 110 kg N/ha every year. The average yield is then:
0:4 2500 þ 0:3 3500 þ 0:3 4500 ¼ 3400 kg=ha
So the average exported N would be 51 kg N/ha and the ratio N

(continued)

– Apply 135.5 kg N/ha every year. The average yield is then:
0:4 2500 þ 0:3 3500 þ 0:2 4500 þ 0:1 5500 ¼ 3500 kg=ha
The average exported N would be 52.5 kg N/ha and the ratio N exported/
N applied would be 0.39.
The best strategy would depend on the price of the grain and the cost of
fertilizer. In any case it is always better to follow a flexible strategy,
i.e. basal application lower than 59 kg N/ha and then apply a top dressing
2 months later. The latter would be omitted if the year came bad, and
would be between 26 and 76 kg N/ha depending on the actual conditions
of the year. By following a flexible strategy, if we can exactly match the
N requirements of each type of year we would apply an average amount
of N:
0:4 59 þ 0:3 85 þ 0:2 110 þ 0:1 135:5 ¼ 84:7 kg N=ha
While the average exported N is 35000.015 ¼ 52.5 kg N/ha and the ratio
N exported/N applied would be 0.62.
(c) After the faba bean crop, the available N at wheat sowing will increase.
Measurement of soil N before sowing would allow quantification of this
N supply. If the measurement is not available, we can calculate the
fertilizer requirement after faba bean. First we calculate N in yield and
residues of the faba bean:
N0yield ¼ Y NCy ¼ 1500 0:037 ¼ 55:5 kg N=ha
N0res ¼ Y ð1 HIÞ=HI NCr ¼ 1500 0:7=0:3 0:016 ¼ 56 kg N=ha
25 þ ð1 þ 0:2Þð33:75 þ 11:25Þ 0:9 1 56 0:2ð55:5 þ 56Þ 5

Nf ¼
0:85
¼ 1:5 kg N=ha
Which is almost zero, i.e. no fertilizer should be applied. The amount of N

to apply for the other yield classes will be 27, 52 and 78 kg
N/ha. Therefore, the basal N application to wheat could be avoided and
the topdressing between 27 and 78 kg N/ha would be applied 2 months
later.
The N fertilizer savings provided by the introduction of the faba bean
would be added to other agronomic advantages of a crop rotation. The
faba bean income will probably be lower than that of the cereal, but the
introduction of the legume crop every 2 or 3 years may increase the
sustainability of the cropping system in the midterm.
25.3 Within-Season Methods for Improved Nitrogen

Management
The well-known spatial and temporal variability of processes affecting soil N

represents a relevant limitation on the accuracy of approaches based on the
N-balance, which is a preplant estimation of N needs for the whole crop cycle.
To overcome this limitation, within-season monitoring approaches have been
proposed based on the improvements of tools and methods for improving N
requirements estimation. These strategies measure the N status of the soil or the
crop during the season, assessing the need for additional N, or recommending
specific rates of supplemental N. They usually involve multiple N applications
and allow the change of fertilizer rates depending on actual measurements of soil or
crop N status. Thus, these methods involve a dynamic strategy able to adapt to
environmental conditions during the crop season.
The most classical method is the pre-sidedress nitrate test (PSNT), performed for
corn in the USA. It is a type of Nmin method based on a soil nitrate analysis (30 cm
depth) when plants are 20–30 cm tall. It represents a point-in-time assessment of the
spring accumulation of NO3 before the crop begins the rapid growth phase. The
PSNT method takes into account the remaining residual N from the previous crops,
N mineralization and N losses before the date of sampling, just before large
amounts of N are required by the crop.
In addition to the methods based on soil available N, there has been a major
development in sensors to determine crop nutritional N status to adjust fertilizer
rates during growing season. The sensors are based on the determination of
transmittance or reflectance of a leaf or the crop canopy at various wave-lengths
and can be hand-held or tractor mounted. If other growth factors are known,
readings can be related with chlorophyll activity and therefore with crop N status.
Comparing with a well fertilized band a sufficiency index (ratio of crop reading/
well fertilized crop reading) can be developed and used to apply variable rates.
25.4 Fertilization of Fruit Trees
25.4.1 Mature Orchards
Mature trees are very efficient in translocating N to reserves (e.g. before leaf fall in
deciduous species) which will be later made available for new growth. Therefore in
mature orchards the calculation of crop N should only consider the amounts of N
exported in yield, or lost by pruning and leaf fall:
N c ¼ Y CN fruit þ Bpruning CN shoots þ Bleaf fall CN senesced ð25:9Þ
Values of N concentration in fruits are presented in Table 24.1. The concentra-

tions for shoots and senesced leaves may be taken as 1 %. The amount of leaf fall
will be equal to total leaf biomass for deciduous trees and around 50 % of leaf
biomass for evergreens with leaf life span around 2 years. Note that this is a
conservative estimate of crop N uptake as N in pruning residues is only lost when
they are burned. The total vegetative biomass production in the growing season (Bv)
is the sum of those in pruning, senesced leaves and growth of permanent structures
(trunk, main branches). Then, using the definition of harvest index we may write:
1 HI
Bpruning þ Bleaf fall ¼ βpl Y ð25:10Þ
HI
where βpl is the fraction of Bv not used in permanent structures. This parameter is
very high (0.8–0.9) for most deciduous species and for evergreens under intensive
management. Therefore, now we can simplify Eq. 25.6 to:

1 HI
N c ¼ Y CN fruit þ βpl CN pl ð25:11Þ
HI
where CN pl is the average concentration of N in pruning residues and senesced

leaves, that may be taken as 1 %. Data on HI of fruit crops indicate that it is usually
above 0.5, so if no information is available for a given species we may take a value
of 0.6.
To calculate the amount of fertilizer to apply we need to consider the fate of the
pruning residues. If they are burned or exported only leaf fall remains on the soil
which will give back mineral N after mineralization some time later. In the best
case, with zero losses of N from the system, we will recover all N from senesced
leaves. Therefore, the minimum amount of fertilizer to apply if residues are burned
or exported will be:

1 1 HI Y 1 HI
Nf ¼ N c 0:5Yβpl CN pl ¼ CN fruit þ 0:5βpl CN pl
Ef R HI Ef HI
ð25:12Þ
If residues are incorporated then the minimum amount to apply reduces to Y CN

fruit/EfR.
Example 25.2 An irrigated vineyard yields 25 t/ha of table grape. Water

content is 80 % (Table 24.1) so yield is 5 t dry matter/ha. N concentration in
fruits is 0.6 % (dry matter basis). Now, using Eq. 25.9:

0:4 N
N c ¼ 5000 0:006 þ 0:9 0:01 ¼ 60 kg
0:6 ha
Assuming Ef ¼ 0.8, if pruning residues are incorporated then the minimum

fertilizer amount would be 37.5 kg N/ha. If residues are exported we should
(continued)

apply at least 56 kg N/ha. In any case fertilizer amounts should not exceed
60/0.8 ¼ 75 kg N/ha. A well established cover crop (e.g. legume or legume/
grass mixture) could fix enough N to supply or reduce vineyard requirements.
A fraction of the N content in the residues would be available for the vineyard
during the growing season after the cover crop is mowed (50 %) or soil
incorporated (70 %).
25.4.2 Young Trees
For young orchards we have also to include the demand of vegetative growth,
which depends on age, species and environmental conditions. It is not easy to
calculate the increase in standing biomass of young trees. For some species
empirical relations have been established between tree biomass and trunk diameter.
A more general and simple approach is to relate canopy growth to transpiration
using the Water Use Efficiency, extending Eq. 14.8 to the whole growing season:
X
T
αw
ΔB ¼ Ep tree ð25:13Þ
0
VPD
where T is the duration of growth, while transpiration is proportional to ET0, the

relative intercepted radiation (fIR) and the transpiration coefficient for full ground
cover (Ktf) (see Eq. 9.20). Therefore:
X
T
αw
ΔB ¼ ET 0 RRi K tf ð25:14Þ
0
VPD
Finally we calculate the N uptake (Nc, g N/tree) required for that increase in
biomass assuming a high N concentration (2 %) which is an upper boundary for
biomass of young trees:
X
T
αw
N c ¼ 0:02 ET 0 RRi K tf ð25:15Þ
0
VPD
For young trees the amount of fertilizer to apply should match the expected uptake,
using a proper value of efficiency.
Example 25.3 In example 3.5 we calculated the relative interception of an

olive tree with radius 0.5 m in Cordoba, Spain on 21 March as RRi ¼ 0.69 m2.
If ET0 is 3 mm day1, taking a value of αw ¼ 7.5 g kPa L1 (Chap. 14) and
Ktf ¼ 1, with VPD¼1.5 kPa:
X
T
αw 6:5 gN
N c ¼ 0:02 ET 0 RRi K tf ¼ 0:02 3 0:69 1 ¼ 0:18
0
VPD 1:5 tree
Assuming E ¼ 0.8, the amount of N fertilizer should be 0.18/0.8 ¼ 0.23 g

N/day/tree.
A similar approach for fertilizing young trees with N may be taken for P
and K by applying the proper concentrations. For deciduous trees we may use
0.2 % of P and 0.9 % of K. For evergreen trees the values are 0.12 % P and
0.9 %K.
Bibliography
Brady, N. C., & Weil, R. R. (2008). The nature and properties of soils (4th ed.). Upper Saddle
River: Pearson Education.
Scheppers, J. S., & Raun, W. R. (2008). Nitrogen in agricultural systems (Agronomy monograph
no. 49). Madison: ASA, CSSA & SSSA.
Chapter 26
Fertilization with Phosphorus, Potassium
and Other Nutrients
Antonio Delgado, Miguel Quemada, Francisco J. Villalobos,

and Luciano Mateos
Abstract Phosphorus and potassium are important nutrients whose management

has some common characteristics since both are non-mobile nutrients in the soil.
Most of P and K in the soil is not available to plants. Reactions involved in their
cycle in the soil imply that only part of applied nutrients remain available for plants
(retrogradation). Adequate fertilizer management must be designed for maximum
recovery of applied P and K, which means that retrogradation is not favored. To this
end, alternatives to basal broadcast application should be taken into account, such
as banding or fertigation applications. Since both nutrients are non-mobile the risk
of leaching is very low. Therefore, management strategies for both nutrients may be
designed at medium- and long-term periods, contrasting with N, whose manage-
ment strategies were designed for short-term periods for each crop. The achieve-
ment of high efficiencies in P and K fertilization is gaining interest since both
nutrients are considered non-renewable resources. An example of what could
happen with the progressive depletion of reserves was the “P crisis” in 2008,
when the price of phosphate rock was multiplied by three due to the shortage of
exportations from China.
Other nutrients are usually applied when deficiency is expected (sufficiency
strategy). Deficiencies in Ca and Mg can be expected in acidic soils; Mg
A. Delgado (*)
Sevilla, Spain
M. Quemada
Madrid, Av. Puerta de Hierro, n 2 28040 Madrid, Spain
F.J. Villalobos
L. Mateos

Agriculture, DOI 10.1007/978-3-319-46116-8_26
deficiencies can be also promoted by high Ca saturation in the exchange complex or

by excessive K fertilization. Micronutrient availability is not only limited by the
amounts in the soil, but also by soil properties affecting plant root functioning in
acquiring and absorbing these nutrients.
26.1 Introduction: Phosphorus and Potassium as Essential

Nutrients
Phosphorus and potassium are the two primary nutrients which are considered
non-mobile in the soil (see Chap. 2). Both nutrients applied as fertilizer are quickly
fixed in the soil through different process. This means that, in contrast to N,
leaching risk is not significant in most soils, except in those with very low fixing
capacity (e.g. very sandy soils). On the other hand, retention reactions in soils imply
that only a fraction of applied fertilizer remains available for plants. As for N,
fertilizer management must be focused on achieving the maximum efficiency in
applied fertilizer. To this end, in the case of N it was very relevant to reduce the risk
of loses, mainly through leaching; in the case of P and K, fertilizer management
must be focused on minimizing the fraction of applied nutrient that is finally fixed
in the soil as non-available forms.
Under a non-limiting N supply, in terrestrial and aquatic systems, P is usually the
limiting nutrient for plant growth. Phosphorus is involved in many biological
processes, including relevant structural functions as part of nucleic acids or phos-
pholipids in membranes. It has also a key role in metabolic reactions, particularly
those involved in energy transfer (phosphorilation reactions). Plants absorb phos-
phorus actively and only in the forms H2PO4 and HPO42, with the second
showing much slower absorption rates.
Phosphorus deficiency has a significant impact on crop growth. Given the high
mobility of P within the plant, deficiency symptoms are first detected in older leaves
that senesce prematurely. In P-deficient plants, purple color in leaves due to the
accumulation of anthocyanic pigments is usual. Normal concentrations in leaves
vary greatly among species in the range of 0.05–0.3 % P (dry matter basis)
(Table 26.1).
High P fertilizer rates can lead to an enrichment of the soil in this nutrient. This
can have adverse consequences not only for crops, such as the increased risk of Zn
and Fe deficiencies, but also on the neighbor ecosystems. Excessive enrichment in P
of agricultural soils may promote high loses of this nutrients to water bodies mainly
bound to eroded particles. High P concentration in surface water triggers the growth
of algae and weeds which results in the eutrophication of surface waters. Therefore
soil erosion may have a large effect on P losses from the soil and pollution of
surface water by the transported sediment.
26
Table 26.1 Critical concentration and sufficiency range for P, K, Ca and Mg in leaves or shoots of crops
Nutrient concentration (% over dry mass)
P K Ca Mg
Crop Stage Critical Adequate Critical Adequate Adequate Adequate
Cotton 45 DAS >0.4 >3.2 0.65–0.8
Faba Flowering 0.19–0.24 0.3–0.55 1.8–2 2.2–4 0.6–1.2 0.24–0.5
Linseeda 53 DAS 0.37–0.69 2.5–3.5 0.96–1.7 0.36–0.65
Oat Tillering 0.24–0.29 0.3–0.5 4.3–4.9 5–5.7 0.21–0.4 0.13–0.3
Ricea Tillering 0.37–0.55 1.6 1.6–3 0.1–0.30 0.14–0.21
Sugar beet 50–80 DAS <0.45 0.45–1.1 2 2.0–6.0 0.5–1.5 0.25–1
Wheat Tillering <0.35 0.35–0.49 <2.3 2.4–4 0.21–0.4 0.13–0.3
Apple Summer 0.10–0.14 0.15–0.2 0.8–1.1 1.2–1.5 1.1–2.0 0.21–0.25
Apricot Summer 0.09–0.13 0.14–0.25 1.0–1.9 2.0–3.5 2.0–4.0 0.30–0.85
Citrus spp. 0.09–0.11 0.12–0.16 0.4–0.69 0.7–1.5 3.0–6.0 0.26–0.60
Fertilization with Phosphorus, Potassium and Other Nutrients
Olive Summer 0.1–0.3 0.4–0.8 >0.8 >1.0 >0.1

Peach Summer 0.09–0.13 0.14–0.25 1.0–1.9 2.0–3.0 1.8–2.7 0.30–0.80
Pear Summer 0.10–0.13 0.14–0.20 0.7–1.1 1.2–2.0 1.5–2.2 0.3–0.5
DAS days after sowing
a
Whole shoot
383
Potassium is absorbed actively as an ion. The absorption is controlled by the

internal concentration. Once absorbed it is transported mainly to young growing
tissues. Potassium has a role as osmoregulator and is involved in stomatal control. It
indirectly promotes photosynthesis and transport of assimilates, and has a direct
action on the activity of some important enzymes. Potassium is also involved in
plant tolerance to cold and water stress.
Symptoms of potassium deficiency are unclear and when they appear, the crop
has already suffered a negative effect. Being mobile, the deficiency symptoms
appear first on older leaves. Table 26.1 provides guidance on the usual and
threshold concentrations of K in leaves of different crops.
Phosphorus and potassium are considered as non-renewable resources because
their fertilizer production comes ultimately from mining. This contrasts with the
case of N because the atmosphere is a near-infinite source of this nutrient for
fertilizer production. Phosphorus and potassium reserves are finite and their
demand will increase in the next decades due to the need of increasing agricultural
production to meet the requirements of an increasing population in the World.
Under this perspective, prices of P and K fertilizers are expected to increase sharply
in the next future. Beside this, P can be considered a strategic resource since most of
the known reserves are in only three countries (USA, Morocco and China). Thus,
the need of a rational management of P and K is a basic requirement to ensure
sustainability of agricultural production in the next decades.
26.2 P in the Soil
Phosphorus is not a major element present in soil since it is not part of commonly
abundant soil minerals, its total concentration in the soil being usually less than 1 g/
kg. It is mainly present as phosphate, which can be found in organic (basically
esters) or mineral forms. Both mineral and organic forms can be found in the soil
solution or bound to the solid fraction. As indicated above, inorganic phosphate
(dissociate forms of orthophosphoric acid) is the form in which plants take up P; in
the soil solution it is in equilibrium in a more or less reversible way with specifically
adsorbed forms (see Sect. 2.4.3) on hydroxylated surfaces (Fe and Al oxides and, to
a lesser extent, borders of clay minerals and carbonates) and with precipitated metal
phosphates (mainly Fe, Al, and Ca phosphates, depending on soil pH). Thermody-
namically stable metal phosphates, such as apatite type in soils with high pH and
high Ca saturation, are insoluble thus contributing little to P in the soil solution;
other less stable precipitates can be present in a lesser extend which can contribute
more to P in solution (Fig. 26.1). Precipitation as insoluble metal phosphates and
adsorption reactions explain that a minor part of applied P as fertilizers remains
available to plants. Organic forms are mainly phosphate-monoesters and
phosphate-diesters, and they may be also adsorbed, sometimes more strongly than
inorganic phosphate, and they can precipitate as well. As in the case of N and other
elements, phosphorus can be used by soil microorganisms and be immobilized, at
26 Fertilization with Phosphorus, Potassium and Other Nutrients 385
Fertilizers Plants Residues

Dissolution Uptake Decomposition
Adsorbed P Mineralization
Specific on hydroxilated Adsorption (Hydrolisis) Organic P
surfaces Inorganic P in Adsorbed or in
Oxides, borders of clay solution (phosphate) solution
minerals, carbonates Desorption Inmobilisation Biomass
Precipitation
Dissolution Weathering
Metal Primary minerals

phosphates (mainly insoluble phosphates;
(precipitates) e.g. apatite)
Fig. 26.1 P cycle in the soil. In italics: physical, chemical or biological processes involved in
nutrients cycle
least temporarily. The reverse process, mineralization, requires the hydrolysis of

ester bounds which is catalyzed by different types of phosphatases, which can be
produced by plant roots and some soil microorganisms. Organic P may account for
a significant fraction of the total P in the soil (usually between 20 and 80 % in
agricultural soils) even in Mediterranean soils with low organic matter
concentration.
Phosphorus concentration in the soil solution is usually very low, typically less
than 1 mg/L in agricultural soils, forcing its replenishment from adsorbed or
precipitated forms as plants absorbs it. The relationship between adsorbed P and
P in the soil solution is determined by the P buffer capacity of soil, which depends,
among other factors, on the adsorption capacity of the soil (Fig. 26.2). A high buffer
capacity indicates a good ability of the soil to replenish P of the solution as it is
absorbed by the crop. If the buffer capacity is low, the concentration of P in the
solution may be high, but the soil’s ability to replenish that used by the crop is
limited. On the other hand, the P supplied as fertilizer initially passes to the soil
solution, but will shift to fixed forms more the higher the buffer capacity. This
means that for restoring a deficiency of P in the solution we need to provide higher
doses of fertilizer the higher the soil buffer capacity.
The degree of ionization of phosphates depends on the pH (Fig. 26.3). In acid
soils monovalent ions predominate while in neutral soil the ratio between mono-
valent and divalent forms is about 1:1. Plants absorb only H2PO4 and HPO42, and
the latter much more slowly. Therefore, the absorption of P is faster in acid soils.
Plants can secrete organic acids to the rhizosphere which release adsorbed P and
dissolve precipitated P, and phosphatases that contribute to mineralize organic
P. Besides this, the interaction between plants roots and microorganisms in the
rhizosphere is very relevant in P uptake by plants because microorganisms can
release organic acids and phosphatases. Symbiotic mycorrhizae can also enhance P
uptake by plants by increasing the absorption surface for P.
Fig. 26.2 Consequences of P buffering capacity in soils: A soil with high buffering capacity; B
(dotted line) with low buffering capacity; red arrow represents the decrease in P concentration in
solution for soil B, and blue arrow the decrease in soil A for the same decrease in adsorbed P
Fig. 26.3 Phosphoric acid PHOSPHORIC ACID SPECIATION

speciation 100
H2PO4– HPO42–
%FORMATION RELATIVETO PO4
80
60
40
20 PO43–
H3PO4
0
2 6 10 14
pH
26.2.1 Available P to Plants in the Soil: The Soil P Test
Labile phosphorus in the soil is that in the solid phase that may be easily released to
the soil solution. It is generally accepted that available phosphorus is the sum of P in
the soil solution and P in labile forms (see Sect. 2.6). Available P in a soil usually
accounts for a minor fraction of total P (frequently less than 10 %). Accurate
determination of available P can be done only by biological methods measuring
the amount extracted by successive crops until evident deficiency in plants appears
(P starvation assays). This is not practical for estimating fertilizer requirements.
However, chemical or biological methods can provide results which are well
correlated with those obtained with P starvation assays. The methods can be
considered as P availability indexes or soil P tests, the most usual being those
involving chemical extraction of P from soil (Table 26.2).
The soil P test is not a real measure of the amount of available P but can provide
information about if the available pool in the soil is enough to cover crop needs.
Soil P tests are crucial for P fertilizer management, but for practical use, critical
values or threshold values for each soil P test should be defined, which is the value
above which no response in crop yield can be expected if P fertilizer is applied.
Thus, below this value, the soil can be considered deficient in P. Critical values can
vary between soil types and crops. Besides this, there is not a universal soil P test
since its efficiency for fertilizer requirement estimation is very affected by soil
properties. This explains why only in Europe more than 6 official P indexes are used
depending on the country: Olsen and lactate (ammonium or Ca) extractions are the
soil P tests more widely used in Europe, while in North America, Olsen, Mehlich
(I and III) and Bray tests are the more usual.
26.3 Potassium in the Soil
Most of the potassium in soils is found as a component of feldspar and mica

minerals, common primary minerals in soils which explain that potassium is an
abundant element in soils (frequently 0.3–3 % in mass). Potassium in primary
minerals is slowly released by weathering. After this release, potassium is found
in soluble, exchangeable, and non-exchangeable forms. Potassium in the soil
solution is in rapid equilibrium with exchangeable K that is retained by electrostatic
attraction to the negatively charged sites located on clays and soil organic matter.
Non-exchangeable K, also defined as “fixed” K, is found within the interlayers and
on the edges of 2:1 clay minerals and cannot be considered readily available to
plants (Fig. 26.4). A part of K supplied as fertilizer is fixed in this form explaining
that not all the applied K remains available to plants; the fraction of applied K that is
fixed is greater with increasing fertilizer rates and increased 2:1 clay content of the
soil. Adsorption as exchangeable or non-exchangeable forms explains why K
moves so little in the soil.
The available K to plants is basically that in the soil solution plus the exchange-
able pool. The potassium concentration in the solution is typically between 0.2 and
10 meq/L, which usually accounts for less than 1 % of exchangeable K and can be
quickly depleted by plant uptake. Potassium in cationic form in the solution is the
unique form that plants are able to absorb. As defined for P, soil buffering capacity
for K is the ability to replenish K in the soil solution from that bound to soil
388
Table 26.2 Different soil P tests

Threshold
levels mg
Method Extractant Conditions to be applied Countries Specific conditions P/kg
Olsen 0.5 M NaHCO3 pH 8.5 Slightly acidic to basic soils UK, part of USA, Rainfed crops in semi-arid 4–12
Mediterranean EU areas on clay soils (mean 8)
countries Rainfed crops in Mediterra- 8–10
nean areas
Rainfed crops in humid 4–5
areas and medium texture
soils
Spring cereals in humid 8–10
areas in medium texture
soils (moderate yield)
Winter cereals in humid 15–20
areas in medium texture
soils (high yield)
Sugar beet 15–20
Corn and wheat in sandy 8–16
loamy soils (China)
General recommendation 8–10
clay soilsa
loamy soilsa
sandy soilsa
Mehlich I 0.05 M HCl þ 0.0125 M H2SO4 Multinutrient extraction for soils Part of USA General 25
with low CEC
A. Delgado et al.
26
Mehlich III 0.2 M CH3COOH þ 0.25 M Multinutrient extraction for wide Part of USA General 50
NH4NO3 þ 0.015 NH4F þ 0.013 M range of soils; well correlated with
HNO3 þ 0.001 M EDTA – pH 2.5 Bray I, Mechlich I and Olsen
Bray I 0.03 N NHF4 þ 0.025 N HCl Soils with acidic and neutral pH Acidic soils in General 30
(Bray- some countries
Kurtz) (e.g. USA)
Bray II 0.03 N NH4F and 0.01 N HCl Soils with acidic and neutral pH General 10–20
Egnér et al. 0.1 M ammonium lactate þ 0.2 M Multinutrient extraction Scandinavia, Cen- General 70–80
acetic, pH 3.75 tral and East
Europe, Portugal
Egnér 0.01 M Ca lactate þ 0.02 M HCl Multinutrient extraction Germany General 100
-Riehm
Morgan 0.7 M NaC2H3O2 þ 0.54 M Multinutrient extraction for acidic General 4–6
CH3COOH – PH 4.8. soils with low cation exchange
capacity
Morgan 0.62 M NH4OH þ 1.25 M Multinutrient extraction for acidic General 4–6
modified CH3COOH – PH 4.8 soils with low cation exchange
capacity
Soltanpour 1 M NH4HCO3 þ 0.005 M DTPA – Multinutrient extraction for P, cat- General 10 (similar
(AB-DTPA) pH 7.5 ions and micronutrients; mainly for to Olsen)
calcareous soils
Fertilization with Phosphorus, Potassium and Other Nutrients
Dyer- 2 % Citric acid Acidic soils General 90–180

Demolon
Schüller 0.05 N Ca lactate þ 0.05 M Ca Acidic soils General 80–140
acetate þ 0.3 M acetic
Truog 0.2 N H2SO4 Non-calcareous soils General 20
a
Multiply by 1.5–2 for high yield crops
389
Fig. 26.4 K, Ca, and Mg cycle in soil. In italics: physical, chemical or biological processes
involved in nutrients cycle
Table 26.3 Treshold levels (STLt) for ammonium acetate extractable K in soils depending on the
cation exchange capacity (cmol(þ)/kg) and on the texture
Cation exchange capacity (CEC) Threshold levels Texture Treshold levels
cmol(þ)/kg mg K/kg soil mg K/kg soil
10 150 Sandy 100
20 180 Loamy 150–175
30 210 Clay 200–300
40 240
The thresholds can be also calculated using STLt ¼75 þ 2.5 CEC (Tri-state recommendation for
corn, wheat, soybean and alfalfa) or STLt ¼ 110 þ 2.5 CEC (general recommendation)
particles, and it is critical to maintain K concentration in the solution in appropriate

ranges for plants. Potassium buffer capacity mainly depends on clay content and
mineralogy, being the highest in clay soils with 2:1 as major clay minerals.
26.3.1 Plant Available Potassium in the Soil
As for P, K fertilizer management must be based on availability indexes. Although

several indexes have been proposed, the most common are those based on the
estimation of exchangeable K. The most widespread method is extraction with
ammonium acetate solution at pH 7 (Table 26.3). This method has generally shown
a good correlation of crop response with soil potassium content. Other methods to
determine exchangeable K, such as buffered BaCl2 or unbuffered NH4Cl provide
similar results. For acidic soils, methods based on extraction with acids
(e.g. Mehlich or Morgan) have been proposed, generally with good correlations
with amounts extracted with ammonium acetate.
26.4 Response of Crops to P or K Fertilizer
The production functions of yield versus applied nutrient (P or K) show similar

responses with decreasing slope (Figs. 26.5 and 26.6). The relationship between
yield and P uptake is similar to that described for N, i.e. more or less straight
initially and then it may reach a ceiling (Fig. 26.7). The relationship between dose
of fertilizer and P uptake is initially linear and can also reach a ceiling where the
absorption is saturated (Fig. 26.7). The intercept with the Y axis of the linear
portion indicates the amount of P the crop may absorb from sources other than
fertilizer (P previously present in soil) and the slope is the recovery efficiency
(or fertilizer efficiency) in the straight section of the relationship.
The relationship between yield and applied K is of the same type as those
described for N and P (Fig. 26.6). The examples shown in Figs. 26.5 and 26.6
indicate that these relations described for P and K may vary with N fertilizer
applications or soil management system. This reveals that nutrition and fertilization
for a given nutrient cannot be considered isolated. Synergistic effects with other
nutrients, such as N, explains that accumulation of P and K can be faster if an
appropriate N supply is performed; this also implies better response in yield to P or
K supply. Adequate water supply also implies better responses of crops to P and K
supply in terms of uptake and yield since a more efficient uptake of nutrients is
achieved. Soil water content is critical in explaining nutrient flux to the roots,
Fig. 26.5 Wheat grain 3.5

yield response to P fertilizer
application with increasing
3.0
rates of N fertilizer in
Australia (Adapted from
2.5
GRAIN YIELD (t ha–1)
Brennan and Bolland (2009)

Crop pasture sci
60:566–577)
2.0
1.5
1.0 Yield 0 kg N/ha
Yield 20 kg N /ha
0.5
Yield 80 kg N /ha
0.0
0 10 20 30 40
P APPLIED (kg ha–1)
Fig. 26.6 Rice grain yield 16

response to K fertilizer
application with increasing 15
rates of N fertilizer in
14
Arkansas (Adapted from
GRAIN YIELD (t ha–1)

Slaton et al. (2010) Rice 13
Response to Nitrogen and
Potassium Fertilization 12
Rate. Arkansas Agric
Extension Service Res 591. 11
available at http:// 10
arkansasagnews.uark.edu/
591–38.pdf) 9
8 Yield 88 kg N/ha
7 Yield 176 kg N/ha
6
0 50 100 150 200
K APPLIED (kg ha–1)

maize biomass production
to P uptake and of P uptake
to P applied for different
levels of N supply
particularly in those that basically move through diffusion mechanisms due to their
low concentration in the solution, such as P and K. The key role of water content
explaining nutrient uptake by plants explains that in Mediterranean dryland areas
(annual rainfall 300–500 mm) threshold Olsen values can vary from 4–5 in rainy
years to 8–9 mg/kg in dry years, reflecting the need of more P present in the soil for
optimal supply to plants in dry years. These examples justify the need of adequate
and integrated management practices to achieve the better crop response to inputs
supplied.
26.5 Phosphorus and Potassium Fertilization Strategies
The principles for establishing the P and K fertilization plan based on soil-test
interpretations vary across countries. In the cases of the sufficiency approach, the
main objective is maximum short-term profitability from applied fertilizer and
minimum risk of environmental impact related to excess fertilizer by accepting
some risk of yield loss. On the other hand, the buildup and maintenance approach
seeks long-term profitability from fertilization, long term maximum returns, and
reduced risk of yield loss due to low fertility.
Due to complex reactions of P and K fertilizers in soils, soil test above threshold
values is the only guarantee of optimal P or K supply to plants since available
reserve of nutrient in the soil is enough for an optimal supply (no fertilizer
response). Below threshold values, there is not guarantee of optimal P or K supply
to crops due to the uncertainty on the fraction of applied fertilizer that can finally
remain available to plants. Thus, according to both strategies of fertilization, an
optimal condition of soil from a point of view of nutrient supply is to stay above
threshold values (or critical values) defined for the recommended soil test. Soil tests
and threshold values can vary widely for P depending on the region/country and
depending on soil properties. In general, the threshold value will depend on the
particular requirement of each crop (high or low P or K extractions), on the clay
content of the soil for P and K, and on carbonate content for P.
26.5.1 Buildup and Maintenance Approach
According to this strategy, if the soil test level (STL) is below the threshold value
(STLt), the fertility level of the soil should be corrected to bring it up to slightly
above this value, according to the references shown in Tables 26.2 and 26.3 for P
and K, respectively. The increase in P or K availability till optimal levels takes
usually several years, being clearly slower if soils have a high fixation and buffering
capacity. It should be remarked that the buildup rate for a given supply strategy
could vary from soil to soil since P and K reactions in the soil cannot be predicted
accurately.
After reaching an adequate level of fertility above the threshold value, fertilizer
applications should compensate removal of nutrients by the previous crop. Nutrient
exports are calculated as the product of the quantity of biomass going out of the
field and its nutrient concentration (Table 26.4). Soil analysis should be performed
every 3-4 years to check the evolution of the level of fertility and varying the
fertilizer rates according to the evolution of soil test levels.
A single formulation of the buildup and maintenance strategy could be:
STL<STLt Add crop exportation þ more (according to models)

STLt<STL<2 STLt Add only exported nutrient (or slightly above exportations)
STL>2 STLt (Maintenance limit) Add less P than crop export (e.g. 50 %) or do not fertilize
Table 26.4 Average phosphorus and potassium concentration (% dry weight) in different
harvested organs and residues for different species
Concentration (% dry matter) Concentration (% dry matter)
Part not
Crop species Part harvested P K harvested P K
Alfalfa (hay) Biomass 0.26 2.10
Apple Fruit 0.05 0.75
Barley Grain 0.42 0.54 Straw 0.1 1.8
Bean (Phaseolus) (dry Seed (dry) 0.54 2.7 Straw 0.14 1.3
seed)
Cotton Fiberþseed 0.41 0.49 Residues 0.1 1.6
Lettuce Leaves 0.75 6.67
Maize (grain) Grain 0.32 0.34 Stover 0.1 1.5
Millet Grain 0.38 0.39 Stover 0.04 1.6
Olives (60 % canopy Fruit 0.14 1.25
cover)*
Orange Fruit 0.14 1.35
Palm Trees fruit bunch 0.09 0.75
Peach Fruit 0.12 1.55
Peas (dry harv.) Seed (dry) 0.48 1.3 Straw 0.3 1.2
Potato Tuber 0.25 2 shoot 0.2 3.95
Rapeseed, Canola Grain 0.62 0.98 residues 0.1 0.8
Rice Grain 0.29 0.28 Straw 0.09 1.5
Sorghum (grain) Grain 0.33 0.39 Stover 0.13 0.73
Soybeans Seed 0.66 1.5 Stover 0.06 0.57
Sugar Beet Root without 0.25 1.54 shoot 0.22 5.8
crown
Sugar Cane (virgin) Stalks 0.01 0.2 Leaves þ stems 0.07 0.12
Sunflower Grain 0.63 0.72 residues 0.14 2.52
Tomato Fruit 0.47 4.28 residues 0.1 1.9
Winter Wheat Grain 0.37 0.46 Straw 0.06 1.2
A more complete list is provided in Appendix 26.1
*This canopy cover is assumed to be that of fully developed trees
Table 26.5 Correction factor Soil texture f range

f for buildup strategy in K
Sandy and sandy-loam 1.1–1.2
fertilization
Loam and silty-loam 1.5–1.7
Clay-loam 2
Clay 2.5–5
For the case of P, buildup models to estimate P fertilizer rate for soil test values
below threshold values could be based on the Tri States fertilizer recommendation:

P 10ρb Z
P rate kg ¼ Exported P þ ðSTLt STLÞ ð26:1Þ
ha N year
STL and STLt are given in mg/kg; ρb is bulk density (t/m3), Z is soil depth to correct
(m). P rate should be less than 100 kg P ha1. Nyear ¼ 1 if P rate <100 kg P ha1;
Nyear should be considered >1 to achieve a final P rates always lower than 100.
In soils with a high P fixation capacity, the application of crop exportations in the
case of STLt<STL<2 STLt could lead to a decrease in STL in the long-term. In this
case, P applications above crop exportations could be recommended, e.g. between
10 and 30 % more than exportations depending on the capacity of soil to fix P.
In the case of K, a buildup equation will be based on the desired increase in soil
K (STLt – STL) which is corrected by a factor fK that depends on the K interlayer
fixing capacity of soils (Table 26.5).

K 10ρb Z f K
K rate kg ¼ Exported K þ ðSTLt STLÞ ð26:2Þ
ha N year
where ρb is bulk density (t/m3), Z is soil depth to correct (m) and Nyear is the number
of years to reach STLt. Nyear ¼ 1 if K rate <275 kgK/ha; N year should be
considered >1 to achieve final K rates always lower than 275 kgK/ha
Buildup for P and K is usually achieved in several years and periodic control of
soil levels (at least every 3 years) are necessary to check when the critical values are
achieved. Massive applications of P and K are less effective, in terms of the ratio
nutrient available to total amount applied. Thus, massive applications allow to
achieve STLt in shorter times, but with a total consumption of fertilizers (and
subsequent cost) greater than with more fractionated applications. This is the reason
why in some areas (e.g. some states in the USA) less than 100 kg P/ha and less than
275 kgK/ha are usually recommended as total rates in a buildup and maintenance
strategy. This is the limit stablished to recommend 1 or more years in the Eqs. 26.1
and 26.2.
Above a certain level, which can be defined as “maintenance limit”, rates lower
than crop exportation (e.g. 50 %) or no fertilizer are applied. The maintenance limit
can vary depending on the recommendation, but in general a reasonable recom-
mendation could be to consider it as twice the threshold value.
Example 26.1 The P concentration in a soil determined by the method of

Olsen in the top 25 cm is 7 ppm while the K concentration is 100 ppm
(ammonium acetate method). In this soil, maize is going to be grown with
an expected yield of 15 t/ha (14 % moisture). Threshold values for P and K are
15 ppm and 250 ppm, respectively. Soil bulk density is 1.4 t/m3. Texture is
clay (45 % clay) and carbonate content 30 %. Calculate P and K fertilizer
rates for these conditions. Calculate also if Olsen P is 27 ppm and acetate
extractable K is 300 ppm.
Fertilization based on Buildup and Maintenance strategy
Crop extraction and exportation; the amount of dry matter exported is:
15000 kg=ha ð1 0:14Þ ¼ 12900 kg grain dry matter=ha
Now, according to Table 26.4, P and K concentrations in maize grain are

0.32 % and 0.34 %, thus, the amount of P and K removed in grain are:
12900 kg=ha 0:0032 kg P=kg ¼ 41:25 kg P=ha

12900 kg=ha 0:0034 kg K=kg ¼ 43:8 kg K=ha
Now we use Eq. 26.1:
P rate ¼ 41:25 þ 10 1:4 0:25 ð15 7Þ ¼ 69:3

69 kg P=ha ð340 kg triple super phosphate=haÞ
If Olsen P is 27 ppm, as it is below the maintenance limit (30 ppm), we may

apply only the exported nutrient:
P rate ¼ 41:25 41kg P=ha ðe:g: 205 kg TS=haÞ
For potassium:
Increase in K ¼ 250 150 ¼ 100 ppmðmg=kgÞ
From Table 26.5 for clay soil we have a range of fK between 2.5 and 5, and
we choose fK ¼ 3.
K rateðkg K=haÞ ¼ 43:8 þ 10 1:4 0:25 3 ð250 150Þ ¼ 1050 þ 43:8

¼ 1093:8 kg K=ha
much higher than 275 kg/ha, then Nyear>1. It is easy to check that to avoid
exceeding 275 kg K/ha in this case we need to assume Nyear ¼ 5, so:
K rateðkg=haÞ ¼ 48:3 þ ð10 1:4 0:25 3Þ=5ð250 150Þ ¼ 258:3kg K=ha
which may be supplied with 258.3/0.5¼516.6 kg potassium chloride
(continued)

In any case, as commented above, this is not a precise estimate and
periodic checks of K status of the soil are recommended to know when the
threshold level in soil has been reached.
If acetate extractable K is 300 ppm, as it is above the threshold and below
the maintenance limit, then we should compensate only the exported nutrient:
K rate ¼ 43:8 kg K=haðe:g: 87:6 kg Cl K=haÞ
26.5.2 Sufficiency Approach
If initial fertility is below the threshold value, this strategy would begin by buildup
applications which may be applied in several years as in the case of the build up and
maintenance strategy. Once the soil test is above threshold values, it is tested every
year, and fertilizer is only applied when the nutrient level is below this value. The
sufficiency approach was initially recommended to avoid environmental problems
derived from an excessive P enrichment of soil which leads to high P concentration
in water bodies triggering eutrophication effects. Beside this, it has another advan-
tage: it promotes an increased use of residual P and K in the soil, i.e. a progressive
starvation of P and K levels in soils may enhance the transformation of
non-available forms to available forms and thus an enhanced use by plants of
non-available forms. In any case, this can be understood as a depletion of soil
nutrient reserves which in the long-term may imply the application of increased
rates of fertilizer.
26.6 Other Nutrients
The application of other nutrients different from N, P and K is usually done

following a sufficiency strategy: they are applied if their deficiency is expected.
Calcium and Mg have a similar cycle to that described for K (Fig. 26.4). Available
amounts are also equivalent to nutrient in solution plus exchangeable pool. All
extraction methods used as soil test for K can also be useful as Ca and Mg
availability indexes. Threshold values for Ca and Mg according to the ammonium
acetate extraction are 250–500, and 30–60 mg/kg, respectively. The threshold
values increase in proportion to cation exchange capacity of the soil. In general
terms, low availability levels of Ca are usually found found in acidic sandy soils
with low base saturation.
Fertilization with K, Ca or Mg must take into account the available levels of the
other alkaline or alkaline-terreous nutrient levels since antagonistic effects can be
Table 26.6 Availability index (soil test) for micronutrients and threshold values
Threshold
values
Micronutrient Method (mg/kg) Conditions of use
Boron Hot water 0.1–2
Copper Ammonium 0.5–2.5 AB-DTPA y DPTA recommended in
Bicarbonate-DTPA soils with basic pH
(AB-DTPA Sultanpour)
DTPA 0.1–2.5
Iron AB-DTPA 4.0–5.0 Soils with basic pH
DTPA 2.5–5.0
“Fast ammonium 350–900 Basic soils; threshold values depend
oxalate” on crop sensivity to Fe chlorosis
(e.g. 350 olive and grapevine)
Non-buffered 10 Soils with basic pH; threshold value
hydroxylammonium defined for sensitive crops to Fe defi-
ciency chlorosis
Manganese AB-DTPA 0.5–5.0 Soils with basic pH
DTPA 1.0–5.0
Molybdenum Oxalato amónico pH 3.3 0.1–0.3
Zinc AB-DTPA 0.5–1.0 Soils with basic pH
DTPA 0.2–2.0
promoted. Exchangeable K/Mg ratios above 0.5 can induce Mg deficiency, and
values lower than 0.1 promote K deficiency; on the other hand Ca/Mg ratios above
10 may promote Mg deficiency, while when they are lower than 2, Ca deficiency
may appear.
Micronutrient availability to plants is not only determined by its amount in the
soil. In the case of metals (Fe, Cu, Mn and Zn), it is clearly affected by soil
conditions affecting their solubility and the ability of plants to mobilize, absorb
and transport them. The paradigmatic case is Fe, a fairly abundant element in soils
whose availability is clearly decreased in calcareous soils. Iron deficiency, known
as Fe deficiency chlorosis, is the consequence of insolubility of Fe compounds at
basic pH and the failure of plant mechanisms to mobilize Fe from soil, to absorb, or
to transport it across cell membranes. Iron chlorosis is a relevant agronomic
problem in agriculture production on calcareous soils.
The most usual soil test for micronutrients is the extraction with the chelating
agent DTPA (Table 26.6.). For Fe, extraction based on ammonium oxalate usually
provides better results than DTPA extractions. In the case of Fe, active calcium
carbonate has been also usually used as an index to predict its deficiency.
26.7 Timing and Fractioning of P and K Fertilizer

Applications
Phosphates have traditionally been used as basal fertilizers before sowing, but can
also be applied at planting which is recommended under no-till by using proper
machinery for sowing and fertilizer application. Under no-till, incorporation con-
tributes to decrease the risk of incidental P losses (e.g. by unexpected rain after
fertilization) and to avoid enriching only the surface layer which would facilitate
losses by runoff or erosion. The application should be performed with more
anticipation the lower the solubility of fertilizer (not less than three months before
planting for poorly soluble such as rock phosphate).
Fractionation of P fertilizer is an uncommon practice. It can be justified in very
sandy soils saturated with P to avoid leaching. It can be also recommended because
it has been found that massive contributions favor retrogradation (conversion to
non-available forms) and decrease the recovery efficiency. Therefore, in soils very
low in phosphorus, or high in calcium or for crops with high P demand, fraction-
ation with soluble P fertilizer can be considered. The main limitation for P fertilizer
fractionation, as it is done with N, comes from its low mobility in the soil; topdress
broadcast P fertilization would lead to an enrichment in P of the soil surface,
increasing P loss risk and constraining root growth in depth. Thus, it seems more
feasible the fractionation of P fertilization when fertilizer can be incorporated into
the soil, particularly close to the root system. Fertigation can meet these require-
ments; this application method not only involves a high fractionation of P fertilizer,
but also location near the roots and watering conditions that favor P movement to
roots. All this results in a much higher efficiency of applied P with fertigation than
with traditional basal applications.
As P shows low mobility in the soil, it can be recommended to locate it close to
the roots by band application at sowing, particularly in low-P soils, to enhance its
use by plants. Localized application can help in reducing retrogradation by satu-
rating the fixation capacity of a more reduced volume of soil, puts the fertilizer
closer to the roots, and usually promotes an early growth of crops (“starting”
effect). The amount of P that can be banded is not limiting except if P is applied
as ammonium phosphate since ammonium can be phytotoxic at high rates; in this
case, less than 40 kg N/ha as ammonium should be applied. Banded fertilizer at
sowing must be located in bands 5 cm to the side and 5 cm below the seed to avoid
germination problems of seeds due to high salt concentration.
Application of P as organic amendments/fertilizer or with organic matter should
be also considered for improved efficiency. Organic matter competes with P for
adsorption sites and decreases the precipitation as insoluble metal phosphates thus
clearly enhancing the recovery of applied P. This explains why greater improve-
ments of soil P test have been found with manure applications when compared with
soluble inorganic fertilizers applications. The application of P as organic
by-products is also gaining interest nowadays as a P recycling strategy to make
agriculture less dependent on a non-renewable resource.
Strategies considered for P can be also practical for K fertilizer which is also
usually applied before or at sowing. Fractionation and located application in bands
are good choices since they contribute to saturating the soil and keeping high
concentration in the soil solution. However, no improvement in recovery efficiency
can be expected with the joint application with organic matter, except in very sandy
soils, since the only contribution of organic matter to the soil K cycle is to provide
more charged surface and thus more retention capacity if clay content is very low.
Appendix 26.1. Average phosphorus and potassium

concentration (% dry weight) in different harvested organs
and residues for different species
% dry % dry
matter matter
Cereals & Pseudocereals Part P K Not P K
harvested harvested
Barley (2 row) Hordeum vulgare Grain 0.35 0.49 Straw 0.08 2.1
Barley (6 row) Hordeum vulgare Grain 0.42 0.54 Straw 0.1 1.8
Buckwheat Fagopyrum Seed 0.35 0.46
esculentum
Maize Zea mays Grain 0.32 0.34 Stover 0.1 1.5
Millet-Foxtail Setaria italica Grain 0.34 0.35 Stover 1.6
Millet-Pearl Pennisetum Grain 0.38 0.39 Stover 0.04 1.6
glaucum
Millet-Proso Panicum Grain 0.34 0.48 Stover 1.6
miliaceum
Oats Avena sativa Grain 0.36 0.44 Straw 0.1 2.3
Quinoa Chenopodium Seed 0.41 1.12
quinoa
Rice Oryza sativa Grain 0.29 0.28 Straw 0.09 1.5
Rice (milled) Oryza sativa Grain 0.3 0.45 Straw 0.09 1.5
Rye Secale cereale Grain 0.38 0.52 Straw 0.09 0.97
Sorghum Sorghum bicolor Grain 0.33 0.39 Stover 0.13 0.73
Triticale X Triticosecale Grain 0.34 0.57 Straw 0.03 1.2
rimpaui
Wheat- Spelt Triticum spelta Grain 0.42 0.44 Straw 0.13 1.4
Wheat-Bread-Hard Triticum Grain 0.43 0.45 Straw 0.06 1.2
type aestivum
Wheat-Bread-Soft Triticum Grain 0.37 0.46 Straw 0.06 1.2
type aestivum
Wheat-durum Triticum durum Grain 0.42 0.5 Straw 0.06 1.2
Grain Legumes Part P K Not P K
harvested harvested
(continued)
% dry % dry
matter matter
Bean Phaseolus spp. Seed (dry) 0.54 2.7 Straw 0.14 1.3
Chickpea (desi) Cicer arietinum Seeds 0.4 1.2 Straw 0.16 2.3
Chickpea (kabuli) Cicer arietinum Seeds 0.4 1.2 Straw 0.16 2.3
Cowpea Vigna Seed 0.52 1.5 Straw 0.28 1.55
unguiculata
Faba bean Vicia faba Seed 0.47 1.2 Straw 0.2 1.6
Lentil Lens culinaris Seed 0.43 0.86 Straw 0.14 1.15
Pea Pisum sativum Seed (dry) 0.48 1.3 Straw 0.3 1.2
Peanut Arachis Pods 0.35 0.56 Straw 0.14 1.38
hypogaea
Soybean Glycine max Seed 0.66 1.5 Stover 0.06 0.57
Forages Part P K Not P K
harvested harvested
Alfalfa (hay) Medicago sativa Biomass 0.26 2.10
Bluegrass-Ken- Poa pratensis Biomass 0.28 1.92
tucky (hay)
Bromegrass (hay) Bromus sp. Biomass 0.16 1.64
Canarygrass-Reed Phalaris Biomass 0.28 2.99
(hay) arundinacea
Clover (white) Trifolium repens Biomass 0.35 2.30
(hay)
Clover-Alsike (hay) Trifolium Biomass 0.25 2.48
hybridum
Clover-Crimson Trifolium Biomass 0.22 2.76
(hay) incarnatum
Clover-Red (hay) Trifolium Biomass 0.26 1.89
pratense
Clover-White Trifolium repens Biomass 0.35 2.25
(hay)
Clover-White- Trofolium Biomass 0.32 2.43
Ladino (hay) repens
Fescue-Tall (hay) Lolium Biomass 0.32 2.36
arundinaceum
Grass (hay) Poaceae Biomass 0.22 1.45
Grass (silage) Poaceae Biomass 0.32 1.88
Maize (silage) Zea mays Biomass 0.20 1.00
Millet-Foxtail Setaria italica Biomass 0.18 1.94
(silage)
Millet-Pearl Pennisetum Biomass 0.26 1.63
(silage) glaucum
Oat (hay) Avena sativa Biomass 0.24 1.26
Orchardgrass Dactylis Biomass 0.18 2.64
(green chop) glomerata
Orchardgrass (hay) Dactylis Biomass 0.25 2.80
glomerata
(continued)
% dry % dry
matter matter
Rye (hay) Secale cereale Biomass 0.22 1.24
Ryegrass-Perennial Lolium perenne Biomass 0.20 1.42
(hay)
Sorghum Sorghum bicolor Biomass 0.21 1.10
Sweetclover (hay) Melilotus sp. Biomass 0.24 1.65
Timothy (hay) Phleum pratense Biomass 0.17 1.63
Trefoil-Birdsfoot Lotus Biomass 0.23 1.89
(hay) corniculatus
Turnip (green chop) Brassica rapa var. Biomass 0.42 3.02
rapa
Vetch (hay) Vicia sativa Biomass 0.36 2.24
Vetch-Hairy (hay) Vicia villosa Biomass 0.36 2.23
Wheatgrass (hay) Poaceae Biomass 0.07 2.70
Sugar, oil & fiber Part P K Not P K
crops harvested harvested
Cotton Gossypium fiberþseed 0.41 0.49 residues 0.1 1.6
hirsutum
Flax Linum Seed 0.57 0.84 residues 0.08 1.74
ussitatisimum
Opium poppy Papaver Capsules 0.6 2.4 Leaves 0.3 3.1
somniferum þstems
Rapeseed Brassica spp Grain 0.62 0.98 residues 0.1 0.8
Safflower Carthamus Grain 0.6 0.75 residues – –
tinctorius
Sugar beet Beta vulgaris Root without 0.25 1.54 shoot 0.22 5.8
crown
Sugarcane Saccharum spp. Stalks 0.01 0.2 Leaves 0.07 0.12
þstems
Sunflower Helieanthus Grain 0.63 0.72 residues 0.14 2.52
annuus
Tobacco Burley Nicotiana Leafþstem 0.31 3.86 stalks 0.31 3.86
tabacum
Tobacco Virginia Nicotiana Leaves 0.27 2 stalks 0.27 2
tabacum
Horticultural crops Part P K Not P K
harvested harvested
Artichoke Cynara scolimus 0.51 2 residues – –
Asparagus (green) Asparagus Stem 0.69 3.4
officinalis
Asparagus (white) Asparagus Stem 0.74 4
officinalis
Beet Beta vulgaris Root 0.32 2.46 shoot 0.44 6.26
Brussels sprout Brassica Leaves 0.51 3.25
oleracea
Cabbage Brassica Leaves 0.35 2.73
oleracea
(continued)
% dry % dry
matter matter
Carrot Daucus carota Root 0.33 2.43 shoot 0.19 1.88
Cauliflower Brassica Head 0.66 3.22
oleracea
Celery Apium Leaves 0.66 4.8 0.66 4.8
graveolens
Chicory Cichorium Leaves 0.23 4
intybus
Cucumber Cucumis sativus Fruit 0.53 4.25 residues – –
Eggplant Solanum Fruit 0.31 3 residues – –
melongena
Endive Cichorium Leaves 0.45 5.6
endivia
Faba bean (green) Vicia faba Fruits 0.5 1.32 residues – –
Leak Allium porrum Bulb 0.21 1.06 residues – –
Lettuce Iceberg Lactuca sativa Leaves 0.5 2
Lettuce Roman Lactuca sativa Leaves 0.75 6.67
Melon Cucumis melo Fruit 0.16 2.58 residues – –
Muskmelon Cucumis melo Fruit 0.36 3.16 residues – –
Parsley Petroselinum Leaves 0.4 2.7
crispum
Pepper (green) Capsicum Fruits 0.35 2 residues – –
annuum
Pepper (red) Capsicum Fruits 0.3 2.4 residues – –
annuum
Pumpkin Cucurbita spp. Fruit 0.39 2.78 residues – –
Radish Raphanus Root 0.4 3.17 residues – –
sativus
Spinach Spinacia Leaves 0.56 5.66
oleracea
Squash Cucurbita pepo Fruit 0.4 3.5 residues – –
Tomato Lycopersicon Fruit 0.47 4.28 residues 0.1 1.9
esculentum
Watermelon Citrullus lanatus Fruit 0.11 1.33 residues – –
Fruit trees, vines Part P K Not P K
and shrubs harvested harvested
Almond Prunus Fruit 0.37 0.75
amygdalus
Apple Malus sylvestris Fruit 0.05 0.75
Apricot Prunus Fruit 0.14 2.17
armeniaca
Avocado Persea Fruit 0.15 2.31
americana
Banana Musa Fruit 0.08 1.54
paradisiaca
Cherimoya Annona Fruit 0.15 1.17
cherimola
(continued)
% dry % dry
matter matter
Cherry Prunus avium Fruit 0.01 1.16
Coconut Cocos nucifera copra 0.3 5
Date palm Phoenix Fruit 0.05 0.84
dactylifera
Fig Ficus carica Fruit 0.07 1.11
Grape (table) Vitis vinifera Fruit 0.05 1.02
Grape (wine) Vitis vinifera Fruit 0.07 0.95
Grapefruit Citrus paradisi Fruit 0.11 1.38
Hazelnut Corylus avellana Fruit 0.33 0.47
Kiwi Actinidia spp Fruit 0.18 1.43
Lemon Citrus limon Fruit 0.12 1.15
Mango Mangifera indica Fruit 0.11 0.95
Oil palm Elaeis fruit bunch 0.09 0.75
guineensis
Olive Olea europaea Fruit 0.14 1.25
Orange Citrus sinensis Fruit 0.14 1.35
Peach Prunus persica Fruit 0.12 1.55
Pear Pyrus communis Fruit 0.07 0.77
Persimmon Dyospiros kaki Fruit 0.07 1.01
Plum Prunus domestica Fruit 0.07 1.16
Pomegranate Punica granatum Fruit 0.1 1.04
Quince Cydonia oblonga Fruit 0.1 0.95
Walnut Juglans regia Fruit 0.22 0.41
Roots, tubers & Part P K Not P K
bulbs harvested harvested
Cassava Manihot Root 0.12 0.77
esculenta
Garlic Allium sativum Bulb 0.44 1.38 residues 0.2 1.3
Onion Allium cepa Bulb 0.35 1.2 shoot 0.38 2.75
Potato Solanum Tuber 0.25 2 shoot 0.2 3.95
tuberosum
Sweet potato Ipomoea batatas Tuber 0.15 1.22
Yam (chinese) Dioscorea Tuber 0.15
opposita
Yam (white) Dioscorea Tuber 0.25 2.3
rotundata
Yam (yellow) Dioscorea Tuber
cayenensis
Bibliography
Black, C. A. (1993). Soil fertility evaluation and control. Boca Raton: Lewis Publishers.
FAO. (2008). Efficiency of soil and fertilizer phosphorus use: Reconciling changing concepts of
soils phosphorus behaviour with agronomic information (FAO Fertilizer and Plant Nutrition
Bulletin 18). Rome: FAO.
Johnston, A. E. (2005) Phosphorus nutrition of arable crops. In J. T. Sims & A. N. Sharpley (Eds.),
Phosphorus: Agriculture and environment (pp. 495–519). Madison: Soil Science Society of
America.
Matar, A., Torrent, J., & Ryan, J. (1992). Soil and fertilizer phosphorus and crop responses in the
dryland mediterranean zone. Adv Soil Sci, 18, 81–146.
McLean, E. O., & Watson, M. E. (1985). Soil measurements of plant-available potassium. In R. D.
Munson (Ed.), Potassium in agriculture. Madison: ASA, CSSA, SSSA.
Pan, W. L. (2012). Nutrient interaction in soil fertility and plant nutrition. In P. M. Huang, Y. Li, &
M. E. Sumner (Eds.), Handbook of soil sciences, resource management and environmental
impacts (2nd ed., pp. 16–1–16–11). Boca Raton: CRC Press.
Sims, J. T., & McGrath, J. (2012). Soil fertility evaluation. In P. M. Huang, Y. Li, & M. E. Sumner
(Eds.), Handbook of soil sciences, resource management and environmental impacts (2nd ed.,
pp. 13–1–13.36). Boca Raton: CRC Press.
Sims, J. T., & Sharpley, A. N. (2005). Phosphorus: agriculture and the environment. Madison:
ASA, CSSA and SSSA.
Vitosh, M. L., Johnson, J. W., Mengel, D. B. (Eds). (1995). Tri-state fertilizer recommendations
for corn, soybeans, wheat and alfalfa. Michigan State University, Ohio State University,
Purdue University. Extension bull E-2567.
Chapter 27
Fertigation
Francisco J. Villalobos, Miguel Quemada, and Antonio Delgado
Abstract Fertigation is the joint application of nutrients along with irrigation

water. It is best suited for high frequency drip irrigation although it may be adapted
to other irrigation methods. Fertigation requires a dosing system and tanks for the
stock solution where nutrients are incorporated. The main characteristics to con-
sider for the fertilizers used in fertigation are concentration, purity, solubility and
pH reaction. The quality of irrigation water has also to be considered in special the
concentration of bicarbonates and calcium. The calculation of stock solutions is
based on the total requirements for N, P and K and total irrigation to be applied.
Specific recipes of fertigation solutions have been developed for several species.
Complete nutrient solutions (e.g. Hoagland-Arnon) are easy to calculate taking into
account the actual concentration of salts in irrigation water.
27.1 Introduction
Fertigation is the joint application of water and nutrients, which requires the
connection of a dosing system of nutrients to the irrigation system. Although this
technique can be applied in principle to all types of irrigation systems, it is generally
used in drip irrigation systems and to a lesser extent in full coverage sprinkler
systems and irrigation machines.
F.J. Villalobos (*)

M. Quemada
A. Delgado
Sevilla, Spain

Agriculture, DOI 10.1007/978-3-319-46116-8_27
If handled correctly, a very high efficiency in nutrient application to crops can be

achieved with this fertilizer management, particularly with high-frequency drip
irrigation. This high efficiency derives from:
(a) Fertigation implies, with drip irrigation, a localized fertilizer application, which
is more efficient in poor-nutrient or high fixing capacity soils. This is relevant
for nutrients that can be strongly bound to soil particles such as P and K.
(b) Drip irrigation systems cause a concentration of the root system within the wet
bulbs; thus the nutrient application is concentrated in areas of high root length
density. This contributes to a better uptake of applied nutrients.
(c) Using high-frequency drip irrigation, the water content in wet bulbs is usually
close to field capacity of the soil; high water content enhances the flux of
nutrients to roots through mass flow or diffusion, thus increasing the efficiency
in applied fertilizers, which means that a larger fraction of applied fertilizers is
finally used by crops.
(d) Fertigation with drip irrigation allows the very frequent application of very low
fertilizer rates which allows adjusting nutrient supply to plant requirements that
can vary depending on the growing stage. This has clear advantages for mobile
nutrients such as N since large soil N accumulation is avoided thus decreasing
losses and opening the opportunity for reducing N application, as can be seen
when fertigation with N is compared with a conventional N fertilization com-
bining planting and two side-dress applications (Fig. 27.1). With non-mobile
nutrients, such as P and K, this application also have advantages since the
frequent application of very low rates enhance fixation in forms (adsorption vs
precipitation in calcareous soils in the case of P) in a more ready equilibrium
with soil solution thus enhancing a greater efficiency of applied nutrients.
Fertigation can be a cost-effective fertilizer application technique when a drip
irrigation system has been installed. Although it requires the installation of a dosing
system, after this initial investment the cost of equipment and labor for applying
fertilizers are usually lower than with other techniques. On the other side, fertilizers
used are usually more expensive.
27.2 Dosing Systems
Fertigation is usually performed by injecting fertilizer in liquid form (fertilizer

solution, commercial or prepared in the farm) into the irrigation water flow to the
emitters. Two types of dosing methods can be distinguished:
(a) Proportional: It applies a constant nutrient concentration, so the amount of
nutrients that enters the system has to be proportional to the irrigation flow.
(b) Quantitative: A total amount of nutrients is added to the irrigation system. The
concentration varies with time.
Fertilizer injection can be done by using different types of devices (Fig. 27.2):
27 Fertigation 409
CONVENTIONAL FERTILIZATION
N lost potential
TOTAL N APPLIED
Kg N/ha
Soil available N
N plant uptake
N at planting 2nd N application 3rd N application
FERTIGATION
N lost potential is
TOTAL N APPLIED
Kg N/ha
reduced
Soil available N
N plant uptake
Several applications adjusted to crop N needs

Days after planting
Fig. 27.1 Time course of crop N uptake and soil N availability for conventional fertilization and
fertigation
Irrigation pipe Irrigation pipe Irrigation pipe

Venturi Injector
FS
FS
FS
Fig. 27.2 Diagrams of fertigation dosing systems. (a) fertilizer tank with pressure differential.
(b) With Venturi device. (c) With hydraulic injection pump
(a) Differential pressure.

A pressure regulating valve is inserted between the inlet and outlet connected to
the fertilizer tank. The pressure difference forces the water to flow through the
tank, from which it takes the nutrients and transports them to the main line.
It can be used to add solid or liquid products and serves only for quantitative
dosing. It is easy to maintain and does not require additional energy but reduces
the pressure in the main pipe. This method is best suited for conditions where
fertigation is performed at irregular intervals.
(b) Venturi system. It connects a Venturi device (narrowing) in parallel to the main
pipe. The depression resulting in the Venturi causes suction of the nutrient
solution. This system allows proportional dosing. The rate of application of
fertilizer can be adjusted and may provide very small amounts but causes
pressure loss in the main pipe.
The depression depends on the water velocity, thus the accuracy of a Venturi
dosing system depends on the capacity of the irrigation system to maintain
constant flow.
(c) Injection Pump. An electric or hydraulic pump may be used. In the latter case
no energy supply will be required. It is a system easy to install and operate,
allows adjusting the dose and does not involve loss of pressure in the main pipe.
It can be used for proportional or quantitative dosing.
(d) The irrigation pump inlet is connected to the fertilizer tank. The concentration
of nutrients is kept constant. If a system for automatic filling of the tank is added
then the concentration will vary with time. This is a simple and inexpensive
method but difficult to automate. You run the risk of sucking air in the pump or
the pump may be damaged by corrosion. This type of system is best suited for
small installations.
Whatever the system chosen it is advisable to install check valves to prevent
contamination of the supply line if return flow occurs and a filter downstream of
nutrient injection.
27.3 Alternatives of Fertigation
Different types of fertigation can be defined depending on the combination of

nutrients applied in each fertigation
(a) With complete nutrient solutions: although it is the usual choice for hydroponic
systems on artificial substrates such as rockwool and perlite, it can also be used
for crops on soil. The nutrient solution applied in each fertigation should
include all macro and micronutrients. This will require the use of several
tanks to avoid incompatibility problems between different fertilizer products:
Tank A: Macronutrients except Ca in an acid medium, usually applying part of
N or P such as nitric or phosphoric acid.
27 Fertigation 411
Tank B: Fertilizers with Ca in neutral or acid medium.

Tank C: Micronutrients in neutral medium.
Another alternative for distribution of nutrients:
Tank A: NPK.
Tank B: N, K, Ca, S and micronutrients.
Tank C: Nitric acid.
If only NPK is required for crop fertilization in soil, only one tank may be
needed. For hydroponics three tanks for applying all required nutrient is the
better choice.
(b) Incomplete solutions: In cropping systems on soil it is usual to apply P as basal
fertilizer before sowing to avoid the high cost of P soluble forms and the high
probability of P precipitation in the system. In this case only N and K are
usually applied through fertigation. Only one tank is required since N and K
combinations are not problematic. It is also possible to apply each primary
nutrient in different fertigations, with a tank for each primary nutrient. This
latter solution reduces the risk of precipitation in the tank or irrigation net.
Whatever the type of fertigation tank, its size and the total amount of fertilized
dissolved is determined by the solubility of the less soluble fertilizer.
27.4 Fertilizers for Fertigation
The main properties considered in fertilizers for fertigation are nutrient concentra-
tion, purity, solubility, pH effect and compatibility. Also the electrical conductivity
in the applied solution should not exceed certain thresholds and the pH should be in
the range 5.0–6.5. In that pH range nutrients are available for root uptake. Above
this range precipitates can be formed (e.g. Ca compounds with phosphate) while
below the root system may be damaged by an excessively acid fertigation solution.
The main properties of fertilizers more widely used in fertigation are shown in
Table 27.1.
The form in which nitrogen is supplied is a critical aspect in fertigation. Nitrogen
cannot be supplied exclusively as NH4+ because (a) it is phytotoxic at high
concentration in the growing media and (b) it can promote a decreased uptake of
other cations such as Ca2 +, Mg2 + and K +. This is caused by competition for
absorption mechanisms and by decreasing electrochemical potential through
plasma membranes which induces an increase excretion of H + by root cells to
maintain electrochemical gradient. Conversely, when N is provided only as NO3,
its absorption promotes an alkalinization of root apoplast and rhizosphere due to the
absorption mechanism of nitrate (symport with H+) which can negatively affect the
absorption of some nutrients such as Fe. Therefore it is recommended to apply N as
80–90 % nitrate and 10–20 % ammonium to maintain the pH of the rhizosphere in
412
Table 27.1 Properties of the main fertilizers used in fertigation. The simplest solid fertilizers are included with the exception of acids that are managed as
liquids
Eq weight Solubility 20 C Concentration (mass percentage)
g/eq kg/m3 N P K S Ca Mg
Monoammonium phosphate NH4 H2 PO4 115 626 12 22.6 0 1.5 1.5 0
Monopotassium phosphate K H2 PO4 136.1 200 0 22.8 28.7 0 0 0
Ammonium nitrate NH4 NO3 80 1920 34 0 0 0 0 0
Ammonium sulphate (NH4)2 SO4 66.1 730 21 0 0 24 0 0
Calcium nitrate (hidrated) Ca(NO3)2 .4H2O 118 1220 12 0 0 0 17 0
Potassium nitrate K NO3 101.1 316 13.4 0 39 0.2 0 0
Magnesium nitrate (hidrated) Mg (NO3)2.6H2O 128.2 1330 10.9 0 0 0 0 9.5
Potassium chloride K Cl 74.6 340 0 0 49.8 0 0 0
Potassium sulphate K2 S04 87.2 110 0 0 41.5 16 0 0
Magnesium sulphate (hidrated) Mg SO4.7H2O 123.2 710 0 0 0 13 0 9.9
Urea CO (NH2)2 60.1 1033 46 0 0 0 0 0
Phosphoric acid 55 %, 1.38 g/cm3 H3 PO4 98 5480 0 17.4 0 0 0 0
Phosphoric acid 75 %, 1.58 g/cm3 H3 PO4 98 5480 0 23.7 0 0 0 0
Nitric acid 57 %, 1.35 g/cm3 NO3 H 63 10,000 12.5 0 0 0 0 0
Calcium nitrate Ca (NO3)2 82 1212 17 0 0 0 24 0
Magnesium nitrate Mg (NO3)2 74.1 770 18.9 0 0 0 0 16.4
27 Fertigation 413
optimum values while taking advantage of the acidification effect of NH4+. The
application of N only as NO3 could be a good choice for fertigation in acid soils.
Temperature is the critical aspect affecting solubility of fertilizers, which is
proportional to temperature. Thus, the maximum concentration of fertilizers in a
solution is determined by the minimum temperature in the tank. Dissolving fertil-
izers is usually an endothermic reaction which decreases the temperature of the
solution. The effect is important for urea and nitrates (ammonium, calcium and
potassium). However, dilution of phosphoric acid is an exothermic reaction which
can be used to compensate the effect of endothermic dissolution reactions thus
increasing the solubility of the fertilizer added afterwards.
The products most widely used are nitrate (calcium, ammonium, potassium) and
potassium chloride, which are very soluble compounds. To ensure the requirement
of high purity and high solubility the fertilizer industry produces specific solid
fertilizers for fertigation. There are also commercial solutions (e.g. N-20 solution,
calcium nitrate, nitric acid, phosphoric acid, ammonium polyphosphates, various
complexes, microelements). Composite solid fertilizers are also produced and
composite liquid fertilizer solutions are presented in a wide range of ratios N: P:
K, with or without micronutrients. Liquid composite fertilizers have a low nutrient
concentration due to solubility limitations.
The use of incompatible fertilizers or the interaction of the fertilizer with
irrigation water, especially if it is hard and/or alkaline water, can cause the
formation of precipitates in the fertilization tank and the clogging of drippers and
filters. These problems can be avoided by a proper choice of fertilizers and proper
management of the irrigation net which must consider appropriate leaching and the
use of acidified fertilizer solutions.
The main incompatibilities among fertilizers in fertigation are those involving
the risk of precipitation of Ca and Mg compounds, such as:
Calcium nitrate in combination with phosphates or sulfates leads to precipitates of
calcium sulfate or calcium phosphate.
Ammonium phosphate in combination with magnesium sulfate leads to magnesium
phosphate (precipitate).
Micronutrients application should take into account the stability of the forms in
which they are applied, usually as chelates which is affected by pH and by the
presence of other cations in high concentration such as Ca. If Fe and P are applied in
acid solutions Fe phosphates can precipitate.
27.5 Quality of Water for Fertigation
The main salts in water are chlorides, sulfates, carbonates and bicarbonates of Ca,
Mg, Na and K. Some waters may contain other ions (nitrates, phosphates, ammo-
nium, etc.) and certain metals (iron, manganese, zinc, lead, etc.) which can be toxic
(Table 27.2). A standard laboratory analysis of irrigation water includes the major
Table 27.2 Main interactions among ions to be considered in fertigation

Ion Toxicity Precipitates with Impairs absorption of Favors absorption of
Ca No SO4, HCO3, H2PO4
Mg No H2PO4 K+a
Na Yes
Cl Yes NO3, H2PO4
2+
SO4 Yes Ca Na
a
This antagonistic effect could be more evident with fertigation than with conventional fertiliza-
tion since fertigation can keep greater concentrations of Mg and K in the soil solution
cations (Ca2 +, Mg2+, Na+ and K +) and anions (Cl, SO4 2, CO32, HCO3).
Boron can also be determined because of its high toxicity even in very small
concentrations. Nitrate concentration (dominant N form in water) should also be
measured to take it into account in the N balance to estimate N requirement by
crops.
The most common problem is the presence of bicarbonate, which combines with
Ca and Mg and may precipitate depending on the pH. Waters high in Ca with
alkaline pH will cause problems with more than 2 meq bicarbonate/L. To correct
these problems and bring the pH to the desired range (5.0–6.5) an acid is added,
leaving around 0.5 meq/L of bicarbonate not neutralized.
Another possible problem is related to P fertilizers. Insoluble Ca and Mg
phosphates are generated in waters high in Ca or Mg when pH is high. These
precipitates are deposited on the walls of the pipes and in the emitters, causing their
clogging. The availability of P to the plants is also reduced. It is therefore
recommended to use acid P fertilizers (mono-ammonium phosphate or phosphoric
acid) to reduce the risk of precipitation of Ca and Mg phosphates.
In any case, dissolved fertilizers remaining in the emitters at the end of the
fertigation can precipitate when water evaporates. To avoid this the duration of
fertigation should be shorter than that of irrigation allowing flushing with water at
the end of the irrigation. To dissolve the precipitates left and unclog the drippers we
may use the acidic reaction of some fertilizers and/or the injection of an acid
solution that also removes bacteria and algae. After injecting the acid the irrigation
and the injection systems should be carefully washed with additional irrigation
water.
Other quality issues in irrigation water may be the following:
(a) The presence of algae (irrigation ponds) or bacteria (groundwater, ponds)
require additional treatments which may be performed with chlorine, copper
sulfate (5 ppm) or potassium permanganate (2 ppm).
(b) Ferruginous underground waters produce rust deposits of Fe or Mn when they
oxidize. They require first pre-treatment such as aeration or chelation, and then
filtration to retain precipitated oxides.
27 Fertigation 415
27.6 Calculation of Stock Solutions
In fertigation systems with proportional dosing, a concentrated solution known as

stock or mother solution is prepared in the irrigation head. Fertigation is
programmed to dilute the solution with the irrigation water in the ratios of 1:100
up to 1:1000, while controlling the pH and EC. This results in the fertilizer solution
that, after being filtered, will reach the emitters. This solution reacts with the
substrate and results in the final nutrient solution that is absorbed by the roots.
The electrical conductivity may be calculated approximately as a function of the
concentration of cations (CC, in meq/ L) or anions or as a function of the salt
concentration (also named total dissolved solids, TDS, in g/L):
ECðdS=mÞ ¼ CC=10 ¼ TDS=0:64 ð27:1Þ
For the calculation of the stock solution we may face different situations:
(a) We know the total amount of N and K to add to the total amount of irrigation.
Therefore we deduce the concentration of the nutrient, and then we convert it to
a quantity of fertilizer to be added using the concentrations indicated in
Table 27.1. The maximum solubility (e.g. 80 % of this) should not be exceeded,
particularly if temperature oscillations are expected. In the final fertilizer
solution the stock solution is diluted M times, so the amount of fertilizer to
add to the stock solution (kg fertilizer/m3) will be:

nutrient requirement kg nutrient
M
ha
3
irrigation mha concentration kg
kg nutrient
fertilizer
(b) We know the ideal concentration of each nutrient (Ne, Pe, Ke) in meq/L and we
want to determine the amount of fertilizers to be added to a tank, considering
that the stock solution will be diluted M times.
(b.1) No acid correction. In the simplest case we have soft water with less than
0.5 meq bicarbonate/L so pH correction is not required. In hydroponics,
dissolved bicarbonate could be the C source for autotrophic nitrifying
microorganisms; thus to ensure conditions for rapid nitrification we
should use a low NH4+/NO3 ratio.
We start from the P fertilizer requirement, as it is often the element of
lower concentration. We will apply P as monoammonium phosphate
(MAP) or monopotassium phosphate to cover the needs. The remaining
needs of N (and/or K) will be completed with potassium nitrate, ammo-
nium nitrate and/or potassium sulphate.
Example 27.1 The ideal solución for a given crop is 4-1-2 meq/L. The stock
solution is diluted 400 times.
With monoammonium phosphate (MAP):
P: 1 meq P/L ∙115 mg MAP meq P1 ∙ 103 g mg1 ∙ 103 L m3 ¼ 115 g
MAP m3
This amount contains also 1 meq/L of NH4+, which is discounted from the
required N concentration, so we still need to add 3 meq N/L which may
achieved using 1.5 mmol/L of ammonium nitrate, AN (as each mol provides
2 equivalents of N).
1.5 mmol AN/L ∙ 80mg AN mmol AN1 ∙ 103 g mg1 ∙ 103 L m3
¼ 120 g AN m3
Finally we satisfy the K requirement using potassium sulfate (PS):
2 meq PS/L ∙ 87.2 mg PS meq PS1 ∙ 103 g mg1 ∙ 103 L m3 ¼ 174.4 g PS m3
The concentrations in the stock solution will be obtained by multiplying
the concentrations above by the dilution factor (400):
MAP 46 kg m3
NO3NH4 48 kg m3
SO4K2 69.8 kg m3
We should check that the concentrations in the stock solution do not

exceed the solubility of the fertilizers used (Table 27.1).
Other possible stock solutions could be:
Monopotassium phosphate 54.4 kg m3
NO3K 40.4 kg m3
NO3NH4 48.0 kg m3
Monopotassium phosphate 54.44 kg m3

SO4K2 34.88 kg m3
NO3NH4 64.00 kg m3
(b.2) Water with more than 0.5 meq HCO3/L: the pH has to be corrected.
Under these conditions, dissolved bicarbonate is not restrictive for nitrifi-
cation so the ratio NH4+/NO3 does not matter.
Acid is used to neutralize bicarbonate leaving only 0.5 meq/L. Then the
procedure is similar to that explained in the previous case taking into
account the nutrients added with the acid.
27 Fertigation 417
Example 27.2 Ideal solution 4-1-2 meq/L. Water with 3.5 meq/L of HCO3.
Dilution 400 times. Correction with phosphoric acid 55 % (density 1.38 g cm3)
To neutralize 3.0 meq/L of HCO3 we need 3.0 meq/L of protons which
can be supplied by 1 mmol/L of pure PO4H3.
1 mmol PO4H3/L∙ 98 mg PO4H3 mmol PO4H31 ∙ 103 g mg1 ∙ 103 L m3
¼ 120 g PO4H3 m3
120 g PO4H3 m3 ∙ 1 g solution/0.55 g PO4H3 ∙ 1 cm3 solution/1.38 g
solution ¼ 158 cm3 m3 (phosphoric acid 55 %)
We have also covered the need of P (1 meq/L).
We will cover now the need for K using potassium sulfate (PS):
2 meq K/L ∙ 87.2 mg PS meq K1 ∙ 103 g mg1 ∙ 103 L m3 ¼ 174.4 g PS m3
To supply 4 meq/L of N with ammonium nitrate, as each mol gives
2 equivalents of N, we will apply 2.0 mmol AN/L:
2.0 mmol AN/L ∙ 80 mg AN mmol AN1 ∙ 103 g mg1 ∙ 103 L m3 ¼ 160 g
AN m3
Then we apply the dilution factor (400) and therefore we come to the
following stock solution:
Phosphoric acid 55 % 63.2 L m3
SO4K2 69.76 kg m3
NO3NH4 64.00 kg m3
Example 27.3 Ideal solution 4-1-2 meq/L. Water with 3.5 meq/L of HCO3.
Dilution 400 times. Correction with nitric acid 57 % (density 1.35 g cm3)
To neutralize 3.0 meq/L of HCO3 we need 3.0 meq/L of protons which
can be supplied by 3 mmol/L of pure NO3H.
3 mmol /L ∙ 63mg mmol1 ∙ 103 g mg1 ∙ 103 L m3 ¼ 189 g m3 pure acid
189 g NO3H m3 ∙ 1 g solution/0.57 g NO3H ∙ 1 cm3 solution/1.35 g
solution ¼ 245.6 cm3 m3 (nitric acid 57 %)
Which contains also 3 meq/L NO3, thus we need a further addition of
1 meq/L of N to complete the required 4 meq N/L.
We apply P as MAP:
1 meq P/L ∙ 115 mg MAP meq P1 ∙ 103 g mg1 ∙ 103 L m3 ¼ 115 g MAP m3
Which contains also 1 meq/L NH4, therefore satisfying the whole needs
of N.
(continued)

Finally we apply K as potassium sulfate:
2 meq K /L ∙ 87.2 mg PS meq K1 ∙ 103 g mg1 ∙ 103 L m3
¼ 174.4 g PS m3
We apply the dilution factor (400) and arrive at the following stock
solution:
NO3H 57 % 98.24 L m3

MAP 46 kg m3
SO4K2 69.76 kg m3
27.7 Fertigation Control
Electrical conductivity (EC) in emitters can be estimated from water analysis and
amount of applied fertilizers. In any case, EC and pH can be measured in emitters to
check the accuracy of the calculations. In systems that allow measuring leachate
volume and the characteristics (pH, EC) of the input and output solutions, we can
check if the fertigation program is correct and amend it if necessary. This would
also serve for automation of the fertigation program.
First we set the leaching requirement (LR) as a function of the nutrient solution
EC. The observed values of the leaching fraction should be similar to LR. Other-
wise the irrigation volume should be adjusted.
A very low nitrate concentration in drainage may indicate that N is limiting, so
its concentration should be increased in the nutrient solution.
A higher value of EC and/or chlorine in the leachate than in the applied
solution indicates an accumulation of salts in the root zone. If the difference
between the EC of drainage and that of irrigation is greater than 0.4–0.5 dS/m,
and/or if the chlorine concentration in the leachate solution is higher than that of
the incoming solution and above 50 mg/L, an irrigation without fertilizers should
be applied to leach salts.
The optimum pH of the irrigation solution is 6–6.5 and can be adjusted by acid
injection. The drainage water pH should not exceed 8.5. Otherwise the NH4+/NO3
ratio of the nutrient solution should be increased up to 0.25.
27 Fertigation 419
27.8 Calculation of Complete Nutrient Solutions
In the case of hydroponics we need complete solutions including micronutrients.

The macronutrients to be added are determined taking into account the composition
of the irrigation water and the composition of the ideal solution. A widely used
reference nutrient solution is the one proposed by Hoagland and Arnon. The
calculation procedure is shown in an example in Table 27.3. Table 27.4 shows
the composition of recommended solutions for different species.
Table 27.3 Example of preparation of the Hoagland-Arnon solution. The negative value for
addition of bicarbonate indicates the need to apply 2.0 meq/L of a nutrient (e.g. N) as acid
Table 27.4 Some examples of recommended concentrations (meq/L) of macronutrients in the

nutrient solution for different species
NO3 NH4+ H2PO4 K+ Ca2+ Mg2+ SO4
Tomato and pepper 15 2 9 10 3 5
(hydroponics)
Tomato, pepper 8.0–12.0 0.9–1.4 1–1.5 4.0–6.0 4.8–7.0
(on soil)
Melon (on substrate) 6.5–11.5 0.7–1.3 1.2 4.0–7.5 3.5–6.5
Melon (on soil)a 9 0.8 – 7 4.5
Strawberrya 7 3.5 1 4.5
Bean 9 1 3.3 6.6 2 2
Cucumber 10.5 1 1 5 7.5 2 2
Lettuce, endive 19 1.2 2 9.0–11 9.0–10 2.0–3.0 2.25
Olive 1.4–2.75 0.6–1.25 1 2.0–4.0 2
Citrus 4–5.5 0.5 0.5 1–1.5 2
Grapevine 2.5–5 0.5–1 1 3.0–6.0
Adapted from Cadahı́a (2005)
a
Supplemented with basal fertilizer for P
The need for adding micronutrients is becoming more frequent as yields have
increased which increases the use of micronutrients, the fertilizers used are more
pure and thus contain less micronutrients and also because the use of manure has
been reduced. The availability of micronutrients usually increases as the organic
matter content of the soil or substrate increases and it is reduced by using hard or
alkaline water for irrigation. In any case, we should be cautious because
micronutrients can become toxic when in excess (Table 27.5). In soils and
substrates, metallic micronutrients (Fe, Mn, Cu, Zn, Ni), are usually present as
oxides and hydroxides, of low solubility at high pH. Boron and molybdenum,
whose concentrations are generally lower than those of metallic micronutrients,
are more soluble and may be present in the irrigation water or organic fertilizers.
Chlorine is also a micronutrient but is rarely scarce and can be toxic at high
concentration.
Micronutrients are added as chelates or salts that can be applied individually or
as ready-made solutions. For some species optimal concentrations of
micronutrients in the nutrient solution have been determined (Table 27.6).
Some authors recommend providing all metallic micronutrients as chelates
although there are some available soluble inorganic salts that can be used
(e.g. CuSO4), but usually are less effective in providing available nutrient to
plants due to oxidation in soil, particularly in the case of Fe. In general inorganic
salts, such as sulfates, are the best option for foliar applications. Table 27.7 shows
the most commonly used products.
27 Fertigation 421
Table 27.5 Allowable concentration of several elements in irrigation water

Long Short
term term
mg/L mg/L Comments
Aluminium 5 20 May turn acid soils into unsuited for cropping. Precipitates
(Al) with pH¼5.5–8.0 which eliminates toxicity
Arsenic (As) 0.1 2 Variable toxicity : 12 mg/L (Sudangrass) – 0.05 mg/L (rice)
Beryllium 0.1 0.5 Variable toxicity : 5 mg/L (cabbage) – 0.5 mg/L (bean)
(Be)
Boron (B) 0.75 2 Toxic for sensitive species (e.g. citrus) from 1 mg/L. Grasses
are tolerant of 2–10 mg/L
Cadmium 0.01 0.05 Toxic for beans, beets, radish with 0.1 mg/L
(Cd)
Chromium 0.1 1 Scarce information. Caution is recommended
(Cr)
Cobalt (Co) 0.05 5 Toxic for tomato with 0.1 mg/L. It is inactivated in neutral
and alkaline soils
Copper (Cu) 0.2 5 Variable toxicity : 0.1–1.0 mg/L
Fluorine (F –) 1 15 It is inactivated in neutral and alkaline soils
Iron (Fe) 5 20 Not toxic in well aerated soils. Induces acidification and
losses of P and Mo
Lead (Pb) 5 10 May inhibit cellular growth at high concentration
Lithium (Li) 2.5 2.5 Most crops are tolerant up to 5 mg/L except for Citrus (limit
0.075 mg/L). Moves in the soil
Manganese 0.2 10 Variable toxicity in acid soils
(Mg)
Molybdenum 0.01 0.05 Not toxic for plants in general. May be toxic for cattle when
(Mo) pastures grow on rich soils
Nickel (Ni) 0.2 2 Toxic for some species at 0.5–1.0 mg/L. Lower toxicity in
neutral and alkaline soils
Selenium (Se) 0.02 0.02 Toxic for plants at low concentration. May be toxic for cattle
when pastures grow on soils with low concentration
Vanadium 0.1 1 Toxic for many species at low concentration
(V)
Zinc (Zn) 2 10 Toxic for many species. Toxicity is reduced when pH>6 and
in clay and organic soils
Table 27.6 Recommended concentrations (mg/L) of micronutrients in the nutrient solution

Mn Fe B Mo Cu Zn
Olive, grapevine, citrus 0.5–1 1–1.4 1–1.1 0.01–0.02 0.05–0.8 0.05–0.2
Tomato, pepper, eggplant 0.5–1 0.8–2 0.3–0.5 0.05 0.05–0.1 0.03–0.1
Strawberry, bean, cucumber 0.5 1 0.3–0.5 0.05–0.1 0.1–0.2 0.1–0.2
Lettuce, endive 0.3–0.5 2.2 0.32 0.05 0.05 0.3
Adapted from Cadahı́a (2005)
Table 27.7 Products commonly used for correcting micronutrient deficiency. It must be noted
that for correction of Fe deficiency chlorosis in calcareous soils most of Fe present in EDDHA-Fe
should be orto-orto
Element Chemical % element Preferred use
Boron H3BO3 17
Na2B4O7.5H2O 20
Na2B4O7.10H2O 11
Ca2B6O11.5H2Oa 10
Copper CuSO4.5H2O 25 Foliar
CuOb 50–75 Soil
Iron FeSO4.7H2O 20 Foliar
FeHEDTA 5–9 Soil
FeEDDHA1 6 Soil
Manganese MnSO4.4H2O 24 Foliar
MnOb 41–68 Soil
Mn oxisulfate 30–50 Soil
Molybdenum Na2MoO4.2H2O 39 Foliar
(NH4)2MoO4 49
MoO3 66 Soil
Zinc ZnSO4.H2O 36 Foliar
Complex ZnSO4 -NH3 10–15
ZnOb 60–78 Soil
Zn oxisulfate 18–50 Soil
ZnEDTA 6–14 Soil
All products are soluble in water except those marked with a(slightly soluble) or b(insoluble)
27.9 Examples of Fertigation Programs
Example 27.4 An orange grove has an annual N fertilizer requirement of

200 kg N/ha. The total amount of irrigation applied is 500 mm. We will
calculate the amount of fertilizer to add to the stock solution (dilution 200)
to meet those N needs.
We assume that we want to apply only N so we rule out NP and NK
fertilizers, and restrict to urea and ammonium nitrate. We could also apply
other fertilizers containing sulfur, calcium or magnesium but they are
discarded because of their low N concentration.
Irrigation water should have a concentration:
200 kg N/5000 m3 ¼ 0.04 kg N m3 ¼ 40 g N m3
The two alternatives would be:
40 g N m3/0.34 ¼ 117.6 g ammonium nitrate m3
40 g N m3/0.46 ¼ 87 g urea m3
(continued)
27 Fertigation 423

The alternative stock solutions for a 200 dilution would have
concentrations of:
117.6 g ammonium nitrate m3 ∙ 200 ¼ 23.52 kg ammonium nitrate m3
87 g urea m3 ∙ 200 ¼ 17.4 kg urea m3
Example 27.5 A citrus orchard requires 200 kg N/ha and 270 kg K/ha with a
total irrigation application of 500 mm. We will calculate the stock solution
(dilution 200) to meet those needs of N and K.
Our first choice is a fertilizer containing both K and N, potassium nitrate.
To supply 200 kg N/ha and 270 kg K/ha, as the concentrations of N and K are
13.4 % and 39 %, respectively, we should add:
270 kg K/ha/(0.39 kg K/kg potassium nitrate) ¼ 692 kg potassium nitrate/ha
692 kg potassium nitrate/5000 m3 ¼ 138.4 g potassium nitrate/m3
That contributes also:
692 kg potassium nitrate/ha ∙ 0.134 kg N/kg potassium nitrate ¼ 92.8 kg N/ha
We still need to add 20092.8 ¼ 107 kg N/ha
That are equivalent to 315 kg ammonium nitrate/ha or 233 kg urea/ha.
If we choose urea, the concentration in irrigation water will be:
233 kg urea/5000 m3 ¼ 46.6 g urea/m3
And the stock solution will be:
27.68 kg potassium nitrate m3 and 9.32 kg urea m3
Alternatively we could have used simple fertilizers (urea and potassium
chloride) and the stock solution would be:
21.68 kg potassium chloride m3 and 17.4 kg urea m3
Considering that the solubility of potassium chloride and urea are much
higher (Table 27.1), concentrations could be an order of magnitude greater.
This implies a greater dilution factor (2000) that would allow a smaller size of
the tank as illustrated in the following example.
Example 27.6 Calculate the minimum size of the fertigation tank for the
previous example considering that the maximum irrigation requirement is
4.5 mm day1 and that the fertilizer is added every day. We will consider only
the option of using urea and potassium nitrate.
(continued)

The required concentrations in irrigation water were calculated in
Example 27.5:
138.4 g potassium nitrate/m3
46.6 g urea/m3
Now by looking at Table 27.1 we see that the maximum concentration to
be allowed in the stock solution (taken as 80 % of solubility) would be;
Urea: 826.4 kg m3 Potassium nitrate: 252.8 kg m3
The most limiting case is that of potassium nitrate which leads to a
maximum dilution factor of:
252,800/138.4 ¼ 1827
The stock solution should be:
Urea: 85.138 kg m3
Potassium nitrate: 252.857 kg m3
The amount of irrigation to be applied is 45 m3 ha1 day1 which requires
24.63 L ha1 day1 of stock solution. This is the minimum volume required
for the tank. For instance in a 10-ha orchard we would require a tank larger
than 246 L.
Bibliography
Burt, C. M., O’Connor, C., & Ruehr, T. A. (1995). Fertigation. San Luis Obispo: IRTC, Cal Pol St
Univ.
Cadahı́a, C. (2005). Fertirrigacion. Cultivos hortícolas, frutales y ornamentales. Madrid: Mundi-
Prensa.
Hagin, J., Sneh, M., & Lowengart-Aycicegi, A. (2003). Fertigation: Fertilization through irriga-
tion. Bassel: IPI Res TopicsInt, Potash Inst.
Kafkafi, U., & Tarchitzky, J. (2011). Fertigation: A tool for efficient fertilizer and water manage-
ment. Paris: Int Fertilizer Ind Assoc (IFA) & Int Potash Inst.
Chapter 28
Manipulating the Crop Environment
Francisco J. Villalobos, Luca Testi, and Luciano Mateos
Abstract Windbreaks are structures that reduce wind speed and may affect turbu-
lence in the protected zone. The maximum efficiency is obtained with windbreaks
of medium porosity that reduce wind speed up to a distance 20–25 times their
height. In the area protected by a windbreak temperature oscillations are larger,
which in some areas may increase frost risk and dew deposition.
Soil temperature can be modified by changing its exposure to radiation, by
artificial heating or by mulching. Mulches can be natural (e.g. crop residues) or
artificial, most notably plastic films. Canopy temperature can be reduced by wetting
with sprinklers although it is only effective with high VPD and implies excessive
water use. Simple models of the energy balance may be applied to calculate the
minimum and the maximum crop temperature. Additional environmental control
may be performed with row covers and greenhouses that create a warmer wind-
protected environment and are increasingly popular in horticultural production.
28.1 Introduction
There are limited possibilities for modifying the aerial environment of crops grown
outdoors. In this chapter we discuss these possibilities starting with protecting crops
from wind and then proceeding with environmental manipulations to modify soil
and crop temperatures.
The main factor that can be manipulated is wind which may be modified by
placing physical structures (inert or living) in the edges of fields. The structures may
form walls, called windbreaks, whose main objective is to reduce wind speed. The
term shelterbelt refers to several rows of trees and shrubs. The structures may be
F.J. Villalobos (*) • L. Mateos

L. Testi

Agriculture, DOI 10.1007/978-3-319-46116-8_28
scattered isolated trees which not only affect wind flow in the area but also have a
protective effect by intercepting rainfall. This association of crops and protective
trees is termed agroforestry. In such systems the trees may yield timber and/or fruit
which directly contribute to farm income, besides protecting the crops.
The use of windbreaks has been a common practice in agricultural systems of
regions with strong winds since long ago. An example would be the protective
windbreaks against the mistral wind in the Rhone Valley in southern France. In the
Great Plains of the U.S. the use of windbreaks were common after the 1930s to
protect the soil from wind erosion after a long period of drought (Dust Bowl). At
present they are only used in regions where wind poses substantial risks to agricul-
tural production. Because they use valuable land, windbreaks are mostly used in
horticulture (fruit tree production).
In contrast to their beneficial protective effects, windbreaks also have negative
effects. First they occupy part of the arable land, and they reduce incident radiation
on the cropped areas close to the windbreaks. If they are living structures, they may
also compete for water and nutrients and may serve as shelter for some pests. But
despite these drawbacks, most studies in windy areas have shown an overall
positive effect of windbreaks on crop yields.
28.2 Effects of Wind on Crops
The effects of wind on crops and soils are diverse:

– Growth: Plant movement due to wind can reduce crop growth rate and increase
plant’s mechanical resistance (shorter and thicker stems, increased root/shoot
ratio). This phenomenon called thigmomorphogenesis, does not require a con-
tinuous stimulus but may be triggered by infrequent movements.
– Mechanical damage: The wind’s force can tear leaves or strip them from the
plant. In dense canopies, abrasion may result from the rubbing of plant leaves
and stems. An indirect mechanical damage may be caused by the impact of soil
particles carried by wind.
– Crop lodging: This is caused by strong winds after wetting the canopy by rainfall
or irrigation, which increases the load on the plant and the bending moment and
decreases the stability of the root plate. The result is that the stems bend or break
at some point near the ground surface and the crop lays on the ground.
– Crop evaporation is proportional to wind speed when crops are well watered.
Therefore crops will use soil water faster in unprotected areas.
– Dry and hot winds may cause grain shriveling in cereals during early grain
filling.
– Salinization in coastal areas may occur due to wind drifts from the sea
– Wind affects the variation of surface temperature. Therefore the risk of frost may
increase in protected areas (see Chap. 29).
28 Manipulating the Crop Environment 427
28.3 Windbreaks
Windbreaks are structures established to reduce wind speed and change its
direction. Hedges may be formed by plants (shrubs, trees or annuals) or inert
structures (hurdles, plastic mesh enclosure walls or other specific structures).
Apart from reducing wind speed, plant windbreaks provide additional benefits
such as providing shelter for wildlife, protecting the livestock from
weather elements and becoming a barrier for sound and smell. In addition some
agricultural operations are improved in protected areas like reduced pesticide
drift and higher uniformity of sprinkler irrigation or pesticide spray application. A
better environment for farm workers is also created. In some areas, environmental
authorities encourage the development of windbreaks as a means to increase
biodiversity and to enhance the landscape by developing specific policies that
provide incentives to farmers. The increase in biodiversity (especially animal) is
often beneficial as birds and insects often prey on pests helping to reduce their
impact. In some cases, however, windbreaks may host detrimental insects,
maintaining populations stocks which can feed on the crops once they are
established. The species composition of windbreaks should then be chosen fol-
lowing an ecological rationale in addition to the aerodynamic considerations
addressed below.
Field windbreaks may be single rows of trees or shrubs or multiple-row shelter-
belts. The later provide better conditions for wildlife and may be formed by four to
five rows of alternating trees and shrubs. Taller species should be placed in the
center of the belt while shorter species can be placed on each side. Deciduous trees
have the disadvantage of losing much of their protective capacity during winter.
Tall annual crops may be used to protect shorter crops.
28.4 Wind and Turbulence in the Sheltered Zone
Windbreak structure -height, density, number of rows, species composition, length,

orientation, and continuity – determines the effectiveness of a windbreak in reduc-
ing wind speed and altering the microclimate.
The effectiveness of the windbreak depends mainly on its width, its height and
its porosity. Effectiveness is measured as the distance downwind, expressed as
number of shelter heights, through which the wind speed is reduced relative to that
in the open.
On the windward side of a windbreak, wind speed is reduced upwind for a
distance of two to five times the height of the windbreak (2H to 5H). On the leeward
side, wind speed may be reduced up to 30H downwind of the barrier (Fig. 28.1).
Windbreak porosity is the ratio of the open fraction of the barrier to its total
volume. Wind flows through the open portions of a windbreak, thus the less porous
a windbreak, the less wind passes through. Low pressure develops on the leeward
100
90
WIND SPEED (% OF OPEN FIELD)
80
70
60
50
deciduous
40 conifers
wall
30
20
–25 –20 –15 –10 –5 0 5 10 15 20 25 30
DISTANCE (number of heights)
Fig. 28.1 Variation of wind speed (as percent of its value in the open) as a function of distance
from the windbreak, expressed as number of heights. Negative and positive values represent the
windward and leeward sides, respectively. The porosity is 70–75 % for deciduous and 20–60 % for
conifers
side of very dense windbreaks, which pulls down air coming over the barrier,
generating additional turbulence and reducing protection downwind. As porosity
increases, so does the flow passing through the barrier, thus turbulence is not
enhanced, and the effectiveness increases, although the magnitude of wind speed
reductions are not as great.
Dense windbreaks (porosity lower than 25 %) show effectiveness (Ew) of
10–15H. With permeability around 50 % the effectiveness increases to 20–25H
(Fig. 28.1), without addition of large scale turbulence. These values of effectiveness
vary however with different factors such as wind speed (Ew is proportional to wind
speed), the atmospheric stability (Ew is larger in unstable conditions), wind direc-
tion (Ew is maximum when wind direction is normal to the barrier). Even when
wind blows parallel to the barriers some effect is observed. Windbreaks with
intermediate porosity (40–60 %) are usually the most effective.
28.5 Establishment and Maintenance of Windbreaks
Trees or shrubs to form windbreaks should grow rapidly, have strong erect stems
able to withstand wind forces and a well anchored root system. They should also be
able to survive under the prevailing abiotic stresses of the area (drought, cold).
Among plant windbreaks, the most commonly used species are conifers such as
cypress (Cupressus spp.), spruce (Picea spp.) or pine (Pinus spp.). Other trees used
are poplars, eucalyptus, etc. Each species and, within species, each variety have
characteristics of adaptation to the environment that determine which is most
appropriate in each case. There are also differences between species and varieties
with regard to competition with the crop, which come mainly from the patterns of
root growth which may be shallow or able to explore deeper soil horizons.
The orientation of windbreaks depends on the design objectives. Farmsteads and
feedlots usually need protection from cold winds and blowing snow during winter.
Field crops and fruit trees usually need protection from hot, dry summer winds, or
wind-blown soil particles, in special during critical growing periods. Windbreaks
for soil erosion control should be normal to the prevailing winds when the soil is
bare (winter and early spring). To recharge soil moisture with drifting snow,
windbreaks should be placed perpendicular to the prevailing winter winds.
Despite of the existence of a predominant wind direction during given periods,
wind direction may vary from day to day or during the day, so the level of
protection by the windbreak may be reduced. A set of multiple windbreaks forming
parallel lines spaced 10–15H provides a larger protected area than a single wind-
break. If protection from several wind directions is required another set of parallel
lines, normal to the first, would be established, resulting in a rectangular arrange-
ment of protected fields.
Sometimes gaps have to exist in the windbreak to allow access to the fields. The
uninterrupted length of a windbreak should exceed the height by at least 10:1. This
is because gaps in a barrier become funnels that concentrate wind flow, leading to
wind speeds in the protected area that may even exceed those in the open.
The plantation of a windbreak follows the same rules of other plantations
although the distance between trees may be smaller (e.g. 3 m between rows, 2 m
between trees in the row) and high survival and rapid growth are critical. It is
therefore very important to replace as soon as possible any tree lost and provide the
young trees with supplemental irrigation during dry periods and protection against
browsing by animals (by planting thorny plants or putting a barbed-wire fence).
Control of weeds is critical in particular during the early years of the plantation.
As the trees grow some pruning may be required to keep the required porosity,
promote vertical growth and eliminate branches damaged by wind or pests. Tree
thinning may be required to enhance trunk diameter growth.
28.6 Microclimate Changes in the Protected Area
Both solar radiation and net radiation are reduced significantly in the area shaded by
windbreaks. The effect is almost nil for distances beyond 1–2H. The effect is
marginal for north-south oriented barriers, as the shaded area is very small around
noon when radiation is at its maximum. Further reduction by shading may occur
early in the morning or late afternoon, but is partly compensated by reflection of
radiation from the windbreak.
The largest effect on radiation occurs in east-west oriented barriers, on the area
to the north (in the North hemisphere) of the windbreak, in special for high latitudes
and winter periods.
The reduction of wind speed, and thus, in turbulence in the protected areas has
several effects on:
(a) temperature: During the day it favors soil heating (Chap. 6), which usually leads
to warmer soil surface and air above. During the night strong temperature
inversions will develop leading possibly to lower minimum temperatures.
This explains the increased frost risk in protected areas.
(b) vapor pressure: It tends to increase close to the canopy during the day, when
plants are transpiring, as mixing is reduced, particularly in calm days. During
the night the higher vapor pressure and the lower temperature enhances dew
deposition in protected areas. The combination of higher vapor pressure or
plants wet by dew with higher temperature may increase the incidence of
diseases.
(c) evapotranspiration: For well watered crops reduced wind speed means higher
aerodynamic resistance (Chaps. 4 and 9) and therefore, reduced ET. This effect
may be offset partly by a reduction in canopy resistance in some species.
However, the improved environment in the protected area may increase crop
growth and hasten depletion of soil water. In rainfed crops subjected to water
stress late in the growing cycle, the overall effect may be a reduction in Harvest
Index in protected areas. However, well-watered crops show the same or higher
yield with reduced ET, which means a higher Water Use Efficiency.
(d) chill factor. Heat losses of livestock, wildlife and structures (farmstead, green-
houses, etc.) due to wind-chill are reduced on the leeward side of a windbreak.
28.7 Scattered Windbreaks
The presence of scattered trees in the field reduces the average wind speed because
of the increase in roughness. The reduction will be proportional to the fraction of
area covered by trees. The main difference of scattered trees and regular rows is the
degree of interaction (including competition) tree-crop which is higher when trees
are scattered. In this case the tree also provides protection to crop and soil from
direct rainfall impact. The negative effects are:
– the increased competition for light as isolated trees will intercept more radiation,
as well as for water and nutrients with the rest of the vegetation.
– additional difficulties for cultural operations as trees become obstacles for the
machinery.
– higher cost of establishing isolated trees, in special when young trees have to be
protected against wildlife or farm animals.
28.8 The Importance of Soil, Air and Crop Temperatures
Soil and air temperatures influence numerous critical processes of the crop. Seed
germination and plant emergence are extremely sensitive to the temperature of the
soil. The time from sowing to emergence increases and seedling growth is also
slower when the soil is cold. Canopy temperature has an important effect on critical
plant processes (development, growth, assimilation) and thus on crop productivity.
The root distribution is also affected by soil temperature. In some species root
growth is restricted to the upper layers when the soil is too cold, while a much
deeper root distribution is observed at higher temperatures, which improves water
and nutrient uptake. Other processes that respond to soil temperature are symbiotic
nitrogen fixation, photosynthesis, water flow in the soil-plant system (water viscos-
ity is high at low temperature), mineralization of organic matter and soil respiration.
In summary, soil and air temperatures have varied effects on crops and their
control may increase yields. Therefore artificial soil and/or air heating is sometimes
used in high value crops. The alternatives for crops grown outdoors are limited but
some may be very effective for manipulating soil temperature.
28.9 Slope and Aspect
The irradiance on a given surface increases as the incidence angle of solar rays
decreases, and is maximal when the radiation vector is normal to the surface. The
effect of slope and orientation (aspect) of the surface will be proportional to the
fraction of beam (direct) radiation reaching the surface, which lies between
0 (cloudy sky) and 0.85 (clear atmosphere with zero zenith angle). Slope and aspect
of a plot is less important when solar zenith angle is high or low, although for
different reasons. The fraction of direct radiation decreases as zenith angle
increases due to the longer path of atmosphere that sun rays cross to reach the
earth. Therefore the slope and aspect of a surface will barely affect irradiance in
high latitudes. On the other hand, for very low latitudes zenith angle is so small that
orientation has little effect on irradiance. The same reasoning may be applied to
seasonal changes: slope and aspect are important in spring or autumn but less
important in summer (low zenith angle) and winter (high fraction of diffuse
radiation).
Figure 28.2 shows the temperature in different places of a field with furrows in
the north-south direction. The temperature at the furrow is lower than on the ridge.
On the sloping sides, the temperature is typically higher than on the ridge, with the
side facing East warmer in the morning and the side facing West warmer in the
afternoon.
North-facing and South-facing slopes are colder and warmer, respectively, in the
Northern hemisphere.
24
22 WEST
RIDGE
20 EAST
18 FURROW
TEMPERATURE (ºC)
16
14
12
10
8
6
4
2
0
0 2 4 6 8 10 12 14 16 18 20 22 24
TIME GMT
Fig. 28.2 Time course of soil temperature at 2.5 cm depth in a North-South ridged sandy loam soil
in Cordoba (Spain)
28.10 Mulching
A mulch is a layer of material covering the soil and acting as a barrier to heat or
water transport. Additional functions of mulches include soil protection against
erosion and weed control. Some common mulches are weed residues, straw and
other crop residues, inorganic mulches (plastic films, gravel, sand) and industrial
byproducts (bark, wood chips, etc.).
The effect of mulching on soil temperature has to be evaluated by considering
first the possible change in net radiation. Black plastic will increase net radiation
(reduced albedo) while straw will reduce it (high albedo). The second aspect to be
considered is water transport. Mulches may reduce water flow or even suppress it
(plastic films). Therefore in many cases more energy will be available for heating
the air (straw) or the air and the soil. A transparent plastic will enhance soil heating
much more than a black plastic (Fig. 28.3). The latter absorbs radiation but trans-
mits and reflects little, so the black plastic is heated and may reach high tempera-
tures. However, conduction of heat to the soil is limited by the air layer between the
plastic and the soil surface. The transparent plastic is transparent to short wave
radiation but blocks little long wave radiation. Therefore during the night the soil
under transparent plastic cools like the control (bare soil) while the black plastic
keeps the soil warmer due to the low transmissivity for long-wave radiation. Straw
transmits little radiation which reduces soil warming during the day and cooling
during the night.
20
18
SOIL TEMPERATURE (ºC)
16
14
12
10
8
TRANSPARENT PLASTIC
6 BLACK PLASTIC
BARE SOIL
4
0 2 4 6 8 10 12 14 16 18 20 22 24
TIME GMT
Fig. 28.3 Time course of soil temperature at 2.5 cm depth under black or transparent plastic films
in a sandy loam soil in Cordoba (Spain)
Transparent plastic films are used widely in horticultural crops during the spring
to increase soil temperature and thus, speed up crop development for early produc-
tion and to reduce season length. Earlier harvest leads to better prices in many
horticultural crops. In addition to the effect on temperature, soil water is conserved
in the upper soil layer, which prevents the appearance of a surface crust and
improves the conditions for germination, emergence and early seedling growth.
Black plastic is also often used in vegetable crops, but the main objective is weed
control . Many weed seeds will remain dormant in the dark and those that do
germinate will die soon due to lack of carbohydrates. Another advantage of a plastic
cover (valid also for transparent films) is preventing the contact of fruits with the
soil thus avoiding diseases.
Most mulches, and especially those of organic origin (straw, crop residues, cover
crops) act as insulators, i.e. they damp the soil temperature waves, and therefore
will keep the soil cooler when applied in spring (Fig. 28.4) or warmer if applied in
the summer. Mulched spring sown crops (e.g. direct sowing with residues) will thus
show a slower development. The effect on the water balance depends on rainfall
distribution: frequent and light rains will wet the mulch and most water will
evaporate directly from it. Heavier and isolated rainfalls will infiltrate better in
mulched soil and soil evaporation will be reduced.
Inorganic mulches like sand or gravel have excellent properties as they do not
reduce soil heating while are very effective in reducing soil evaporation and
increasing infiltration.
32
30 Bare
Covered with residues
28
Mean daily soil temperature (ºC)
26
24
22
20
18
16
14
12
10
8
60 70 80 90 100 110 120 130 140 150
Day of the year
Fig. 28.4 Mean daily surface soil temperature for a sandy loam soil in Cordoba (Spain) in 2011,
covered with straw or bare
28.11 Artificial Soil Heating
In very special situations (nurseries, sport stadiums, high value horticulture) the soil
may be heated using electrical cables or pipes with hot water. The cost may be
reduced when hot water from cooling operations in industry or power plants is
available as a byproduct.
28.12 Modifying Canopy Temperature
Windbreaks increase temperature oscillations in the protected area (see 28.5). An

alternative for reducing canopy temperature of crops grown in the open is wetting
the plants. Some time ago a group of researchers in the USA proposed the use of
frequent irrigation to keep the canopy wet, and thus bring canopy temperature
closer to its optimum in order to increase productivity.
Using the equations for sensible and latent heat flux we may compute the
difference in temperature between the canopy and the air above as:

1 γ ð1 þ rc =ra ÞðRn GÞr a
Tc Ta ¼ VPD ð28:1Þ
Δ þ γ ð1 þ r c = r a Þ ρ Cp
Now, if the canopy is wet, canopy resistance is zero, so canopy temperature is now
given by:

1 γ ðRn GÞr a
T cw Ta ¼ VPD ð28:2Þ
Δþ γ ρ Cp
The difference in temperature between the dry and the wet canopy will be:

γ rc Δ ðRn GÞ VPD
Tc T cw¼ þ
½Δ þ γ ð1 þ rc =ra Þ ðΔ þ γ Þ ρ Cp ra
ð28:3Þ
Δ ðRn GÞ þ ρ Cp VPD=ra γ rc
¼
Δþγ ρ Cp ½Δ þ γ ð1 þ rc =ra Þ
The cooling due to wetting the canopy is proportional to radiation, VPD and to
canopy resistance. Note that the left term in Eq. 28.3 is the latent heat flux according
to the Penman-Monteith equation for zero canopy resistance (LEw). So the increase
of evaporation when the canopy is wet may be written as:

Δ ðRn GÞ þ ρ Cp VPD=ra ΔðRn GÞ þ ρ Cp VPD=ra
LEw LE ¼
Δþγ Δ þ γ ð1 þ r c =r a Þ
ð28:4Þ
γ rc =ra
¼ LE
Δþγ
Therefore the relative increase of evaporation due to wetting is proportional to rc/ra.

We can also deduce the reduction in canopy temperature per unit increase in latent
heat flux:
T c T cw ra
¼ ð28:5Þ
LEw LE ρ Cp
According to this equation, the efficiency of cooling a crop, taken as the ratio of
temperature decrease and the increase in water use, will be higher for smooth
(short) crops and low wind.
Example 28.1 An irrigated crop with rc ¼ 40 s/m and ra ¼ 40 s/m at midday

in summer (Rn-G ¼ 600 W/m2) in an arid area (air temperature 40 ºC and
relative humidity 30 %). At 40 ºC, ρ Cp ¼ 1140 J K1 m3
es ¼ 0.6108 exp[17.27 40/(40 + 237.3)] ¼ 7.37 kPa
ea ¼ es HR/100 ¼ 2.21 kPa
VPD ¼ 7.37–2.21 ¼ 5.16 kPa
Δ ¼ 4098 ∙ 7.37/(40 + 237.3)2 ¼ 0.39 kPa/K
(continued)

If the canopy is dry, latent heat flux according to the Penman-Monteith
equation is
ΔðRn GÞ þ ρ Cp VPD=ra 0:39 600 þ 1140 5:16=40

LE ¼ ¼ ¼ 727 W m2
Δ þ γ ð1 þ r c = r a Þ 0:39 þ 0:067 ð1 þ 40=40Þ
and canopy temperature is:

1 0:067 ð1 þ 40=40 Þ 600 40
T c ¼ 40 þ 5:16 ¼ 35:5 C
0:39 þ 0:067 ð1 þ 40=40 Þ 1140
If we wet the canopy, LE will be:
ΔðRn GÞ þ ρ Cp VPD=ra 0:39 600 þ 1140 5:16=40

LEw ¼ ¼ ¼ 834 W m2
Δþγ 0:39 þ 0:067
And the cooling effect would be:

γ rc
T c T cw ¼ LEw
ρ Cp ½Δ þ γ ð1 þ rc =ra Þ
0:067 40
¼ 834
1140 ½0:39 þ 0:067 ð1 þ 40=40 Þ ¼ 3:7 K
So the temperature of the wet canopy would be 31.8 ºC.
Apart from increasing crop water use, frequent wetting can have adverse effects
by promoting some diseases and reducing crop nutrient uptake (less water flowing
from the soil through the plant).
28.13 Minimum Crop Temperature
Air temperature is routinely measured in weather stations with shielded sensors at

standard height (e.g. 1.5 m). However, actual canopy temperature (Tc) will usually
differ from that value (Chap. 5). The main factors determining the difference
between air temperature at the weather station (Taw) and Tc are net radiation,
wind speed, air humidity and aerodynamic roughness (Chap. 4). The calculation
of this difference for the minimum temperature may be performed using a rather
simple model as during nighttime net radiation is only long wave, wind speed is
Table 28.1 Calculated difference between crop and air minimum temperature when air temper-
ature at the weather station is 0 ºC and wind speed is low
Short crop h ¼ 0.1 m Tall crop h ¼ 1 m
RH Cloudy Clear Cloudy Clear
70 1.37 4.05 0.80 2.53
100 0.64 2.41 0.32 1.25
Two conditions of cloudiness (completely overcast or clear) and two conditions of relative
humidity (rather dry, 70 % and saturated 100 %) have been considered for crops of height 0.1 m
and 1 m
usually low and relative humidity is high. An example of model results are
presented in Table 28.1 for two crops with height 0.1 m (e.g. grass) and 1 m
(e.g. cereal) when air temperature is 0 ºC and wind speed is low. It is clear that
the difference Taw-Tc is smaller for the taller crop, when the sky is overcast and
when humidity is high.
28.14 Calculation of Maximum Canopy Temperature
Maximum canopy temperature also differs from maximum air temperature. High
canopy temperatures may be detrimental to critical reproductive stages
(e.g. pollination in cereals) so it is important to understand how it is affected by
environmental factors. Again turbulence is a main determinant of the difference in
temperature between the crop and the air above: as wind speed increases the
difference between the two is reduced. However, water deficits that induce stomatal
closure decrease transpirational cooling and thus increase canopy temperature.
Therefore, the impact of high temperatures on crops is amplified by water deficits.
Here we present a simple procedure for calculating expected maximum crop
temperature from standard weather data.
Using Eqs. 9.10 and 9.12 we compute maximum canopy temperature as:
r aH
T cx ¼ T ax þ ½ð1 f G ÞRn LE ð28:6Þ
ρ Cp
where Tcx and Tax are maximum canopy and air temperature (ºC), respectively, fG is
the fraction of net radiation invested in soil heat flux (taken as 0.1 during the
daytime), LE is latent heat flux, raH is aerodynamic resistance for heat exchange, ρ
is air density and Cp is specific heat of air.
This equation is evaluated at the time of maximum temperature, which is
assumed to occur 3 h after solar noon. At that time, solar radiation on sunny days
is approximately 84 % of the value at solar noon. We also assume that on average,
net radiation is 60 % of solar radiation, so:
π 106 Rsd
Rnx ¼ 0:6 0:84 ð28:7Þ
2 3600 N
where Rnx is net radiation (W m2) at the time of maximum temperature, Rsd is
daily solar radiation (MJ m2 day1), and N is daylength (hour).
The time course of latent heat flux (LE) along the daytime is assumed to follow a
sine function, with the maximum occurring at the time of maximum air tempera-
ture. Therefore, maximum LE (LEx, W m2) is computed as:
π 2:45ET
LEx ¼ 6 ð28:8Þ
2 10 3600N
where ET is daily actual evapotranspiration (mm day1).

Calculation of raH for unstable conditions may be performed using a simplified
equation derived from the model of Thom and Oliver as a function of wind speed
over grass (U2g):
25:9
r aH ¼ ð28:9Þ
1 þ 2:3 U 2g=cw
The coefficient cw depends on crop height and lies between 7.3 (h ¼ 0.5 m) and 5.5
(h ¼ 3 m).
The calculation of wind speed at the time of maximum temperature is performed
by assuming that mean daytime wind speed is twice the mean value during the
nighttime, and that the value during the day is a sine function. Therefore, at the time
of maximum temperature, wind speed at the weather station is:
π U 2gm
U 2gx ¼ 1 þ ð28:10Þ
2 1 þ N=24
where U2gm is the mean (24-h) wind speed measured at the weather station.
For field crops typical values of raH are between 6 (high wind speed) and 12 s/m
(moderate wind speed) with only a minor effect of crop height.
Example 28.2 Wheat at flowering stage at Cordoba, Spain.

11 April 2014. Cw ¼ 6.6, daylength ¼ 13 h
Maximum air temperature 29.7 ºC, Reference ET¼ 5.1 mm/day,
Rsd ¼ 22.7 MJ m2 day1
Mean wind speed (weather station) ¼ 2.3 m/s
The maximum crop coefficient of wheat during this period is 1.1 (Chap. 9)
so:
ET ¼ 1.1 ET0 ¼ 5.6 mm/day
(continued)

Maximum wind speed (weather station):
π U 2gm
U 2gx ¼ 1 þ ¼ 1:667 2:3 ¼ 3:84 m=s
2 1 þ N=24
Aerodynamic resistance:
25:9
r aH ¼ ¼ 11:1s=m
1 þ 2:3 3:84=6:6
Maximum net radiation over the crop:
π 106 Rsd
Rnx ¼ 0:6 0:84 ¼ 384 Wm2
2 3600N
π 2:45 ET
LEx ¼ 6 ¼ 460 Wm2
2 10 3600 N
So maximum canopy temperature will be:
r aH 11:1
T cx ¼ T ax þ ½ð1 f G ÞRnx LEx ¼ 29:7 þ ½ð1 0:1Þ384 460
ρ Cp 1184
¼ 29:7 1:1 ¼ 28:6 C
If water deficit occurs so actual ET is only 50 % of maximum ET

(LE ¼ 460/2 ¼ 230 W m2):
11:1
T cx ¼ 29:7 þ ½ð1 0:1Þ384 230 ¼ 29:7 þ 1:1 ¼ 30:8 C
1184
If the crop had maximum stress (zero ET):
11:1
T cx ¼ 29:7 þ ½ð1 0:1Þ 384 0 ¼ 29:7 þ 3:2 ¼ 32:9 C
1184
28.15 Greenhouses
The highest level of control of the aerial environment of plants is achieved in

growth cabinets and growth chambers which are only used in research and breeding
programs due to their high cost. The lowest level corresponds to windbreaks. A
second step in environmental control is achieved with mulches and row covers
which are pieces of clear plastic stretched over low hoops enclosing the rows of
plants. Floating row covers are those supported by the plant itself. Also, shading
nets and other types of covers are being used now in fruit tree production as
protective covers against hail, to improve fruit quality, and for limiting access to
some insect pests. The use of nets of different colors that alter the light spectrum are
being tested for improving certain fruit quality features and to disrupt insect flights.
The third level is that of greenhouses which are structures covered with a transpar-
ent material. A wide range of designs differing in cost, level of control and frame
and cover materials is available. The simplest case is the plastic unheated green-
house for horticulture production in mild-winter areas (e.g. Almeria in Southern
Spain). The most sophisticated designs are metallic structures with glass or rigid
plastic panels that have artificial heating, supplementary lighting and CO2 fertili-
zation (e.g. the greenhouse industry of The Netherlands). Greenhouse horticulture
is increasingly using hydroponics instead of natural soils, a technology where plants
are grown with or without mechanical support on an artificial medium (sand, gravel,
rock wool, peat moss, etc.) and watered with a nutrient solution that is recycled
through the system.
Glazing materials for greenhouses may be plastic films (e.g. polyethylene, PE),
rigid plastic panels (e.g. polycarbonate) or glass. Glazing materials show high
transmittance for PAR (above 80 %), but they may be transparent to Infrared
(PE) or not (PE with specific additives, glass, any material covered by condensa-
tion). In the former case (IR transparent) radiative cooling at night is almost the
same as in the open. This may be mitigated by using IR opaque curtains. During the
day the problem may be the opposite due to excessive heating of the air and plants
inside the greenhouse during late spring or summer. In that case it is possible to use
shade cloth to reduce irradiance, increase ventilation or use cooling systems.
In Sect. 9.10 we saw that evaporation inside unheated plastic greenhouses
approaches equilibrium evaporation, i.e. it is mostly related to radiation inside the
greenhouse. In this case it is easy to deduce the sensible heat flux inside as the
difference between net radiation and evaporation. This sensible heat flux will be
equal to the transfer of sensible heat between the air inside and the air outside. For a
given relative renovation rate (RR, hour1) and mean greenhouse height (hg, m) we
can calculate the difference in temperature between the inside and the outside as:
ð1 kL ÞΔ þ γ 3600 kRN Rsi CT Rsi

T inside T outside ¼ ¼ ð28:11Þ
Δþγ ρCp hg RR hg RR
where γ is the psychrometric constant (approx. 0.067 kPa K1), Δ is the slope of the
saturation vapor pressure function versus temperature (kPa K1, see Eq. 9.22), ρ is
air density, Cp is specific heat of air at constant pressure (see 5.7), kRN is the ratio
net radiation/solar radiation inside the greenhouse, which may be taken as 0.7 and
Rsi is solar radiation inside the greenhouse (W m2). The coefficient kL represents
the fraction of evaporation as compared to equilibrium evaporation. For instance if
75 % of the area the greenhouse is covered by well watered vegetation then

kL ¼ 0.75 and CT varies from 1.16 at 10 ºC to 0.85 at 30 ºC.
Example 28.3 A 3-m high greenhouse of 200 m2 area located at latitude

37ºS is ventilated with an air flow of 5 m3 s1 (18,000 m3 h1). Therefore the
relative renovation rate is 18,000/(200 ∙ 3) ¼ 30 h1. On June 21 solar
declination is 23.45º so the maximum solar radiation outside at noon on a
clear day is 507 W m2 (Chap. 3). The cover is polyethylene with transmis-
sivity 0.7, so estimated radiation inside is 355 W m2. The temperature
outside is 20 ºC and kL is 0.75 so we can deduce CT ¼ 0.99. Therefore the
expected increase of temperature at that time is:
CT Rsi 0:99 355

T inside T outside ¼ ¼ ¼ 3:9 K
hg RR 3 30
So the temperature inside will be 23.9 ºC.
This simple model is only a first approximation to characterize the micromete-

orology of greenhouses as it ignores other processes that contribute to heat
exchange (e.g. conduction through the cover) but is very useful to illustrate the
possibilities of climatic control in unheated greenhouses, namely changing the
radiation inside by putting shade cloth or whitewash painting on the cover or by
manipulating ventilation via opening/closing vents or using fans. Apart from
keeping the temperature within the optimal range for plant growth, ventilation is
needed to prevent excessive air humidity inside the greenhouse as it enhances the
risks of fungal diseases. This is especially important at night when temperature
approaches the dew point temperature so condensation occurs on plants and on the
inner surface of the cover.
Bibliography
Burke, S. (1998). Windbreaks. Port Melbourne: Inkata Press.

Castilla, N. (2013). Greenhouse technology and management (2nd ed.). Wallingford: CABI
Publishing.
Grace, J. (1988). Plant response to wind. Agriculture, Ecosystems and Environment, 22(23),
71–88.
New York: Wiley.
Chapter 29
Frost Protection
José Paulo De Melo-Abreu, Francisco J. Villalobos, and Luciano Mateos
Abstract Frosts affect agricultural production reducing yields and/or product

quality. To avoid frost damage the best strategy is to use passive methods which
imply a good choice of species, cultivars, planting dates and locations, keeping the
soil compacted, wet and smooth, among other options. This requires the knowledge
of the frequency distributions of minimum temperatures and the evaluation of the
effect of low temperatures on crop performance (i.e., critical damage temperatures).
However, the mechanisms of frost damage are rather complex and depend partly on
plant hardiness, so predictions of damage are very uncertain. When frost occurs
there are active (protective) methods of control that minimize the damage, includ-
ing the reduction in long wave radiation loss (e.g. plastic covers), direct heating by
burning fuel, air mixing (e.g. wind machines), and overhead irrigation for releasing
the heat of fusion.
29.1 Introduction
A frost is the occurrence of air temperature equal or lower than 0 C at a height

between 1.25 and 2 m, measured in an appropriate shelter. Most agricultural
systems of temperate climates are affected by frost. The limitation to crop produc-
tion due to frost is usually characterized by the mean frost-free period, which is the
time from the last spring frost to the first fall frost. This period limits the growing
J.P. De Melo-Abreu (*)

DCEB, ISA & LEAF, ULisboa, Tapada da Ajuda 1349-017, Lisboa, Portugal
F.J. Villalobos
L. Mateos

Agriculture, DOI 10.1007/978-3-319-46116-8_29
444 J.P. De Melo-Abreu et al.
season of many crop species and determines the possibility of growing a crop in a
given region.
The potential of frost risk increases as we move away from the Equator, with a
band between the two tropics where frost does not occur (except at high elevations).
Areas with frost-free periods over 240 days are between 12 and 40 latitude and
include the most important agricultural regions. In these regions frost damage is
often prevented on fruit and horticultural crops. Areas with frost-free periods of
180–240 days extend to 50 , although they may be found in higher latitudes due to
sea influence. When the frost-free period is less than 90 days, agriculture is very
limited and most food crops cannot be grown.
Frost is the weather hazard that is responsible for the greatest crop losses in the
United States and, probably, in the World. Among frost protection methods, the
most useful are preventive methods, such as right choice of species/cultivar, site
selection, right choice of sowing date, and appropriate management techniques to
keep susceptible organs away from the soil surface, adequate plant nutrition, and
measures to enhance soil thermal conductivity. Protective methods, that are
implemented in the night of frost are usually expensive and thus can only be
performed in high-value crops in horticulture and fruit production.
29.2 Effects of Frost on Crop Production
Frost damage depends on many factors such as the species, the cultivar, the degree
of acclimation, the state of the plant tissues (which depends on the stage of
development, and on irrigation and fertilization practices, among other factors),
the height of the canopy, the type of pruning, the rate of temperature decrease, the
duration of the frost and the minimum temperature achieved. This complexity
makes it difficult to predict frost damage. An additional problem is that the
minimum temperature recorded at weather stations (according to standard rules)
is not the same as the canopy temperature in a given field nearby.
The resistance of crops to cold is evaluated according to the lowest average
minimum temperature at which they can survive. In tropical areas, plants are
generally tender, and damage may result from exposure to low temperatures
above the freezing point, sometimes as high as 12 C. The plant tissue damage
caused by low temperatures above 0 C is termed chilling injury. Chilling injury is a
particular problem in horticultural plants when unseasonal weather causes damage
to chilling-sensitive species (most tropical vegetable crops). As with frost, many
physiological and environmental factors affect the magnitude of the injury, for
instance, immature fruits are more sensitive than mature fruits. Contrary to frost,
chilling injury symptoms may be reversed, at least partially, if exposure to low
temperatures is brief.
Frost damage in plants occurs below 0 C, at temperatures ranging from about
1 C down to 196 C. This offset is explained by two mechanisms: avoidance
and/or tolerance of freezing. Plants avoid intercellular freezing either because the
29 Frost Protection 445
solutes outside the protoplast lower the freezing point of these aqueous solutions or
because there is supercooling (the temperature of the liquid drops below its freezing
point without becoming solid), due to absence of freezing nuclei. Tolerance hap-
pens when, despite the occurrence of intercellular freezing and the concomitant
shrinking and dehydration of the protoplast, after thawing there is full recovery of
the protoplast structure and function. Intracellular freezing, if it ever occurs under
natural crop growing conditions, is always lethal for the cell. When freezing occurs
it starts in the intercellular spaces due to the lower concentration of solutes outside
than inside of the protoplasts. The decrease in water potential due to freezing and
solute concentration induces loss of water by the protoplast, resulting in its shrink-
age and increase of solute concentration inside the protoplast.
Many mechanisms of frost injury have been put forward, but it is possible that in
general tissues are either injured by direct mechanical injury inflicted by the ice
formed outside cells or by the shrinkage and dehydration experienced by the
protoplast. In this case, denaturation of nuclear proteins may be the ultimate
cause of injury.
Initially, critical damage temperature (Tcrit) was defined as the maximum
temperature that results in frost injury to a plant organ that is subjected to it for
more than thirty minutes. The term has been extended to specific levels of injury.
For example, T10 for apple flowers refers to the Tcrit that inflicts 10 % loss of the
total flowers, and T90 would correspond to 90 % loss.
Let us issue some general considerations for cereals and fruit trees that are not of
tropical origin. During rest, plant organs have often very low Tcrit. However, after
bud burst critical temperatures slowly approach their upper limit. During active
growth, most plant organs have Tcrit that are only a few degrees below 0 C.
Moreover, the difference T10–T90 tends to decrease as plant phenological develop-
ment progresses attaining a minimum that occurs, usually, around grain/fruit set
(see Fig. 29.1). During the rest period and onset of growth, plants have a consid-
erable capacity to keep low (or actively lower) Tcrit, in response to the continued
occurrence of low temperatures, in a process called hardening or acclimation. From
flowering onwards, hardening capacity is inexistent or reduced. On the other hand,
after exposure to a period of high temperatures de-hardening may occur.
The nature of freeze damage varies with the plant/organ affected. Most vegeta-
bles that are injured present either a “burned” appearance, or seem “soggy”, or
present changes in color or texture. Under rigorous winters, when the protective
snow cover is insufficient, winter cereals may get leaf injury or even tillering node
injury. After emergence, frost damage is usually at flowering or at grain set. In
temperate climates, in general, deciduous fruit trees and the vine during winter
frosts are not affected. Only when temperatures are very low, in some extreme
environments or when there is substantial de-hardening, there is frost damage to
dormant buds or, even more rarely, to tree trunks. After bud burst, in the case of
apples, pears and stone fruits, flowers and small fruits are very tender and sensitive
to frost that causes substantial losses at those stages. Sometimes, however, after
pollination there is only partial loss of seeds which affects the growth of the fruits,
particularly in the case of stone fruits that only have one or two seeds per fruit.
CRITICAL TEMPERATURE (°C) –2
–4
–6
–8
–10
–12
10 % kill
–14
90 % kill
–16
–18
–20
1-Mar 11-Mar 21-Mar 31-Mar 10-Apr 20-Apr
DATE
Fig. 29.1 Typical 10 % and 90 % bud kill temperatures for cherry trees corresponding to average
dates observed at the Washington State University, Prosser Research and Extension Centre
(Adapted from Proebsting and Mills (1978))
When a small fruit experiences light freeze injury, a coarse russet tissue grows and
covers a portion of the fruit, resulting in the deterioration of fruit quality.
A detailed list of critical temperatures may be found in the review by Snyder and
De Melo-Abreu (2005). Most vegetables and fruits have maximum freezing tem-
peratures between 0.4 and 2.7 C, which represents the upper limit of critical
temperature, since the heat capacity of the structure and some (small) degree of
supercooling may result in actual freezing temperatures that are somewhat lower.
More juicy tissues tend to have higher critical temperatures.
A series of crops of tropical origin (tobacco, tomato, cucumber, peanut, rice,
melon and cotton) present Tcrit, for all the crop cycle, that decrease from 0 to 2 C.
Millet and corn have a Tcrit that is around 2 C or 3 C at germination and grain
filling, and one degree higher around flowering.
Table 29.1 shows some critical temperatures in relation to stage of small grains,
silage and forage crops, sugar beet and the olive crop. Table 29.2 shows critical
temperatures in relation to stage for some important deciduous fruit trees and
grapevines.
29.3 Frost Types
Radiation frosts occur on calm nights with a clear and dry atmosphere, which
enhances long wave radiation losses (Chap. 3). The low wind speed determines a
temperature inversion. Figure 29.2 shows an example of evolution of temperature
Table 29.1 Critical temperatures (ºC) in relation to stage of different crop species (Adapted from
Snyder and De Melo-Abreu (2005))
Crop Critical temperaturea Stage
Alfalfa 6/14
Barley (winter)b /17.3 to 12.9 Tillering
Barley (winter)b 1 to 2 Flowering
Barley (winter)b 2 to 4 Grain filling
Oat /10.5 to 6.5 Tillering
Oatb 8 to 9 Germination
Oatb 1 to 2 Flowering
Oatb 2 to 4 Grain filling
Olive 12.4 to 4.1/19.3 to 8.1 Rest
Potatob 2 to 3 Germination
Potatob 1 to 2 Flowering
Rye (winter)b /19.5 to 25 Tillering
Ryegrass (Italian) /8.4 to 7.4 3–4 leaf-stage
Ryegrass (Perennial) /13.95 to 10.31 Mature
Soybean /4.5 Seedlings
Soybeanb 1 Pod filling
Subterranean clover 5.5/7.8 Seedlings
Sugar beetb 6 to 7 Germination
Sugar beetb 2 to 3 Flowering
Sunflowerb 3.9 Bud formation
Sunflowerb 0.6 to 0 Flowering
Triticale /17.5 to 9.2 Tillering
Wheat (spring) 2/5.5 Tillering
Wheat (winter) 3/18 Tillering
White clover 7.7 to 4.9/20.3 to 7.4
a
Before the slash (/) temperatures correspond to unhardened plants and after to hardened
b
Under field conditions
profiles during the night. When the wind blows at night the temperature profile
becomes more uniform and the air temperature rises (Fig. 29.3). When the wind
stops, the temperature drops again (Fig. 29.3).
Advection frosts occur as a result of large-scale transport of cold air masses.
They occur on cloudy days or nights with moderate or strong wind coming in the
wake of a cold front. Temperature inversions are not present, at least in the first
phase of such events. Later, after the wind weakens, an inversion may develop if
surface cooling conditions occur.
Hoar frost occurs when ice crystals appear on the crop by deposition of water
vapor or freezing of dew. Both processes release heat and therefore delay freezing
of crop tissues. If the amount of ice is large the term white frost is used. When the
concentration of water vapor in the air is very low (dew point below the minimum
Table 29.2 Critical temperature (ºC) values for several deciduous fruit tree crops and grape vines.
The 10 % kill and 90 % kill imply that 30 min at the indicated temperature is expected to cause
10 % and 90 % kill of the plant part affected during the indicated phenological stage. The values
for bloom are the average from early to late bloom (Adapted from Proebsting and Mills (1978) and
Snyder and De Melo-Abreu (2005))
Crop Stage 10 % kill 90 % kill
Apple Silver tip 11.9 17.6
Bloom 2.4 3.9
Peach First swell 7.4 17.9
Bloom 2.8 4.9
Pear Scales separate 8.6 17.7
Bloom 2.9 5.3
Grape First swell 10.6 19.4
Bud burst 3.9 8.9
First leaf 2.8 6.1
Fourth leaf 2.2 2.8
14
18h
12 22h
10 04h
HEIGHT (m)
06h
8
08h
6
0
–2 0 2 4 6 8
TEMPERATURE (°C)
Fig. 29.2 Development of a temperature inversion over an apple orchard in Northern Portugal
(Adapted from Snyder and De Melo-Abreu (2005))
temperature), there is no dew to freeze and no possibility of deposition and the

tissues are affected without prior formation of ice, causing necrosis of the tissues
(“black frost”). The presence of white frost indicates that damage may occur, but
black frost is the visualization of the damage that already occurred.
5
TEMPERATURE (°C) WIND SPEED (m s–1) 4
3
2
1
0
–1
–2
–3
–4 U
t
–5
–6
0 2 4 6 8 10 12 14
TIME (hours after sunset)
Fig. 29.3 Time course of temperature and wind speed over bare soil during the night. Espiel
(Spain), February 3–4, 2012
29.4 Climatology of Frosts
The frost-free period is the time between the last frost (late winter or spring) and the
date of the first autumn frost. This period is highly variable from year to year, so it is
of limited value to assess the risk of frost damage, which should be based on the
frequency distribution of frost dates. The dates of the first and the last frost may be
considered as independent random variables that follow the normal distribution.
This allows the calculation of the probability of frost during specific periods. For
instance, the probability of spring frost after a given day is:

t mLF
Pðfrost after day tÞ ¼ Py P z > ð29:1Þ
sLF
where Py is the fraction of years when frost occurs, mLF is the mean date for the last
frost, sLF is the corresponding standard deviation and z is the standard normal
distribution which can be calculated using tables or the following approximate
function:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi!
2x2
Pðz xÞ ¼ 0:5 1 1 exp ð29:2Þ
π
where the positive root is used if x >0 and the negative root when x <0. We have to
remember that Pðz > xÞ ¼ 1 Pðz xÞ.
Similarly, the probability of autumn frost before a given day is:

t mFF
Pðfrost before tÞ ¼ Py P z < ð29:3Þ
sFF
where mFF and sFF are the mean and standard deviation for the date of the first frost.
Note that these statistics are only computed for years when frosts occur.
Example 29.1 The dates of the first and last frost (expressed as days from
September 1) during 15 years are given in Table 29.3 for two locations
(Gibraleon and Jerez del Marquesado) in Southern Spain.
Table 29.3 Dates of the first and last frost (expressed as days from September 1) during
15 years for two locations (Gibraleon and Jerez del Marquesado) in Southern Spain. The
minimum temperature observed each year is also shown
Gibraleon Jerez del Marquesado
Date of frost Tmin Date of frost Tmin

Year First Last C First Last C
2000 2 69 181 2.9
2001 1.7 70 185 5.7
2002 133 133 0.6 94 217 8.6
2003 0.7 88 223 7.3
2004 119 168 3.9 73 223 15
2005 166 166 0 75 222 10.2
2006 147 147 0.6 98 219 7
2007 1.2 77 212 7.1
2008 131 132 2.6 60 226 6.1
2009 106 242 1 45 247 5.9
2010 156 156 0.2 77 197 5.4
2011 122 167 3.1 104 229 10.4
2012 0.8 88 243 4.7
2013 1.1 76 208 4.6
2014 123 129 0 96 207 4.7
Average 133.67 160.00 0.30 79.33 215.93 7.04
Std. deviation 19.24 34.52 1.75 15.65 18.54 3.02
In Gibraleon no frost occurred in 6 out of 15 years, so the statistics for the

mean and standard deviation (shown also in Table 29.3) are computed for the
remaining 9 years and Py ¼ 0.6. The mean dates for the first and last frost are
day 134 (January 12) and 160 (February 10). The probability of frost after
March 1 (day 182) will be:
(continued)

t mLF
Pðfrost after day 182Þ ¼ Py P z >
sLF
182 160
¼ 0:6 P z > ¼ 0:6 P½z > 0:64 ¼ 0:6 ð1 P½z 0:64Þ
34:5
¼ 0:6 ð1 0:74Þ ¼ 0:156
Now we will calculate the probability of frost before December 1 (day 92):

t mFF 92 133:7
Pðfrost before 92Þ ¼ Py P z ¼ 0:6 P z
sFF 19:24
¼ 0:6 P½z 2:17 ¼ 0:6 0:012 ¼ 0:007
In Jerez del Marquesado the mean dates for the first and last frost are day
79 (November 18) and 216 (April 4). The probability of frost after March
1 and before December 1 are 97 and 80 %, respectively.
29.5 Risk of Extreme Cold Temperatures
Historically, farming has been pushed towards the environmental limits where risks
of extreme events are on the increase. Many agricultural decisions have to be based
on the probability of damaging events that can kill the plants or reduce yield
substantially thus making farming unviable. For frost risk analysis we distinguish
the probability P(T<Tc) of occurrence of temperature below a critical threshold in
any year and the risk (R) which is the probability of the event occurring at least once
over a design period (nd) (nd would be the expected duration of the orchard in years,
for example). Instead of risk we may use certainty (C¼1R) which is then the
probability of the event not occurring over the design period. Assuming a Bernouilli
distribution, the certainty (C) is related to the probability of having a temperature
below Tc in any given year:
C ¼ ½1 PðT < T c Þnd ð29:4Þ
For example, if the probability of temperature below 10 C in any given year is
0.003 (i.e. it happens three times in 1,000 years) then the certainty for 20 year
project duration is 0.94, i.e. we are 94 % certain that temperatures will never fall
below 10 C in 20 consecutive years.
The probability of an extreme event occurring in any given year should be
calculated as the ratio of the observed extreme events over the number of years
of record. As they are rare events we would require a very long weather record
(e.g. more than 1,000 years) which is never available. Instead, for limited data sets
we calculate the parameters of the underlying statistical distribution. Haan

recommended the type I extreme value (Gumbel) probability distribution:

Tc β
PðT < T c Þ ¼ 1 exp exp ð29:5Þ
α
where α ¼ σ/1.283, β ¼ μ + 0.577 α, μ is the average of the minimum temperatures

recorded each year and σ is the corresponding standard deviation. The parameter β
is the mode (most frequent value) of the distribution.
From the equations above we may deduce another for calculating the certainty:
nd
Tc β
C¼ exp exp ð29:6Þ
α
Example 29.2 The minimum temperature in Jerez del Marquesado (Spain)

has an average of 7.04 C with standard deviation 3.02 C (Table 29.3).
Therefore the parameters of the Gumbel distribution are α ¼ σ/1.283 ¼
2.355 C and β ¼ μ + 0.577 α ¼ 7.04 + 0.577∙2.355 ¼ 5.68 C. If we are
planning to establish an orchard during 20 years and the critical temperature
is 12 C the certainty will be:
20
12 ð5:68Þ
C¼ exp exp ¼ 0:255
2:355
which is very low. This value indicates that the risk of failure of our orchard is
75 %. In the other location (Gibraleon) the certainty will be 0.998 which
indicates a negligible risk for the orchard.
29.6 A Simple Model for Nocturnal Surface Cooling
Soil surface cooling during the night may be analyzed using the energy balance of
the soil (Chap. 6):
Rn ¼ C þ H þ LE þ G ð29:7Þ
where Rn is net radiation, C is heat released by freezing of water, H is sensible heat

flux and LE is latent heat flux. Combining the equations of heat flux in the soil
(Chap. 6) and the loss of long wave radiation (Chap. 3) Brunt proposed a very
simple model to calculate the variation of surface temperature during the night. The
assumptions include LE ¼ 0, H ¼ 0 and an isothermal soil profile at sunset. This
implies that soil cooling during the night is equivalent to the loss of long wave
radiation while the other processes are ignored. Brunt also assumed that net
radiation does not change during the night.
The analytical solution for Ts0 is
2 1 pffi
T s T s0 ¼ pffiffiffi pffiffiffiffiffiffiffiffiffi Rn t ð29:8Þ
π k Cv
where Ts0 is surface temperature at sunset, CV is specific heat of the soil per unit
volume (J/m3/K), k is thermal conductivity (W/m/K) and t is time elapsed after
sunset (s). Note that the square root of k Cv is the thermal admittance (Chap. 6).
This equation implies that temperature during the night will decrease in proportion
to long wave radiation loss and the square root of time after sunset. Soils with low
admittance (sandy, dry) will cool faster.
Example 29.3 For a loam soil at Permanent Wilting Point admittance is

1,057 W m2 s1/2 K1 (Table 6.2). For Rn ¼ 70 W/m2, the drop of temper-
ature after sunset would be 10, 14.2 and 17.4 K after 5, 10 or 15 h, respec-
tively. If the soil was at saturation (thermal admittance 1,835 W m2 s1/2 k1)
the temperature would decrease only 5.8, 8.2 and 10 K.
The above example illustrates the important effect of soil water content on
nocturnal cooling.
29.7 Frost Protection
Frost protection methods include those methods that are implemented before the
frost night in order to avoid or minimize frost damage (i.e., passive, indirect, or
preventive), and methods that are implemented during the frost night (i.e., active,
direct, protective). Often, the effect of passive methods, which are relatively cheap,
adds up to the effect of the active methods. Therefore, passive methods should
always be considered and, when suitable, implemented in conjunction with one or
more active methods, or in isolation. A complete description of most of the existing
methods and related computational tools is available in Snyder and De Melo-
Abreu (2005).
29.7.1 Passive Protection Methods
29.7.1.1 Site Selection
Section 29.6 described the physics of the cooling process in a specific location
ignoring the horizontal movement of air. As the air cools, its density increases, and
will tend to flow to areas of lower density, typically downwards to valleys and
depressions. The degree of accumulation of cold air or ventilation depends on the

topography, the wind speed and the temperature gradients.
Frost sensitive crops should not be placed in locations where cold air accumu-
lates due to orography (depressions) or to the existence of artificial obstacles
(fences, windbreaks).
The presence of large bodies of water (lakes, sea) in the direction where cold
winds come from can reduce the risk of frost due to heat exchange between water
and air.
If possible the most critical areas should be detected using maps of minimum
temperatures which could be obtained using remote thermal infrared imagery. As a
rule of thumb places where radiation fog is more frequent are also those where
radiation frosts are more likely.
29.7.1.2 Selection of Species, Cultivars and Cultural Techniques
Crop species differ in their sensitivity to cold and frost, and genetic variability may
exist within each species. Taking into account the climatology of frosts in the
location and the critical temperatures for the crop alternatives to that location, we
will choose the species and the cultivar in order to reduce the risk of frost damage.
Some cultural techniques may be beneficial, when they explore the knowledge
of the biometeorology of frost. In deciduous fruit trees, pruning is usually done in
winter but, in locations prone to severe frost, late pruning is advisable from the
viewpoint of frost damage prevention, since plants are more sensitive just after
pruning and the probability of severe frosts decreases as the spring approaches. In
the same regard, pruning promotes bud burst, hence late pruning exposes new
growth to less-frequent and less-severe frosts. Training pruning techniques that
elevate the level of tender organs are beneficial because in frost nights there is,
usually, a temperature inversion, which results in lower levels being colder.
Delayed bud break has been achieved in pome fruits by periodic overhead irrigation
in late winter that cools the buds and delay their development until past the most
frost risky period. Nitrogen fertilization and high water status tend to elevate the
critical temperature. Hence, when there is a strong probability of frost in the
upcoming days it is not wise to N-fertilize or irrigate (but see also next section).
Some bacteria called Ice Nucleating Active (INA) may act as freezing nuclei and
therefore initiate the freezing process. These bacteria are often concentrated on the
cover crops and weeds present in the orchard and their removal may help in
preventing frost damage.
29.7.1.3 Soil Management
Minimum surface temperature may be increased by increasing the thermal admit-

tance (irrigating, compacting the soil) or increasing the radiant energy reaching the
soil during the daytime (avoid opaque mulches or cover crops). Soil heating during
the day depends also on the partitioning between G and H (Chap. 6) so a smooth soil
surface (high aerodynamic resistance) will improve soil heating as compared to a
rough soil surface. Therefore tillage operations (reduce thermal admittance,
increase surface roughness) are not desirable if frost is expected.
Irrigation increases soil thermal admittance, which reduces nocturnal cooling
but reduces also diurnal soil heating as the energy spent in evaporation from the soil
surface increases (Chap. 6). The best situation would be to have a wet soil covered
with transparent plastic. If that’s not possible the best choice is to irrigate some days
in advance to let the upper soil layer dry, and thus reduce soil evaporation, while the
rest of the profile is wet (high thermal admittance).
Soil heat flux is also increased after removal of cover crops or weeds in orchards.
Their removal should preferably be done by using herbicides that eliminate the
cover but do not include tillage and the concomitant change in soil surface bulk
density. When tillage is used it should be done well in advance of the sensitive frost
period for the soil to settle and the residues to decompose.
29.7.2 Active Protection Methods

29.7.2.1 Increasing Radiation Interception
Radiation frosts occur when long wave radiation loss is high (absence of clouds,
low air humidity). Thermal radiation may be intercepted partly to reduce long wave
loss by spraying water. For the water spray to be effective, the diameter of the
droplets or particles should have a diameter of the same order as that of thermal
radiation (8–12 μ).
Artificial clouds of smoke may be produced by burning different materials (tires,
wood or fuel) but are inefficient and therefore rarely used nowadays (small diameter
of particles, rapidly vanishing, high energy cost, pollution problems).
Another way to trap the longwave radiation is by using commercial solid acid
clouds or aerosols that have a suitable particle size and are produced in situ by
combining several products.
Some materials which are almost opaque to long wave radiation (e.g. thermal
blankets) may be used to cover high value crops.
29.7.2.2 Air Mixing
Inverted temperature profiles typical of nights with radiation frost may be homog-
enized by mixing air of different heights, thereby increasing the temperature at
canopy height. The effectiveness of air mixing will be proportional to the temper-
ature gradient.
Although helicopters have been used, the best choice for air mixing is a wind
machine which consists of a steel tower with a large rotating fan near the top. Fans,
with blades of diameter between 3 and 5 m, are located about 10–11 m above
ground and are oriented to blow at a slight downward angle (e.g. 7 ) to improve
mixing.
29.7.2.3 Heating the Air or the Canopy
The losses of energy from a crop during frost may be compensated burning fuel
(solid, liquid or gas) in heaters, which transfer energy by thermal radiation and
convection. The energy loss from the crop is usually in the range 20–40 W m2,
while input from heaters is typically between 140 and 280 W m2, which indicates
a very low efficiency. The best conditions for this method are no wind and the
presence of a strong inversion.
The temperature of air leaving the heater is very high, so it will rise rapidly
mixing with colder surrounding air, until it reaches the height where the air has the
same temperature. Eventually, the mixed air will cool, become denser and descend,
which creates a circulation pattern within the inversion layer. When there is a strong
inversion (i.e. a low ceiling), the heated air rises to a lower height and the volume
influenced by the heaters is smaller so efficiency is higher. Efficiency is low when
heaters are too big or hot as the warmed air can break through the top of the
inversion layer.
29.7.2.4 Irrigation
Irrigation is a very useful tool for protecting crops against frost and it relies mostly
on the release of heat by water cooling (4.18 103 MJ/K/kg) and by freezing
(0.334 MJ/kg), but some heat may be spent in evaporation. In the best case, when
we irrigated with water extracted from wells it has a temperature close to the mean
annual air temperature at the site so the contribution of water cooling is very small
compared with that of freezing. This is the basis of using sprinkler irrigation for
frost protection which usually requires lower application rates (ca. 1 mm h1) than
typical sprinkling systems for water supply to the crop. One needs to supply water
continuously to the whole area to be protected, so that there is always a thin layer of
water over the ice that is formed, thus keeping the temperature at the freezing point.
Surface irrigation may also be used and is also based upon the same principles (use
of the heat of fusion and specific heat of water), but the heat is liberated to the air
near the soil surface, and it may not be sufficient to prevent frost damage.
Bibliography
Haan, C. T. (1979). Risk analysis in environmental modifications. In: Barfield, B. J., & Gerber,
J. F. (Eds.), Modification of the aerial environment of crops. American Society Agriculture
Engineer (ASAE) (pp. 30–51), Monograph No. 2. St Joseph.
Proebsting, & Mills. (1978). Journal of American Society Horticulturae Scientia, 103, 192–198.
Hoboken: Wiley.
Snyder, R. L., & de Melo-Abreu, J. P. (2005). Frost protection: Fundamentals, practice and
economics (Vol. 1). Rome: FAO.
Snyder, R. L., de Melo-Abreu, J. P., & Matulich, S. (2005). Frost protection: Fundamentals,
practice and economics (Vol. 2). Rome: FAO.
Chapter 30
Control of Weeds and Other Biotic Factors
Abstract Weeds are plants whose presence is undesirable at a time and/or place
because they compete with crops for resources, deteriorate the quality of the
harvested product and can hinder harvesting. The most important weed species
include C4 perennials with vegetative propagation. Usually weeds are able to
produce many seeds that often present dormancy, which generates a soil seed
bank which germinate over many years. This prevents weed eradication and forces
us to use control techniques to keep weed populations at tolerable levels. Weeds
adapt in a few years to the cropping system, in particular to control methods.
Control techniques include the use of herbicides and of cultural practices such as
tillage, mulching, mowing, and crop rotation. Crop management has an important
effect on the incidence of pests. Irrigation method and irrigation frequency deter-
mine the germination of weeds and influence the infection by aerial or soil patho-
gens. Biological control is effective with invasive weeds and some insects. The
ability of weeds to evolve in response to the selection pressure exerted by control
measures forces us to establish long-term strategies (weed management) which
should be based on detailed knowledge of the ecology of the weed species. Then
different types of control should be alternated to improve the efficiency of control.
30.1 Introduction
Weed control has always been an important part of agricultural practices and is
often considered as part of agronomy. Traditionally, only manual weeding was
specifically aimed to control weeds, but many practices, such as tillage, burning and

L. Mateos

Agriculture, DOI 10.1007/978-3-319-46116-8_30
rotations, contributed in some way to this control. In 1944, when 2,4-D was
introduced as a herbicide, Weed Science appeared as a discipline within the
techniques of Plant Protection, although much more tied to crop production
techniques.
It is estimated that not more than 250 species have become important weeds for
agricultural production.
In this chapter we will review the main ecological characteristics of weeds and
the methods of control with the exception of pesticides which will be dealt with in
next chapter. For some specific control techniques we will also mention their
impact on other pests.
30.2 Characteristics of Weeds
A weed is a plant growing where it is not wanted, i.e. in the wrong time or location.
This is a relative definition, as seeds left from a crop after harvest may lead to weeds
for the next crop. However, the most troublesome weed species usually show
special characteristics that allow their dispersal and persistence in agricultural
systems and increase their ability to compete with crop plants.
30.2.1 Dispersal of Weeds
Weeds use the general dispersal mechanisms of plants (wind, animals, water,
gravity) but also present among their invasion strategies the so called anthropocory
(dispersal by human action). The commerce of agricultural products, seeds, and
other materials has contributed to the dispersal and homogenization of weeds
worldwide. The most damaging species are found in many environments and
systems all around the world. Table 30.1 shows the most important weed species
on a global scale. Most of this species are C4 perennials.
The invasion of a new site is usually based on a small but continuous flow of
propagules transported from short distances. However, the most effective dispersal
strategy of a weed is based on adaptations that ensure the return of weed seeds
along with the crop seed at the time of planting.
30.2.2 Persistence of Weeds
The persistence of weed populations is often guaranteed by the production of a

large number of seeds per plant which often exhibit dormancy mechanisms. These
allow germination to spread over a long period (years) and ensures the survival of
the population.
30 Control of Weeds and Other Biotic Factors 461
Table 30.1 Top 10 worst weed species on a global scale. The number of crops and countries
where it is consider an important weed are also presented
Species
(photosynthesis Common #
type) name Crops #Countries Cycle Propagules
1 Cyperus rotundus Purple 52 92 Perennial Rhizomes with
(C4) nutsdege tubers
2 Cynodon dactylon Bermuda 40 80 Perennial Rhizomes with
(C4) grass stolons
3 Echinochloa crus- Barnyard 36 80 Annual Seed
galli (C4) grass
4 Echinochloa colona Jungle rice 32 60 Annual Seed
(C4)
5 Eleusine indica (C4) Indian goose 46 60 Annual Seed
grass
6 Sorghum halepense Johnson 30 53 Perennial Rhizomes and
(C4) grass seeds
7 Imperata cylindrica Cogon grass 35 73 Perennial Rhizomes
(C4)
8 Eichhornia crassipes Water 1 Perennial Stolons
(C3) hyacinth
9 Portulaca oleracea Purselane 45 81 Annual Seeds, rooting
(C4) at nodes
10 Chenopodium album Lambs 40 47 Annual Seed
(C3) quarters
The community of buried propagules (soil seed bank) consists of rhizomes,

stolons, bulbs and seeds of different species. The seeds entering the soil may be
dormant or enter into secondary dormancy. The most important factors for germi-
nation or dormancy release are: water, temperature, light, NO3, O2 and CO2.
The permanence of some form of dormancy in buried seeds is an adaptive
advantage as it allows adjusting seed germination to ecological conditions favor-
able to the survival of seedlings.
The loss of seed dormancy in the soil occurs especially if the water content and
aeration are adequate. Germination is very sensitive to the quality of light (phyto-
chrome system), especially to that resulting from radiation interception by vegeta-
tion. Thus, by detecting the light quality, the seed not only receives information
about the depth at which it is buried, but can also detect the presence of a canopy
above, which may limit the growth of the seedling. Apart from that, some species
respond to short light pulses as those that the seed may experience during secondary
tillage.
Other factors contributing to end the seed dormancy is a low concentration of
CO2, a high concentration of nitrates and the oscillation of temperature. The latter
allows the seed to “detect” its depth in the soil because the temperature range
decreases with depth. A buried seed which is subjected to a very wide temperature
variation is probably near the surface and it is likely to stay on a clear area without
vegetation, since the presence of the latter buffers the variation of soil temperature.
On the other hand, weeds often show a great capacity for acclimation and
adaptation. In many cases this is associated to annual cycles, reduced number of
chromosomes and self- pollination or vegetative propagation.
The most important selective force acting on weeds in agricultural systems is
human action including type, depth and timing of tillage, sowing date, timing and
type of herbicide application, etc. There are selection pressures that select weed
genotypes suitable to endure in the system due to adjusted mechanisms of dor-
mancy and germination, and, more important, that there are similarities to the crop
in height, seed size and maturity period to ensure the joint harvest with the crop. It is
clear then that the main determinant of weed flora in a given area will be the main
crops grown in addition to management practices.
In some cases, besides selective forces, the system provides genetic information
that can contribute to the evolution of the weed (crossing between weeds and crop
plants).
Unlike insects and fungal pathogens weed populations do not suffer sudden
changes in population density due to the damping exerted by the seed bank.
Therefore, weeds are usually a chronic but not an epidemic problem, but the
variability is large. Some species with high reproductive potential but with low
seed dormancy and low persistence in the soil, such as Alopecurus myosuroides
(green foxtail) are typically aggressive opportunistic invaders, but are also easily
removable. By contrast, other species with low reproductive potential but with
strong dormancy in seeds, as Veronica hederifolia (speedwell), tend to persist and it
is difficult to alter the size of their populations. These differences in the population
dynamics of different weed species have implications for their control which should
try to eradicate populations of the first type with intense short-term measures while
maintaining reduced population sizes for the second type.
30.3 Classification of Weeds
According to their life cycle weeds are classified as:

– Annual weeds: they complete their life cycle in 1 year. We may distinguish two
groups:
i. Winter annuals, that germinate in the fall and set seed in spring or summer.
They are usually found in winter cereals. Examples: Avena sterilis (wild oat),
Sinapis arvensis (wild mustard).
ii. Summer annuals that germinate in spring and set seed in autumn.
Examples: Amaranthus retroflexus (redroot pigweed), Portulaca oleracea
(common purslane).
– Biannual weeds: they require 2 years to complete the cycle, devoting the first
year to vegetative development and storage of carbohydrates (rosette stage in
many species) and flowering in the second spring.
Example: Thistles
– Perennial weeds have cycles of several years: in some cases they show vegeta-
tive reproduction from organs (roots, stems, rhizomes, stolons, tubers, bulbs)
that remain dormant until suitable conditions for sprouting occur.
Examples: Cyperus rotundus (purple nutsedge), Cynodon dactylon (Bermuda
grass) and Sorghum halepense (Johnson grass).
Weeds can also be classified according to the habitat where they are usually
found (crops, pastures, orchards or forests, surface waters, roadsides and waste
places).
30.4 Crop-Weed Interactions
Weeds always compete for resources (water, light, nutrients) and often show certain
competitive advantages such as high density, earlier emergence or high early vigor.
Some morphological (taller plants, deeper roots) or physiological (C4 photosyn-
thesis, allelopathy) mechanisms may also enhance weed growth when in competi-
tion with crop plants.
Apart from yield losses due to competition, weeds have other negative effects
such as hindering harvest or degrading crop quality by altering the color, smell,
taste or adding toxins.
In all cases the level of competition and thus of yield loss will be directly
proportional to the earliness of weed emergence relative to that of the crop, as
discussed in Chap. 12. Therefore it is difficult to establish general relationships
between crop yield and weed density (Dpw). If we use the reciprocal yield law
(Chap. 12) we may calculate yield (kg/ha) as:
10 HI Dp
Y¼ ð30:1Þ
b1 þ b2 Dp þ b2w Dpw
where b1 and b2 are empirical coefficients that determine the response of the crop to
plant density. The factor 10 in Eq. 30.1 is used to convert g/m2 to kg/ha. Now we
have added a term (b2w Dpw) that incorporates the competition of the weed. If the
crop and the weed are very similar in form and cycle then the coefficients b2 and b2w
should be the same.
Example 30.1 Let’s calculate the yield of a cereal crop as a function of weed
density. We’ll asume that:
b1 ¼ 0:01 plant=g, b2 ¼ b2w ¼ 0:001m2 =g, HI ¼ 0:4 and Dp

¼ 200 plants=m2 :
According to Eq. 30.1 yield may be calculated as:
10 HI Dp 800
Y¼ ¼
b1 þ b2 Dp þ b2w Dpw 0:01 þ 0:20 þ 0:001Dpw
Values of yield for weed densities of 0, 100 and 200 plants m2 would be
3,810, 2,581 and 1,951 kg/ha.
Many studies indicate that the 30–40 days after emergence are critical for
determining yield losses due to weeds. If the crop is kept clean during this period
any later invasion will not cause yield reduction but may lead to other problems
(e.g. harvest problems). On the other hand, if weeds are completely removed at the
end of this period, yield reduction should be small.
30.5 Economic Threshold
An economic threshold or action threshold for a given pest is the density of the pest
at which control should be applied. Otherwise the pest will reach a higher density
level (called Economic Injury Level) at which the cost of control equals the
economic loss due to the yield reduction effected by the pest, quality loss or
harvesting difficulties. Note that if the action level is exceeded there will be an
economic loss. For weeds the action threshold may be equal to the EIL but for
insects it may be much lower if the insect population increases rapidly.
Using Eq. 30.1 we may calculate the yield loss (YL, kg/ha) for weeds as:
10 HI Dp 10 HI Dp
YL ¼ ð30:2Þ
b1 þ b2 Dp b1 þ b2 Dp þ b2w Dpw
One important difference between weeds and other pests is that weeds have a
negative effect for any density as they always use resources (water, nutrients).
Other pests (e.g. insects) may not have any negative effect in terms of yield loss
when populations are low.
If the selling price of the harvest is PY (euro/kg) and the cost of weed control is
CH (euro/ha), then the economic threshold (taken as equal to EIL) corresponds to a
yield loss of CH/PY. Using Eq. 30.2 we may deduce that it will occur when the weed
density is:
10 HI Dp CH
Dpwu ¼ ð30:3Þ
b2w Y x ðPY Y x CH Þ
where Yx is yield in the absence of weeds. We see that the economic threshold is
directly proportional to planting density and to the cost of weed control, while it is
inversely proportional to weed competitive ability (b2w), to crop yield and to selling
price.
Example 30.2 Using the data of Example 30.1 we may calculate the eco-
nomic threshold if the grain price is 0.25 euro/kg and the cost of weed control
is 150 euro/ha:
10 0:4 200 150

Dpwu ¼ ¼ 39 plants m2
0:001 810ð0:25 3810 150Þ
In this case control measures should not be taken unless weed counts exceed
39 plants m2.
The economic threshold thus defined is valid in the short term as it affects only
the current crop. Lower values of economic threshold will result if one considers a
longer perspective. For example, in organic farms in the Netherlands the main
problems of weeds in certain crops (e.g. onion) are due to seeds from some weed
plants that had not been controlled in the previous wheat crop. This implies the need
for more intensive weed control during the wheat season at a higher cost that this
crop would need.
Some species (e.g. Chenopodium album) are easy to control with hormonal
herbicides but difficult to eradicate, due to their long seed persistence. In these
cases we may use the short term economic threshold. In other species that are
difficult to control with herbicides but have little seed persistence (e.g. Avena
sterilis) it is better to apply intense control measures using a long-term economic
threshold.
The effect of a given weed density on yield depends on many factors, in special
the time of weed emergence relative to the crop. Furthermore weeds can be
gradually emerging, which further complicates the prediction of its ability to
compete, given also that there is substantial spatial variability in the distribution
of weeds within a field. Thus the empirical equations between yield and weed
density have a limited validity, and economic thresholds deduced from them, too.
The concept of economic threshold may be difficult to apply to many situations
as the effect of pest densities on crop yield has to be known a priori. However it
serves to illustrate some important concepts on pest control. First, low population
levels of the pest do not have a negative effect on farmer’s profits, i.e. complete
suppression of the pest is not usually the best alternative, unless we can ensure long
term eradication. Second, the tolerable population density depends on both biolog-
ical (competitive ability of weeds, damage capacity of insects, response of plants to
damage, etc.), economic (market value of crop, cost of control) and agronomic
(e.g. planting density) factors.
30.6 Pest Control
Pest control aims at reducing populations of pests to acceptable levels.

The complete elimination of weed plants and seeds is expensive and almost
impossible due to the soil seed bank. Fumigants will control nematodes, soil borne
fungi and bacteria, soil insects, and weeds (both weed seeds and germinating
seedlings). Their use is restricted by their high cost and the toxicity risks. A less
aggressive option than to fumigate the soil is called soil solarization whereby a
transparent plastic sheet is placed on the soil surface following irrigation in
summer. The high temperatures (>60 ºC) and high humidity that is generated by
the treatment eliminate most of the soil borne pests down to about 30 cm or deeper.
As eradication of weeds is not possible we need to keep their populations low by
preventing the entry of new propagules to the field and by reducing their population.
Prevention is based on using seed free of pest propagules and viruses and clean
machinery and keeping the field margins free of dangerous weed species which may
also host other pests.
30.6.1 Mechanical Control
Tillage in general stimulates the germination of weed seeds in the soil. The main
effects of tillage on emerged weed plants are due to the burial of the aerial part
(reduced assimilation), mechanical wounding of shoots and roots and uprooting of
the weed that then dies by desiccation. Tillage is effective against annual weeds, but
may increase infestation of creeping perennials by breakage and spreading of
propagules (rhizomes, stolons) when the soil is wet. The effect of tillage depends
on the type and depth, with deep moldboard being very effective in killing plants by
burial but also promoting the transfer of seeds from deep to surface layers. Vertical
tillage will act mostly by mechanical wounding but has little effect on seed
distribution.
In earlier times the only measures available for weed control in growing crops
were pulling, harrowing and hoeing. Pulling is very effective against annuals and
tap-rooted plants but may not be so if the plant is able to re-sprout from root
segments. Therefore its effectiveness depends on the ability to remove as much root
system as possible, which is quite difficult in perennials.
Mowing reduces the growth of weeds and is very effective in preventing seed
production. However, mowing alone selects creeping genotypes or species that
escape control.
Burning stubble or crop residues has been used since ancient times not only to
control weeds, insects and pathogens but to facilitate seed bed preparation. How-
ever, burning is being restricted in many areas because of several disadvantages
(wildfire risk, smoke control, loss of organic matter and nitrogen). High tempera-
tures kill seedlings and may kill or reduce the viability of weed seeds, insects and
fungi close to the soil surface. This may also be achieved by solarization as
discussed above. Another possibility is flaming of weeds using a torch that directs
the flame to the weed for a short period, which causes plant death after some time
and is more effective on weed seedlings.
Flooding prevents the germination of seeds and kills submerged plants. It is the
basic weed control measure in continuous flooded rice crops.
Opaque mulching will prevent seed germination and kill weeds by carbon
starvation. For that one may use black or gray plastic films or thick layers of
other materials (gravel, sand, sawmill residues).
30.6.2 Other Control Techniques
Crop management affects weed populations in many complex ways. The crop
rotation may help in reducing weed problems, e.g. alternating winter and spring
crops. Rotations are also very effective for reducing the incidence of soil borne
diseases. The management of residues plays also a major role in the control of pests
as they serve as a reservoir for pest propagules.
Irrigation management also affects pests. The germination of weeds will not
occur in dry soil, thus partial wetting or underground irrigation will be helpful. On
the other hand preplant sprinkler irrigation may be used to induce germination of
weeds which are then controlled using herbicide or shallow tillage. This system is
called stale or false seed-bedding. Some soil borne fungi are promoted by contin-
uous wetting of the soil so reducing the frequency of irrigation may be an effective
control measure. Wetting of plants shoots promotes the infection by aerial fungal
diseases (e.g. rusts) but may help in controlling other pests (e.g. mites).
Selection of adequate genotypes may help in improving the competitive ability
of the crop. In the case of aerial diseases the use of resistant genotypes is one of the
major alternatives for control (see Sect. 30.7). Cultivars which are Genetically
Modified Organisms (GMO) have been bred that include resistance to certain
herbicides (e.g. soybean resistant to glyphosate) thus allowing the application of
the specific herbicide to the field areas where weeds appear after the crop has been
established. In other cases, bacteria (Bacillus thuringiensis) genes that produce a
toxin have been inserted in cultivars of some major crops such as corn and cotton,
and these cultivars produce the Bt toxin that kills Lepidoptera larvae feeding on the
crop. In all cases, weed and insect control costs have been greatly reduced in the
GMO cultivars.
Planting density also affects the incidence of pests. On the one hand a high
planting density may compensate for plant losses due to pests. On the other, crops
with high LAI suffer a more humid microenvironment and stay wet for longer after
wetting which increases the incidence of aerial pathogens.
Biological control involves the introduction of organisms that are consumers,
pests or diseases of other pests. The agent should not affect the crop and should be
able to adapt successfully in the area where it is introduced. The most famous
example of biological control of weeds was that of Opuntia stricta in Australia
which was controlled successfully using a moth (Cactoblactis cactorum). Insect
control may be performed by Coccinellidae (ladybugs, ladybirds) which feed on
aphids and scale insects.
Chemical control is one of the main alternatives for pest control and will be
considered in detail in the next chapter.
30.7 Using Cultivars with Resistance to Pests
One of the best alternatives for reducing the impact of insects and diseases is using
resistant cultivars. We distinguish two types of resistance:
(a) Qualitative or vertical resistance is that controlled by one or a few genes so
we find distinct resistant and susceptible cultivars. The continuous use of the
resistant cultivar may lead to the appearance of a new race of the pathogen
for which resistance is lost, forcing the development of a new resistant
cultivar.
(b) Quantitative or horizontal resistance is that controlled by many genes so there is
a continuous variation in the level of resistance of different genotypes. This
type of resistance holds for many races of the pathogen so it will last much
longer than vertical resistance, but is difficult to transfer between genotypes.
We can also distinguish between resistance, if the pest does not infect the host
plant, or the infection is very limited, and tolerance, when infection occurs but the
impact on yield is very low.
Several alternatives may be used to reduce the selection pressure on the patho-
gen so the resistance holds longer:
(a) Alternating resistant and susceptible cultivars in the crop rotation.
(b) Sowing a mix of resistant and susceptible cultivars
(c) Using a multiline which is an ensemble of cultivars with resistance to different
pathogen races
(d) Including the resistance to different races in a single cultivar
Options a and b are not very attractive to farmers, while options c and d require
huge efforts by breeding companies, so in the end plant breeders keep track of new
pathogenic races and breed for new cultivars in a never ending race against the
adaptation of pathogens. This is called maintenance breeding and is essential for the
sustainability of agriculture even though it is not high in the priority list of
agricultural research in many countries.
30.8 Weed Management
Weed management is a set of strategies of weed control aimed at controlling the

weed populations in the long term (years) in a cropping system. It involves a more
comprehensive and longer term approach than weed control. This technology is
based on better knowledge of the ecology of weed populations, their critical periods
for the formation of propagules and their interactions with cultural practices. With
that knowledge it is possible to design long-term strategies to keep weed
populations at acceptable levels from the point of view of crop production. For
example, models of germination of weeds may be used to adjust the dates of tillage,
herbicide application or crop planting before the emergence of the weeds.
Another important aspect of weed management is to develop long-term strate-
gies that are necessary if we look at weeds as a phenomenon at the farm or even at
the agricultural system level. In this case we should be concerned about the
production and dispersal of propagules. For instance, the better the separation
between crop and weed seeds in the combine, the better the dispersal of the weed
will be.
Strategies for long-term management must also consider the evolution of weeds
which may be very fast (several years) in response to the strong selective pressure
exerted by agricultural practices (including herbicides). One interesting choice is
alternating control measures. An example would be the application of herbicides for
several years leading to a reduction of genetic variability, which often leads to very
strict temperature requirements for germination. After several years we may exploit
the characteristics of the target population using tillage at the right time or by
changes in cultural practices (e.g. earlier sowing). In this second phase the weed
population is selected to broader forms which include greater sensitivity to
herbicides.
Bibliography
Pimentel, D., & Peshin, R. (2014). Integrated pest management. Pesticide problems (vol 3).
Dordrecht: Springer.
Radosevich, S. R., Holt, J. S., & Ghersa, C. M. (2007). Ecology of weeds and invasive plants:
Relationship to agriculture and natural resource management (3rd ed.). Hoboken: Wiley.
Zimdahl, R. L. (2013). Fundamentals of weed science (4th ed.). New York: Academic.
Chapter 31
Application of Herbicides and Other Biotic
Control Agents
Abstract The main characteristics of pesticides to consider from an agronomic

viewpoint are selectivity, mobility within the plant and toxicity on non-target
plants, insects and other fauna. Pesticide application should be aimed at maximiz-
ing the protective effect while preventing drift, which is increased with high wind
speed, unstable conditions and high evaporative demand. There is a general trend
towards using molecules in pest control that have less persistence and less negative
impacts on the environment. Pesticide doses for trees may be adjusted depending on
the actual canopy volume (TRV method). In the case of annual crops the dose may
be calculated as a function of Leaf Area Index.
31.1 Introduction
In this chapter we discuss the types of chemicals normally applied to crops and the
environmental factors to consider for their application. The main types of pesticides
used in agriculture are herbicides, insecticides and fungicides for controlling weeds,
insects and pathogen fungi, respectively. The term pesticide also includes other
products which are not exactly control agents like defoliants, desiccants and plant
growth regulators (Table 31.1).
Pesticide use is almost as old as agriculture with elemental sulfur being the first
known pesticide as it was employed by Sumerians to control insects and mites
about 4500 years ago.


Agriculture, DOI 10.1007/978-3-319-46116-8_31
472 F.J. Villalobos and E. Fereres
Table 31.1 Classification of Pesticide type Target

pesticides according to the
Algicide Algae
target organism
Avicide Birds
Bactericide Bacteria
Fungicide Fungi
Herbicide Weeds
Insecticide Insects
Miticide Mites
Molluscicide Snails, slugs
Nematicide Nematodes
Piscicide Fish
Rodenticide Rodents
Plant growth regulators
Dessicants
Defoliants
31.2 Pesticides
In general a small amount of pesticide active ingredient (a.i.) has to be applied

uniformly over a large area. To improve distribution and application uniformity the
active ingredient is usually made up into a formulation, which combines the active
ingredient with various inert carriers, and other ingredients to improve shelf-life,
enhance dispersal in water and prevent clumping (solids). The type of formulation
can affect the toxicity, the persistence and the rate of release.
The concentration of active ingredient in a formulation may be expressed as %
weight (% w/w) or g/100 g for solids. For liquids there are several alternatives: %
weight (% w/w), g/100 g, % volume (% w/v), g/100 cm3 or g/L.
Pesticides can be presented as liquids (solutions or suspensions) or solid (powder
or granules). They may be applied with spray bars or dispensers, either over the
entire surface, or localized over only a fraction.
It is common to use mixtures of pesticides to expand the range of action (control
a larger number of species), save on applications and in some cases produce
synergism (the action of a pesticide enhances the effect of another). Compatibility
is the possibility of mixing two pesticides without reducing their efficiency.
According to their mobility in the plant, pesticides may be called systemic when
they are absorbed and translocated inside the plant or contact when they affect only
the plant part in contact with the chemical.
We may also classify pesticides according to their persistence (residual or
non-residual) or their selectivity (selective versus non-selective) depending on the
number of species affected.
31 Application of Herbicides and Other Biotic Control Agents 473
31.3 Application of Pesticides
Most pesticides are applied by spraying liquid formulations but other methods may
also be used (Table 31.2), like baits, fumigants, dusts or addition to irrigation water.
Spray methods may be classified according to the volume application rate (VAR,
L/ha) (Table 31.3).
Spray application is a relatively complex process that can be divided into a
transport phase and another of interaction with the surface to be treated. A set of
droplets, characterized by a diameter distribution and an initial velocity, is released
from the nozzle at a given height. These droplets suffer a vertical force, resultant of
the forces of gravity and friction, and a horizontal force (wind drift) that determine
the trajectory and hence the drop point. During transport, droplets are subjected to
direct evaporation so the diameter decreases. The fall speed decreases as does the
diameter of the droplets. Very small droplets may remain practically suspended in
the air and fall very slowly, which increases horizontal displacement and thus, drift.
The interaction phase is that occurring when the drops reach the surface. If the
droplet is too large it tends to bounce or drain down to the ground. Small droplets
will likely stick to the crop surfaces.
Therefore, optimal droplet diameter is generally in the 150–250 μm range
depending on the type of product (Table 31.4). The density of impacts required
also depends on the type of product. For example, non-systemic fungicides and
preventive contact insecticides for very mobile pests require full coverage of plant
organs and therefore a very high density of impacts. In the case of insecticides that
are toxic by ingestion with highly mobile insects the required density will be much
lower.
31.4 Drift
Drift is a side effect of pesticide use associated with ground and aerial application
and is an important environmental concern. Drift is the uncontrolled airborne
movement of spray droplets, vapors, or dust particles, away from the intended
point of application, therefore reducing the actual dose applied. Drift can cause
injury to non-target plants and animals, and has the potential for contaminating
non-target sites, in special, surface waters, sensitive crops, warehouses, populated
areas and flowering crops with bees present.
Any pesticide application may produce some amount of drift. The actual amount
will depend on the formulation of the material applied, the application method, the
volume used and the weather conditions during application.
Pesticide drift is usually greater when application height is large (e.g. by aircraft)
so it is recommended not to exceed 1.2 m (above the crop) for ground applications.
Drift is also more important when particles are light (e.g. dusts, low volatility oils)
or drops are small. For sprays, droplet size increases with nozzle opening and
Table 31.2 Classification of pesticides according to presentation

Applied
Name Description Advantages Disadvantages Typical use as
Solids
Bait Mixture of Easy to Risk for pets Insects, Solid
a.i. and food spottreat and and wildlife rodents, birds,
that attracts apply by hand. slugs
pests. Made as Ready to use
meal, pellets or
liquid
Dust Finely ground Ready to use. Visible on Spot or seed Solid
inert particles No mixing plants. Easily treatment
(e.g. talc) with inhaled. Risk
1–10 % a.i. of drift
Granules and Dry inert mate- Ready to use. May be eaten Soil treatment Solid
pellets rials (e.g. corn No mixing. by birds. May for insect or
cob) with Minimal drift. need weed control.
2–25 % a.i. incorporation. Baits
Impregnated Granular fertil- One step Special equip- Soil Solid
fertilizer izer containing application. ment required application
a low concen-
tration of
pesticide
Soluble Dry powder or Agitation not Hazardous if Mostly sprays Liquid
powder granules which needed after inhaled for insect and
dissolve in mixing. weed control.
water to make Few formula-
spray solution. tions
Often > 50 % a. available
i.
Wettable Finely ground Less hazard Hazardous if Sprays for Liquid
powder inert ingredi- than emulsifi- inhaled. Dusty insect, dis-
ents with able ease, and
>50 % concentrate weed control
a.i. forms a
suspension in
water
Dry flowable, Mixture of Less dusty and Spray mix Sprays for Liquid
water dis- a.i. and inert lower inhala- requires insect, dis-
persible material tion hazard agitation ease, and
granules forming pellets than powder weed control
or granules.
Forms suspen-
sion in water
Liquids
Aerosol (A) Usually contain Low concen- Inhalation Inside Gas
small amounts tration of a.i. hazard. greenhouses
of a.i. and a
petroleum
solvent
(continued)

Applied
Emulsifiable Contains a.i., High concen- Amount of Sprays for Liquid
concentrate petroleum sol- tration of a.i. increases insect, dis-
vent, and emul- a.i. Easily mixing hazard ease, and
sifiers. Pesti- mixed weed control
cide is
suspended in
spray which is
milky coloured
Flowable Finely ground No dust Spray mix Sprays for all Liquid
particles needs constant types of
suspended in an agitation pesticides
inert liquid car-
rier. Forms sus-
pension in
spray mix
Gel Semi liquid Used with Herbicides Liquid
emulsifiable water soluble and
concentrate packaging insecticides
Micro-encap- Consist of pes- Reduced haz- Agitation Insecticide Liquid
sulated ticide ard to applica- needed. High and phero-
materials surrounded by a tor. Easy to risk for bees mone sprays
plastic coating. mix and apply
Mixed with
water and
sprayed
Solution A.i. comes Easily mixed High concen- Spray of Liquid
dissolved in tration of herbicides
liquid. Forms a a.i. increases
solution in mixing hazard
spray mix
Ultra low Liquid with Requires little Needs special Insecticide Liquid
volume con- very high con- or no mixing. application sprays inside
centrate or centration of Few formula- equipment greenhouses
sprayable a.i. Used as it is tions available
concentrate or slightly
dilluted
Gases
Fumigants Volatile liquids Wide spectrum Highly toxic. Greenhouses, Gas
or solids pack- of pests and Treated area granaries.
aged to release stages of must be sealed Pre-plant soil
a toxic gas development. treatment for
Good penetra- resistant pests
tion of struc-
tures and soils
(continued)

Applied
Package
Water-solu- Pre weighed Safe and easy Sprays for Liquid
ble packets amount of wet- to use insect, disease
table or soluble and weed
powder in plas- control.
tic bag that dis-
solves in water
Table 31.3 Classification of Trees and shrubs Annual crops

spray methods according to
Dose (L/ha)
the volume of spray applied
High volume >1000 >700
Medium volume 500–1000 200–700
Low volume 200–500 50–200
Very low volume 50–200 5–50
Ultra low volume <50 <5
Table 31.4 Desired parameters for spraying the main types of pesticides. The volume of spray for
LAI ¼ 1 has been calculated for nearly optimal conditions (10 % loss by drift and 10 % not
intercepted by the canopy)
Required droplet Impact Spray volume for Main
diameter density LAI ¼ 1 concern
Pesticide Type μm cm-2 L/ha
Herbicide Contact 300 60 106 Drift
Herbicide Systemic 700 20 449 Drift
Fungicides Contact 150 60 27 Areas
wetted
Fungicides Systemic 250 20 20
Insecticides Contact 200 60 63 Impact
number
Insecticides Systemic 350 20 56
Fertilizers >1500
decreases with pressure. Droplets with diameter lower than 100 μm favor drift so
the 150–200 μm range is recommended. Thickeners reduce the frequency of small
droplets.
Fumigants and highly volatile formulations may produce vapors which easily
drift. Vaporization (volatilization) is proportional to evaporative demand and
inversely proportional to drop size. Water-based sprays will volatilize more quickly
than oil-based sprays. However, oil-based sprays can drift farther because they are
lighter, especially at high temperatures.
To minimize drift it is better to work under low evaporative demand (temper-

atures below 32–35 ºC), and low wind (1–2 m/s) with consistent direction and
close to neutral conditions. These conditions usually occur in the early morning
and late afternoon. Unstable conditions favor the rise of droplets and therefore the
horizontal displacement. Under stable conditions (temperature inversion) a highly
concentrated cloud is formed which can move outside the target area. This is
highly risky for herbicides but might be more acceptable for insecticides or
fungicides. Spraying of pesticides, in special herbicides, should be avoided for
calm (wind speed below 1 m/s) and windy conditions (above 2.5–3 m/s). Note that
atmospheric instability adds thermal turbulence to mechanical turbulence due to
wind so for any given wind speed the conditions for spraying get worse as
instability increases.
Pesticide applications should also be avoided when rain is expected in the short
term. Wind direction should be taken into account to analyze the risk of drift
arriving at sensitive areas. Untreated buffer zones should be established if needed.
31.5 Persistence of Pesticides
Persistence of pesticides is important as it determines the duration of the protective

effect and the safety period in terms of harvesting the crop after the application. The
persistence of a given herbicide also may preclude planting a sensitive crop species
in the same field.
The degradation of pesticides is caused by:
– Microbial decomposition so environmental conditions that favor microbial
activity contribute to the degradation of the pesticide.
– Chemical decomposition by oxidation, reduction or hydrolysis.
– Photolysis, i.e. chemical degradation due to light absorption by the product.
Pesticides can also be immobilized in the soil (adsorption by soil colloids). Most
adsorption occurs with organic colloids, so soils high in organic matter or clay soils
will require larger doses of herbicide.
Pesticides can be lost by leaching or volatilization. Leaching depends on its
solubility and its adsorption to colloids. To avoid volatilization the more volatile
products should be incorporated into the soil. Residual herbicides (e.g. atrazine) can
cause serious environmental problems as they accumulate in groundwater after
leaching which has led to their banning in some countries (e.g. Atrazine was banned
in the European Union in 2004).
31.6 Herbicides
31.6.1 General Characteristics
Herbicides are phytotoxic products. They may be classified as total or selective,

when they affect all or some species. In terms of time of application we may
distinguish among preplant, preemergent and postemergent (in reference to the
crop) and in terms of mobility we have systemic and contact herbicides.
The dose of herbicide to be applied depends on its phytotoxicity to the weeds and
the crop, and the density of weeds present.
Herbicides may be absorbed by the roots, the leaves or the stems. Leaf absorp-
tion is increased by adding wetting agents. The absorption of polar herbicides is
more effective by the roots than by the leaves. Herbicides interfere with some
metabolic process of the plant (photosynthesis, respiration and metabolism of
nucleic acids and proteins). Visual symptoms of herbicide damage include reduced
growth, malformation of leaves and plant wilting.
The translocation of systemic herbicides can follow two paths. If it is absorbed
by the roots, the herbicide passes to the xylem, and then it is distributed by the sap
flow. If it is absorbed by the leaves it moves via symplast to the phloem, and then it
is transported to the growing tissues.
31.6.2 Selectivity of Herbicides
The selectivity of an herbicide implies its ability to control weeds without causing
significant damage to the crop. The selectivity depends on several factors:
(a) Plant: Sensitivity is higher in young and fast growing plants. The highest
sensitivity in seedlings is due to their thinner cuticle. The morphology of the
plant may be responsible for the selectivity. It may be due for example to
differences between the crop and the weed in the root system or the location of
sensitive organs such as the apical meristem.
(b) Climate: The selectivity of foliar absorbed herbicides is reduced at high tem-
peratures. However in these conditions the rate of metabolism of the herbicide
in the plant is higher which helps reduce its negative effects. In conditions of
high humidity and/or high soil water content the leaf cuticles have a higher
degree of hydration which facilitates herbicide leaf absorption and can increase
the potential damage to the crop.
(c) Soil: Texture and organic content determine the degree of fixation of the
herbicide to soil colloids and hence the concentration of the active ingredient
in the root zone.
The high sensitivity of seedlings is the basis of the technique of split dosing
which involves making a number of small dose post-emergence applications when
weeds are emerging so that the crop hardly suffers their effects while weed
seedlings are killed. The main groups of herbicides and their characteristics are
presented in Table 31.5.
31.7 Insecticides
Insecticides are agents that control insects by killing them or preventing them from
destructive behaviors. Insecticides may be natural or artificial and are applied in a
wide range of formulations and delivery systems (sprays, baits, slow-release diffu-
sion, etc.). Some G.M.O. genotypes incorporate bacterial genes coding for synthesis
of insecticidal proteins.
Table 31.6 shows a classification of insecticides according to several criteria.
By 1940 only some inorganic (e.g. sulfur) or botanical (e.g. pyrethrum) insec-
ticides were available. Then the first synthetic organic insecticide, DDT, an organ-
ochlorine, appeared. After that the history of insecticides has been a mixture of
success (high levels of control at reduced cost, development of more specific
products with low persistence) and failure (e.g. organochlorines are toxic for
animals, accumulate in the trophic chain and had to be banned).
31.8 Fungicides and Control of Diseases
Plant diseases are best controlled by integrating a number of practices including

sowing date, planting density, crop rotation, cultivar selection (using disease-
tolerant or disease-resistant genotypes), fertilizer and irrigation programs, micro-
climate modification and application of fungicides.
Knowledge of the disease cycle of the pathogen is important when developing
disease forecasting systems or economic thresholds. Forecasting systems are based
on temperature and relative humidity or leaf wetness. Thresholds-based programs
involve periodical monitoring of symptoms (e.g. number of disease spots per leaf).
Important aspects of the disease cycle include the number of generations per year of
the pathogen and the time between infection and the appearance of symptoms
(latent or incubation period) which may be a few hours for aerial diseases up to
several weeks for soil borne pathogens.
Fungicides are products that kill or prevent the growth of fungi and their spores.
They may be classified according to different criteria.
(a) Mobility in the plant.
Contact fungicide is that that remains on the surface where it is applied but does
not enter the plant and has no after-infection activity. Repeated applications are
needed to protect new growth and to replace losses by rain or irrigation, or
degraded by environmental factors.
480
Table 31.5 Classification of herbicides according to their main characteristics

1 2 3 4 5 6 7 8
Photosystem II inhibitors Triazines (eg Atrazine), Ureas (eg Diuron) S, Temp., placement, Lf Ap M- L
Lf metabolism H
Photosystem I inhibitors Benzotidiazoles. (Ex. Bentazon) Lf Metabolism Lf No L L
Membrane synthesis inhibitors Bipyridylium (Ex. Diquat, Paraquat) Lf Temp., placement, Green No Nil H
metabolism tissue
Membrane disrupters Diphenylether, Dinitrophenols Lf Pubescence Green No L M
tissue
Aminoacid synthesis inhibitors Sulfonylureas Lf Metabolism Enzimes Ap, M- L
(selective) Sy H
Aminoacid synthesis inhibitors (not Aminoacid derivatives (Ex. Gliphosate) Lf Placement, genotype Enzimes Sy L L
selective)
Lipid synthesis inhibitors Aryloxyphenoxypropionate and cyclohexane Lf Not affecting dicots Meristem Sy Nil M
derivatives
Growth regulators Phenoxy-carboxylic acids (eg 2,4-D), benzoic S, Not affecting monocots Auxines Ap, V M
acid, piridines Lf Sy
Cell division inhibitors Dinitroanilines (eg Trifluraline) S Placement Contact No V M
tissue
Protein synthesis inhibitors Chloroacetamide and thiocarbamates S Placement, metabolism Seedlings Little L- M
H
Pigment synthesis inhibitors Piridinones S Placement, metabolism Chloroplast Ap M- L
H
F.J. Villalobos and E. Fereres
31
1: Herbicide type
2: Chemical groups
3: Activity S: soil, Lf: leaf
4: Basis for specificity
5: Site of action Lf: leaf
6: Translocation Ap: Apoplast, Sy: symplast
7: Persistence L: low, M: medium, H:high, V: variable
8: Toxicity L: low, M: medium, H:high
Application of Herbicides and Other Biotic Control Agents
481
Table 31.6 Classification of insecticides according to their main characteristics

Toxicity for other
Type Main effect on insect organisms Main targets
Organochlorine Nervous system High Broad spectrum
Organophosphate Neuromuscular system Very high Broad spectrum
Organosulfur Ovicidal Very low Mites
Carbamates Nervous system High, in special to Broad spectrum
fish
Formamidines Nervous system Medium Organophosphate and
carbamate-resistant pests
Dinitrophenols Synthesis of ATP Slight to moderate Withdrawn from use
Organotins Synthesis of ATP High for aquatic Mites in trees
life
Pyrethroids Nervous system High for fish Most agricultural insects
Nicotinoids Nervous system Low Sucking insects, soil
insects, whiteflies
Spinosyns Nervous system Low Caterpillars, lepidopteran
larvae, leaf miners, thrips
Pyrazoles Synthesis of ATP Low Psylla, aphids, whitefly
and thrips
Pyridazinones Mitochondrial electron Medium for Mites
transport aquatic life
Quinazolines Blocking the synthesis of Medium Broad spectrum
chitin at larval stage
Botanicals Pyrethrum – nervous Low Lice
systems
Nicotine – nervous Low Aphids and caterpillars
systems
Rotenone – respiratory High to fish Fish
enzyme inhibitor
Limonene – nervous Low Fleas, lice, mites, and
systems ticks
Neem - reduces feeding Low Moth and butterfly larvae
and disrupts molting
Antibiotics Blocking the neurotrans- Toxic to fish and Spider mites, leafminers
mitter GABA bees
Fumigants Act as narcotics Depends on Depends on compound
compound
Inorganics Dependent upon type of Depends on Depends on compound
inorganic compound (e.g. sulfur for mites)
Biorational Act as attractants, growth Very low Very specific
regulators or endotoxins
Benzoylureas Insect growth regulators Some may affect Caterpillars and beetle
other larvae
invertebrates
Systemic fungicides are absorbed into the plant and move within it. They
may offer some after-infection activity. Very few fungicides are truly systemic
(the group of the phosphonates) but some are acropetal penetrant (moving
upwards in the xylem), and some are localized penetrant (i.e. redistribute within
the treated leaf).
(b) Role in protection (some fungicides can fall into more than one of the following
categories)
– Preventive fungicides offer a protective barrier that prevents infection.
– Early-infection activity: the product enters the plant and stops the pathogen,
being usually effective until several days after infection. This type has also
preventive activity and is most effective when applied before infection.
– Eradication: a few products have the ability to stop disease development after
symptoms have developed. Even then the damage caused by the disease on
the plants often does not disappear.
– Anti-sporulant activity (ability to prevent spore formation). In this case,
disease continues to develop, but spores are not produced, so the amount of
inoculum available to infect surrounding plants is reduced.
(c) Mode of action.
The mode of action is how a fungicide acts on a target fungus, i.e. which is the
specific process in the metabolism that is affected (e.g. damaging cell mem-
branes, inactivating critical enzymes). A single-site fungicide affects only one
point in one metabolic pathway or a single critical enzyme or protein. These
fungicides are less phytotoxic and tend to have systemic properties, but show a
higher risk of pathogens developing resistance, as the mode of action is so
specific that small genetic changes in fungi can overcome the effect of the
fungicide.
On the other hand, a multi-site fungicide affects a number of different
metabolic sites within the fungus.
31.9 Calculation of the Volume Application Rate

of Pesticides
The minimum volume of pesticide to apply or Volume Application Rate (VAR, L

ha1) may be calculated as a function of the desired impact density (N, impacts
cm2) and the droplet diameter (d, μm) as:
107 π d3 N
VAR ¼ f ðLAIÞ ð31:1Þ
6 1 ps p d
where ps and pd are the probability of droplets not being intercepted by the canopy
and the fraction of droplets lost as drift, respectively. The f (LAI) function depends
on the type of pesticide. For contact fungicides and contact insecticides f (LAI) ¼ 2
LAI, while for systemic fungicides and insecticides f (LAI) ¼ LAI. For herbicides f
(LAI) ¼ 1. The same value should be adopted if the calculated value of f (LAI) is
lower than 1.
Example 31.1 We want to apply a contact fungicide on a wheat crop with

full ground cover and LAI ¼ 3. The required impact density is 60 impacts cm
2
and our sprayer generates droplets of diameter 150 μm. We will perform
the application with good conditions (drift loss 10 %). With full ground cover
we can assume that almost all droplets will be intercepted by the canopy
(ps ¼ 0.1). With a contact fungicide we take f (LAI) ¼ 2 LAI ¼ 6. Therefore:
107 π d3 N 107 π 1503 60

VAR ¼ f ðLAIÞ ¼6 ¼ 80 L ha1
6 1 ps pd 6 1 0:1 0:1
31.10 Calculation of Spray Volumes for Fruit Tree

Orchards
The Tree-Row-Volume (TRV) system was developed for hedgerow apple orchards
in the United States. Instead of using a standard volume application rate (3741 L/
ha ¼ 400 gal/acre), the VAR was corrected in proportion to tree volume per unit
area (TRV, m3 ha1):

VAR L ha1 ¼ 0:0937 TRV ð31:2Þ
where 0.0937 is the volume (in liters) of spray necessary to wet 1 m3 of crown
(of low leaf area density). This value increases to 0.1337 L m3 for dense trees. This
same approach may be applied also to orchards where tree crowns do not overlap.
However, the specific factor of volume per unit volume may change for high
density orchards or differences in leaf anatomy (e.g. for grapevines, 0.3 L m3).
If the recommended dose of pesticide is δp (kg active ingredient ha1) for a
standard VAR of 3741 L/ha, then the actual dose of pesticide should be:
0:0937 TRV δp
Actual dose kg a:i: ha1 ¼ ð31:3Þ
3741
Example 31.2 An apple orchard has trees spaced 7 5 m. We want to apply

a fungicide with recommended dose of 2 kg a.i./ha. The horizontal radius of
(continued)

the trees is 2.0 m and the vertical radius is 1.5 m. The trees have a low leaf
area density. The tree volume per unit area is:
4
π 2 2 1:5 4 2 1
TRV ¼ 3 10 m ha ¼ 7181 m3 ha1
75
For low leaf area density the volume application rate is:
VAR ¼ 0:0937 TRV ¼ 0:0937 7181 ¼ 673 L ha1
And the actual dose:
0:0937 TRV δp 0:0937 7181 2

Actual dose ¼ ¼ ¼ 0:36 kg a:i: ha1
3741 3741
Bibliography
Fernandez-Cornejo J, Nehring R, Osteen C, Wechsler SJ, Martin A, Vialou A (2014) Pesticide use
in U.S. agriculture: 21 selected crops, 1960–2008. Econ Info Bull EIB-124, ERS-USDA.
Furness, G. O., Magarey, P. A., Miller, P. H., & Drew, H. J. (1998). Fruit tree and vine sprayer
calibration based on canopy size and length of row: unit canopy row method. Crop Protection,
17(8), 639–644.
Matthews, G. A. (1999). Application of pesticides to crops. London: Imperial College Press.
Pimentel, D. (2005). Environmental and economic costs of the application of pesticides primarily
in the United States. Environment, Development and Sustainability, 7, 229–252.
Chapter 32
Harvest and Conservation
Abstract Harvesting is the key operation in farming that culminates the season’s
efforts. It represented an important fraction of all labor used in agriculture until
recently and the equivalent in production costs. With the advent of mechanical
harvest, costs have decreased dramatically contributing greatly to the reduction of
food prices in recent decades. Determining the time of harvest is normally a com-
promise between factors that increase profits (e.g., delaying it for greater biomass)
and increased risks (e.g., lower product quality or persistent bad weather). There are
yield losses during harvest that must be prevented and also during postharvest, in this
case related to storage conditions. Drying grains and storing them under low relative
humidity (RH) minimizes the incidence of fungal diseases that deteriorate the
product. The water content of the seeds after harvest tends to an equilibrium value
in storage which depends on the RH and air temperature, such seed water content
may be determined through Moisture Release Isotherms. Forage crops may be
consumed directly by animals or cut and conserved as silage, haylage or hay, with
each process requiring more drying of plant materials before safe storage. Harvest of
fruits for the fresh market is still done by hand and quality considerations are
important in determining its timing. The harvest of fruits and vegetables for
processing has been extensively mechanized; determining its timing is based on a
tradeoff between quality factors and yield and marketing objectives.
32.1 Introduction
Harvest is a key activity as it culminates all farming operations leading to the

ultimate goal of agriculture, producing food and obtaining useful materials. The
first agricultural practice in history was harvest. Long before actual crop husbandry
began humans were collecting seeds from some grass species as part of a diverse


Agriculture, DOI 10.1007/978-3-319-46116-8_32
diet. These hunter-gatherers processed the seeds to produce ale and bread. A great
advantage of seeds was the possibility of long term storage for later use in contrast
with the very limited time that meat from hunting lasted in storage. Furthermore the
low efficiency of collecting seeds from natural populations where useful plants are
scattered probably fostered the idea of agricultural crops. Someone must have
wondered: what if this field would be filled only with the plant that I want every
year? What if all the plants ripened at the same time, with larger seeds that would
not shed before I collected them? Thus, harvesting the edible parts of plants in
natural environments probably was pivotal to the invention of agriculture.
The great advances in crop productivity of recent decades have often been
associated with a reduction in harvesting costs through mechanization. In many
cases harvest costs limit the economic viability of crops. Traditionally harvesting
was based only on human labor with the partial support of beasts of burden for
gathering and transport. The availability of labor for harvest often limited the area
of arable land. For instance hand harvesting 1 ha of wheat required about 130 -
man-hours of work (13 days of 10-h days) while current mechanized harvesting
requires only 0.4 man-hours. Therefore if the time available to harvest is between
30 and 60 days it follows that one person could harvest 2.3–4.6 ha manually but as
many as 750–1500 ha using a standard combine harvester.
One important decision for the farmer is when to harvest. Physiological maturity
is defined as the time when biomass of the harvested part reaches a maximum.
However, the best conditions for harvesting (harvest maturity) may occur much
later because, in most cases, reducing further the water content of the product is
beneficial in terms of handling less weight and to reduce the risk of fungal diseases..
In the case of fruits the level of ripeness involving color, accumulation of secondary
products, and taste has also to be considered. In the case of many vegetables the
date of harvest may not be associated to physiological maturity when special
quality characteristics are the main concern.
The harvest method and associated operations also affect other components of
the farming system. Normally the harvested product represents only part of the
biomass and the rest must be managed in such a way that it contributes to the
subsequent crop and does not interfere with subsequent farming operations. Actu-
ally, the management of crop residues has a major impact on the organic matter and
nutrient balances of the soil and offers great opportunities for contributing to soil
conservation. Harvesting can also have adverse effects on the agricultural system
like the spread of pathogens, insects and weeds and soil compaction due to the
traffic of the heavy machinery used.
32.2 Harvest Operations
Harvest can affect the whole plant or only part of it. In this case the harvested organ
may be aerial or underground (Table 32.1). The crop residues may be also harvested
and exported outside the field, either separately, or still attached to useful parts, for
32 Harvest and Conservation 489
Table 32.1 Classification of products harvested in agriculture according to the final use or the
harvested organ
Use Harvested organ and products
Food Fresh Shoot Vegetative
Transformation Reproductive (flowers,
fruits, seeds)
Fodder Sap (e.g. maple)
Industrial products Textile Latex (e.g. rubber,
opium)
Chemical (oil, paint, Resins (e.g. pine)
varnishes)
Fuels Subterranean Roots (e.g. beet, carrot,
organs cassava)
Perfumery and Bulbs (e.g. onion)
cosmetics
Pharmaceutical Tubercules (e.g. potato)
Rizhomes (e.g. ginger)
Seeds and propagules for
agriculture
Table 32.2 Some basic operations in the harvest of different crops

Harvesting operations Crops
Mowing Forage crops, medicinal crops
Reaping, threshing, and winnowing Cereals, pulses, oil crops
Combing and pulling Fruits
Digging, sieving and loading Tubercules, roots
Cutting Vegetables
Gathering Fruits
Shaking, sweeping, loading Nuts, olive
further separation. In the best case only the useful organs are exported while crop
residues remain in the field.
Decisions about when and how to harvest depend on the final use of the product
(Table 32.1). For example, vegetable and fruit products for fresh consumption are
frequently gathered by hand and harvest may be performed in several passes as the
crop ripens.
The operations involved in crop harvesting are varied and depend on the species
and its use (Table 32.2). In most cases the harvest consists of cutting part or the total
above ground part of the plant and separating the useful fraction (e.g. seeds) from
the residues. These two processes may occur at about the same time (e.g. combine
harvest of grains) or the cut plants may be left in the field for drying and then either
combined or transported out of the field for threshing and winnowing. Before
harvest it may be necessary to prepare the crop (using defoliants or abscission
promoters) and/or the soil (compaction, irrigation, etc.).
After harvesting some postharvest operations may be performed in the field/
farm or outside, like drying, cleaning, removal of plant parts, sorting by size, etc.
Finally the harvested product may go directly to the packing plant or to the
storage facility, which may require aeration and/or temperature control. For horti-
cultural crops a controlled atmosphere is usually employed, that is, with low oxygen
and high carbon dioxide concentration that greatly slow down fruit ripening.
32.3 Yield Losses During Harvest
Not all crop biomass but only part of it (yield, Y) is useful. The ratio yield/biomass
was defined as the Harvest Index in Chap. 13. However, the harvested yield (Yc) is
always less than Y (measured at physiological maturity) because of losses:
(a) Before harvest, due to:
– Respiration of the harvestable organ, which is proportional to the time the organ
is in the field after reaching maturity and to its water content and temperature.
– Loss of harvestable structures (dehiscence of pods, abscission of fruits,
consumption by herbivores, fire, hail, etc.).
(b) During harvest:
– Inaccessible parts of the plant: e.g. plant parts below the cutter bar.
– Not captured by the harvest system or dropped by it.
– Deterioration of structures collected (broken grains, damaged fruits)
– Rejection due to low quality: Although not exactly a loss, part of yield may be
left uncollected because of poor quality or excessive harvest costs. For instance a
very low grain yield may not compensate for the cost of mechanical harvesting.
(c) After harvest (during packaging, transport or storage)
– Respiration of the harvested organ: dependent on water content and
temperature
– Consumption or deterioration caused by pests and diseases: dependent on the
sanitary conditions of the storage facility and on the water content and
temperature of the product and air humidity.
– Rejection due to postharvest quality criteria: The collected product (or a
fraction of it) may not meet quality standards implying a loss or an additional
cost for sorting and separation of those products not meeting the standard.
Box 32.1 The Year Without a Summer

Weather conditions in 1816 turned extremely cold as a result of the volcanic
eruption of Mount Tambora in Indonesia that ejected an enormous amount of
ash into the atmosphere, thereby reducing transmissivity and global radiation.
(continued)

It is estimated that global average temperature decreased after the eruption and
for almost 3 years between 1.5 and 3ºC. Heavy rainfall and cold temperatures
during the summer in Western Europe resulted in the delayed development of
many crops that did not reach maturity. Those that did could not be harvested in
many cases, and when harvested, the grain was lost later during storage. At that
time, subsistence agriculture was common and people in the cities lived from
harvest to harvest depending on fragile agricultural systems that did not
produce sufficient food. Food shortages were dramatic during 1816 and the
two following years, not only in Europe but globally. In Asia, the monsoon
patterns were disrupted and that caused not only famine but epidemics that
killed millions, and political unrest was at a high point in China.
32.4 Agricultural Operations Affected by Harvest
Harvesting normally requires the concentration of efforts by the farmer and its labor
force thus restricting other simultaneous farm operations. The date and method of
harvest of a given crop determines the possible choices for the next crop in the
rotation. The harvest method determines the amount and distribution of residues in
the field, which determines the need for additional operations (burning, chopping,
removal) and thus, the time required for land preparation before sowing the next crop.
Irrigation may be used to improve soil conditions before harvest (e.g. before
digging of tubers or roots). Irrigations may also be stopped to promote abscission
(of leaves or fruits) and to prevent soil compaction due to traffic on wet soil during
harvest.
Some tillage operations may be required before harvesting (e.g. surface soil
compaction to facilitate sweeping of fallen fruits after tree shaking of almond trees).
Pesticide treatments must be stopped for some time before harvest to comply with
the safety periods. In other cases (e.g. cotton) defoliants are applied to facilitate
mechanical harvesting.
32.5 Harvest of Grains and Seeds
While there is some leeway in harvest time, in winter cereals delaying harvest while
leads to lower grain water content and therefore, reduced drying requirements, it
has negative effects by increasing:
– dry matter losses and/or reduced grain quality
– seed losses by dehiscence.
– likelihood of deterioration by adverse conditions (lodging, hail) or destruction
by fire
Table 32.3 Status of seeds of major agricultural crops as a function of water content
Water Water
content g potencial Activity of Degradation under
water/g Mpa Seed status biotic factors storage
>0.41 >1.5 Physiological
maturity
0.30–0.40 5 to 1.5
0.20–0.30 11 to 5 High respiration rate High (bacte- Fast
ria, fungi)
0.13–0.20 100 to 11 High mechanical High Fast
resistance (insects,
(0.13–0.16) fungi)
Fit for combine
harvesting
0.10–0.13 120 to High Slow
100 (insects)
<0.10 < 120 Low Slow (may increase
with high temperature)
– production of mycotoxins, toxic chemicals produced by naturally occurring

moulds and some plant pathogens like Fusarium. The most common type is
aflatoxins, generated by moulds of Aspergilus flavus and A. parasiticus. The
probability of contamination with aflatoxins is higher when harvest is delayed
during wet weather or when grain is harvested with high moisture and drying is
delayed. Water stressed crops are more susceptible to infection. Aflatoxins
contamination not only occurs in cereal grains, but also in other agricultural
products (sunflower seeds, nuts, cassava, cottonseed, spices, pepper, hay). Maize
harvest can be delayed longer than winter cereals but only if weather conditions
are dry, thereby reducing drying costs without risking aflatoxin contamination.
In many cases harvest is performed before reaching the minimum water content
required for safe storage and then the seed water content must be reduced after-
wards out of the field. In species where pod shedding is an issue (e.g. rapeseed), or
when time is limited, harvest is performed in two stages. The first is swathing,
i.e. cutting the crop and leaving it forming windrows where it will dry for 5–10 days
until the desired water content in seeds is achieved, when the second stage to
separate the seed from the pods (combining) is performed. In general, grain can
be stored safely when it is clean, dry, healthy and intact. Conditions improve if it is
cold at the time of storage.
The water content of the grain affects (Table 32.3):
(a) The risk of physical damage during harvest
(b) The incidence of insect pests and fungal diseases
(c) The metabolic activity of the seed, which affects its rate of deterioration, the
release of oxygen by respiration and heat generation in the stock.
20
Wheat
SEED MOISTURE (% fresh basis) 18 Sunflower
16
14
12
10
4
0 10 20 30 40 50 60 70 80 90 100
RELATIVE HUMIDITY (%)
Fig. 32.1 Seed moisture isotherms for seeds of wheat and sunflower at 20 C
The water content of the seeds after harvest in storage tends to an equilibrium
value which depends on the relative humidity (RH) and air temperature. Strictly
equilibrium is reached when the water potential is equal in the air and the seed,
which can be calculated as
RT
Ψseed ¼ ln aw ð32:1Þ
Mw
where Ψseed has units of J/kg (or kPa), R ¼ 8.3143 J mol1 K1, T is air temperature
(K), Mw ¼ 0.018 kg mol1 and aw is the water activity, which is equivalent to the
equilibrium RH/100. Therefore, for a given air temperature and RH the grain will
reach an Equilibrium Moisture Content (EMC). This relationship is called Moisture
Isotherm (Fig. 32.1) and is applied not only to seeds but to any type of material. The
main factor determining EMC is RH, while temperature plays a minor effect. For
instance cereal grains at 25 C and 70 % RH show EMC between 13 % and 14 % (wet
basis). Using Eq. 32.1 this corresponds to a water potential of 49,095 kPa
(49.1 Mpa). One of the main factors affecting the relationship between EMC and
RH is the oil content: seeds rich in oil will show a lower EMC for a given RH. This is
clearly seen when comparing seeds of wheat and sunflower (Fig. 32.1). It is important
to point out that seed moisture may be expressed on a dry basis or wet basis:
mass water
EMCd ¼ 100 ð32:2Þ
mass dry seed
mass water
EMCw ¼ 100 ð32:3Þ
mass wet seed
Therefore:
EMCd
EMCw ¼ ð32:4Þ
100 þ EMCd
The seeds are classified as orthodox (can be dried to very low water content
without being damaged) as in most agricultural species and recalcitrant (desiccation
kills the seed) like those of Quercus spp., oil palm, chestnut and cacao. The seed
longevity in orthodox seeds decreases linearly with water potential from 350
(2–6 % water) to 14 MPa, which is equivalent to a range in aw from 0.1 to 0.9,
regardless of species. Seed longevity decreases also with temperature. Simple rules
have been proposed for seed storage. For instance, the James’ rule establishes that
the sum of temperature ( C) and relative humidity (%) should be lower than 60.
According to Harrington’s rule seed longevity decreases by one half for every 1 %
increase in seed moisture content or every 6 C increase in temperature.
To be on the safe side the RH in the space between grains has to be lower than
67–70 %, so that spore germination of pathogens is prevented. Therefore keeping a
low RH will ensure that the seeds do not lose viability or mass (by respiration) and
are not attacked by pathogens and insects.
The grains may require conditioning (drying, cleaning) before they can be stored
safely. Conditioning systems are divided into aeration systems, natural air drying or
low temperature (heating of 3–7 C) systems and high-temperature systems.
Postharvest losses of grains due to different causes are significant, particularly in
developing countries where much of the production is conserved at the household
level. Although figures as high as 40 % are frequently cited, the World Bank
estimated that in 2010, the value of postharvest losses amounted to 15 % of grain
production in Sub-Saharan Africa.
Table 32.4 presents values of seed water at physiological maturity and
recommended for harvest and storage.
32.6 Harvest of Forage Crops
In pastures and some forage crops harvest may be performed directly by grazing
animals. In this case it is important to adjust the density of animals to maximize
productivity. Animals may also graze field crops during specific growth stages or
after harvest to use the stubble. The former is the case of dual purpose crops
(cereals, rapeseed) that may be grazed during vegetative growth, as the meristems
are not affected, so crop growth can resume afterwards and lead to seed production
if the season length is sufficient.
In general the harvested forage crops are stored as silage (50–65 % water
content), haylage (30–50 %) or hay (15–25 %). The best nutritive quality in most
forage crops is achieved around flowering time. Delaying mowing after that time
implies greater (and drier) biomass but lower quality. The other factor that
Table 32.4 Water content of harvested organ at physiological maturity and at the time when it is
suitable for harvest and recommended values for harvesting and long term storage
recommended water
content for harvest
Physiological Suitable for No air Air Long term
Crop maturity harvest drying drying storage
Winter 35–40 <17–18 12 13.7–15.2
cereals
Maize 25–30 18–23 12.5–15.5
Rice 30–33 22–28
Rapeseed 40 <15 8 8
Sunflower 30–40 9–10 7.0–8.3
Lentil <13
Soybean 50 14–20 14 18 12
Sorghum 25–30 12–25 14 20–25 12
Bean 38–44 30–40
Pearl millet 30 <20 <13.5
determines mowing dates is the weather, as conditions after mowing must be dry
enough for successful drying.
For silage the crop is cut and chopped and taken to the silo where it is compacted
to exclude air and then sealed with plastic to ensure anaerobic conditions. These are
required for lactic bacteria to operate and reduce pH thus ensuring long term
conservation. Some crops (maize, sorghum) may be taken to the silo just after
cutting while others are left in the field drying for 1–2 days before silage starts. For
haylage the crop is left longer in the field for drying and then it is wrapped tightly in
plastic-covered bales.
Hay production requires a longer drying period. After mowing (and sometimes
conditioning) the harvested parts (crushing of plant material to speed up drying),
tedding (mixing and upturning) and windrowing (piling the plant material in rows)
are followed by baling. The time to dry has to be minimized to reduce respiration
losses. This time is proportional to forage biomass and inversely proportional to
swath area and evaporative demand.
32.7 Harvest of Underground Organs
In this category we include species that are harvested for their storage organs which
are located underground such as tubers (e.g. potato, yam), roots (sugarbeet, turnip,
cassava) and bulbs (onion, garlic).
Harvesting date of underground organs should be performed around physiolog-
ical maturity when little green area is left, although there are tradeoffs between
maturity and market targets; for instance, in some areas potatoes may be harvested
earlier than at maturity to fetch better prices, even though some yield is sacrificed.
Removal or killing of the shoot before harvest enhances periderm thickening of
tubers which reduces the risk of peeling or bruising during harvest. Some root crops
such as carrots are harvested when size is adequate for the market.
The harvest process of underground organs includes cutting, digging and lifting.
The product may be transferred directly to a trailer after separating soil and plant
residues or left in the field for drying.
As opposed to seeds, desiccation of harvested underground organs should be
prevented under storage. As water loss is proportional to Vapor Pressure Deficit, it
will be reduced by applying cool moist (almost saturated) air.
32.8 Harvest of Fruits and Vegetables
Fruits for the fresh market are usually collected by hand at the time when they reach
the desired size and are approaching ripening. Quality considerations are critical in
some crops such as wine grapes. Here, harvest is delayed until a certain level of
sugar content is reached in the grapes, as determined by periodic monitoring, and/or
the desired colors are achieved as required by the enologists. In olive oil production,
early harvest produces higher quality oil as demanded by markets (more fruity) at
the expense of lower oil content in the fruit and thus lower oil yields. An important
difference between species is the production of ethylene for ripening. Climacteric
fruits (e.g. pear, Appendix) can ripen off the plant once they have reached physi-
ological maturity, so they can be harvested at any time after reaching marketable
size and ripened later or they may be harvested when fully ripe. Non-climacteric
fruits (e.g. orange) have to be on the plant to complete ripening, so harvest must be
delayed until that time.
Fruits for processing (canning, dried, preserves, oil extraction, juice) may be
mechanically harvested (e.g. using shakers) when ripe.
Harvest of nut crops occurs after physiological maturity considering two factors:
the decreasing water content and the formation of an abscission zone to promote
fruit detachment at the time of harvest. Waiting for too long leads to significant fruit
drop, causing a fraction of the fruits to be on the ground with increased costs. Nut
harvesters are usually based on shaking the tree and collecting the fruits on inverted
umbrellas or lateral boards before they are transferred to a trailer.
Appendix: Composition of Harvested Products
Percentage of dry matter, gross energy per unit dry mass and main composition of
harvested products. Fresh fruits are also classified according to the capacity for
ripening after detaching from the plant at physiological maturity (climacteric, C) or
the lack of it (non climacteric, NC)
Part
harvested DM GE Protein Fat Ash
Cereals and MJ/ % Over
pseudocereals % kg dry mass
Barley (2 row) Hordeum Grain 86.5 18.4 14 3 3
vulgare
Barley (6 row) Hordeum Grain 88.5 18.4 12 2.5 3
vulgare
Maize Zea mays Grain 86 18.7 9.4 4.3 1.4
Millet-Foxtail Setaria italica Grain 90.5 18.8 11.9 4.9 3.6
Millet-Pearl Pennisetum Grain 89.5 18.8 12.4 4.9 2.7
glaucum
Millet-Proso Panicum Grain 90.5 19 14.2 5.5 3.7
miliaceum
Oats Avena sativa Grain 91 19.5 11 5.4 3
Rice (milled) Oryza sativa Grain 87.5 18 10.4 0.5 0.6
Rice Oryza sativa Grain 88 17.6 8.3 2.1 5.9
Rye Secale cereale Grain 87 18 10.3 1.4 2
Sorghum Sorghum Grain 88 18.8 10.8 3.4 2.1
bicolor
Triticale X Triticosecale Grain 89 18.1 11.7 1.5 2.1
rimpaui
Wheat- Spelt Triticum spelta Grain 89 19 12.2 3.9 2
Wheat (bread) Triticum Grain 87.5 18.2 12.6 1.7 1.8
aestivum
Wheat-durum Triticum durum Grain 87.5 18.5 16.5 2 2.1
Quinoa Chenopodium Seed 89 19.4 15.2 7.3 3
quinoa
Buckwheat Fagopyrum Seed 89.7 19 18.5 4.9 4.2
esculentum
Part
MJ/ % Over
Legumes % kg dry mass
Bean (dry harvest) Phaseolus spp. Seed 89 18.6 24.8 1.7 4.6
Chickpea (desi) Cicer arietinum Seed 89.5 19.6 22.1 5 3.3
Chickpea (kabuli) Cicer arietinum Seed 89.5 19.6 22.3 6.4 3.5
Cowpea Vigna Seed 90 18.7 25.2 1.6 4.1
unguiculata
Faba bean Vicia faba Seed 90 18.7 29 1.4 3.9
Lentil Lens culinaris Seed 89 18.5 26.9 1.6 3.8
Pea (dry harvest) Pisum sativum Seed 90 18.3 23.9 1.2 3.5
Peanut Arachis Pod 93 27.5 27 39 2.6
hypogaea
Soybean Glycine max Seed 87.5 23.6 39.6 21.3 5.8
(continued)
Part
MJ/ % Over
Forages % kg dry mass
Alfalfa (hay) Medicago Biomass 25 18.2 18.2 2.1 10.7
sativa
Clover (white, hay) Trifolium Biomass 25 17.4 22.7 2.2 12.3
repens
Maize (silage) Zea mays Biomass 30 18.9 8.1 2.6 4.8
Sorghum (silage) Sorghum Biomass 26 18.1 6.7 2.6 8.8
bicolor
Part
Sugar, oil and fiber MJ/ % Over
crops % kg dry mass
Cotton Gossypium Fiber 91 23.8 21.8 19.7 4.4
hirsutum +seed
Flax Linum Seed 93.5 27 22 34
ussitatisimum
Rapeseed Brassica spp. Seed 91 28.8 20.9 46 4.3
Safflower Carthamus Seed 92 26.1 15.6 32.2 2.4
tinctorius
Sugarcane Saccharum spp. Stalk 26 19 0.8 1.1 0.6
Sunflower (for oil) Helianthus Seed 91.5 28.7 20 44 4
annuus
Sunflower (for seed) Helianthus Seed 91.5 24 24 25 3
annuus
Tobacco Virginia Nicotiana Leaf 8 35 19
tabacum
Part
MJ/ % Over
Horticultural crops % kg dry mass
Artichoke Cynara Flowers 17 13.1 21.8 1 7.5
cardunculus
Asparagus (white) Asparagus Stems 7 12.5 32 1.8 8.5
officinalis
Been (green) Phaseolus Pods 8.7 11.7 24.1 4.6 8.0
vulgaris
Beet Beta vulgaris Root 12 14.5 13 1.4 9
Broccoli Brassica Flower 11.8 12.4 36.4 5.1 5.1
oleracea heads
Brussels sprout Brassica Leaf 13 12.8 24 2.1 9.8
oleracea
Cabbage Brassica Leaf 9.9 10.2 12.1 1.0 7.1
oleracea
(continued)
Carrot Daucus carota Root 11 13.3 4.5 3.6 5.5

Cauliflower Brassica Head 8.9 14.4 28.1 4.5 7.9
oleracea
Celery Apium Leaf 5 14.6 15 3.7 16.3
graveolens
Chicory Cichorium Leaf 8 12 21 3.8 16.3
intybus
Cucumber Cucumis Fruit – NC 3.5 17.1 28.6 17.1 11.4
sativus
Eggplant Solanum Fruit – NC 7 13.5 12.7 2.3 8.6
melongena
Endive Cichorium Leaf 6 11.5 20.2 3.2 22.7
endivia
Faba bean (green) Vicia faba Fruit 27 13.4 29 2.7 4.1
Leak Allium porrum Bulb 17 15 8.8 1.8 6.2
Lettuce Iceberg Lactuca sativa Leaf 5 13.2 20.5 3.2 8.2
Lettuce Roman Lactuca sativa Leaf 3.9 12.3 30.8 2.6 10.3
Melon Cucumis melo Fruit – C 12 14.7 5.3 1.4 4
Muskmelon Cucumis melo Fruit – C 10 14.4 8.6 1.9 6.6
Parsley Petroselinum Leaf 10 12.6 24.8 6.6 18.3
crispum
Pepper (green) Capsicum Fruits 7.2 13.8 11.1 2.8 8.3
annuum
Pepper (red) Capsicum Fruits 7.1 16.1 11.3 2.8 11.3
annuum
Pumpkin Cucurbita spp. Fruit – NC 9 13 11.9 1.2 9.5
Radish Raphanus Root 6 13.2 13.6 2 11
sativus
Spinach Spinacia Leaf 6.5 10.6 40.0 9.2 30.8
oleracea
Squash Cucurbita pepo Fruit – NC 14 13.8 7.4 0.7 5.9
Strawberry Fragaria x Fruit – NC 9 15.1 7.4 3.3 4.4
ananassa
Tomato Lycopersicon Fruit – C 5.4 11.3 9.3 1.9 7.4
esculentum
Watermelon Citrullus Fruit – NC 9 14.8 7.1 1.7 2.9
lanatus
Part
Fruit trees, vines MJ/ % Over
and shrubs % kg dry mass
Almonda Prunus Fruit 93 21.4 11.4 22.4 5
amygdalus
Apple Malus Fruit – C 13.8 15.6 4.3 3.6 1.4
sylvestris
(continued)
Apricot Prunus Fruit – C 14 14.8 10.3 2.9 5.5

armeniaca
Avocado Persea Fruit – C 27 25 7.5 55 5.9
americana
Banana Musa Fruit – C 26 17.1 5.5 1.3 4.5
paradisiaca
Cherimoya Annona Fruit – C 21 15.2 7.6 3.3 3.2
cherimola
Cherry Prunus Fruit – NC 19 14.9 6 1.1 2.7
avium
Coconut (copra) Cocos nucifera Fruit 92 32.1 8.6 66 2.5
Date palm Phoenix Fruit – NC 77 14.7 2.3 0.2 2.2
dactylifera
Fig Ficus carica Fruit – C 21 14.8 3.6 1.4 3.1
Grape (table) Vitis vinifera Fruit – NC 17.3 15.2 4.0 1.2 2.9
Grape (wine) Vitis vinifera Fruit – NC 18.9 15.1 3.2 0.5 2.6
Grapefruit Citrus Fruit – NC 11 14.7 6.9 1.1 3.4
paradisi
Hazelnuta Corylus Fruit 91 24.5 13 33 1.9
avellana
Kiwi Actinidia spp. Fruit – C 17 13.3 6.7 3.1 3.6
Lemon Citrus limon Fruit – NC 13 11 10 2.7 2.7
Mango Mangifera Fruit – C 18 15.2 5 2.3 3.2
indica
Oil palm Elaeis Fruit 58 23.5 7.8 47 3.6
guineensis bunch
Olive Olea europaea Fruit – NC 50 24 4.2 53 11
Orange Citrus sinensis Fruit – NC 18 14.6 7.2 1.7 3.4
Peach Prunus persica Fruit – C 12 14.6 8.1 2.2 3.8
Pear Pyrus Fruit – C 14.8 12.3 2.0 0.7 1.4
communis
Persimmon Dyospiros Fruit – C 20 14.9 2.9 1 1.7
kaki
Pinapple Ananas Fruit – NC 14 14.9 3.9 0.9 1.6
comosus
Plum Prunus Fruit – C 15 15.1 5.5 2.2 2.9
domestica
Pomegranate Punica Fruit – NC 25 15.7 7.6 5.3 2.4
granatum
Quince Cydonia Fruit – C 16 14.7 2.5 0.6 2.5
oblonga
Walnuta Juglans Fruit 93 24.7 12.3 32.1 2.4
regia
(continued)
Part
Roots, tubers & MJ/ % over
bulbs % kg dry mass
Cassava Manihot Root 37.6 17.1 2.6 0.8 2.8
esculenta
Garlic Allium sativum Bulb 39 15 15.4 1.2 3.6
Onion Allium cepa Bulb 10.9 13.8 9.2 0.9 3.7
Potato Solanum Tuber 23.5 16.9 10.8 0.5 7
tuberosum
Sugar beet Beta vulgaris Root w/o 20 16.9 7.8 0.5 6.9
crown
Sweet potato Ipomoea batatas Tuber 30 17.4 5.5 1.1 3.6
White yam Dioscorea Tuber 26.2 17.1 5.9 0.5 4.3
rotundata
Yam chinese Dioscorea Tuber 18.6 17.3 8.7 0.5 4.2
opposita
Yellow yam Dioscorea Tuber 16.6 17.3 6.2 0.4 3.2
cayenensis
a
Including shell
Bibliography
ASABE Standards. (2007). D245.6. Moisture relationships of plant-based agricultural products.

St. Joseph: ASABE.
Bakker-Arkema, F. W. (Ed.). (1999). CIGR handbook of agricultural engineering (Vol. IV).
St. Joseph: ASABE.
Lucia, M., & Assennato, D. (1994). Agricultural engineering in development: Post-harvest
operations and management of foodgrains. Rome: FAO.
Chapter 33
New Tools and Methods in Agronomy
Pablo J. Zarco-Tejada, Luciano Mateos, Elias Fereres,

Abstract As developments in information and communication technologies

increase, new opportunities are becoming available to refine agronomic manage-
ment. Precision agriculture, or site-specific crop management, is a farm manage-
ment concept based on observing and responding to intra field variability. This
concept is being increasingly applied to irrigation and fertilizer management with
the aid of remote sensors of vegetation in the visible and thermal bands and
automatic guidance systems based on precision GPS. Remote sensing is the acqui-
sition of information about an object without making physical contact by using
satellites, aircraft, and/or land platforms. Other tools include crop simulation
models which are computer programs with equations that represent the response
of the crop to management practices (e.g. planting density) and the environment
(meteorology, soil). They may be combined with remote sensing data for large scale
analyses and predictions and for refining site-specific management.
33.1 Introduction
Agriculture started around 10,000 years ago and since then, improvements of
agricultural technology have contributed to increasing the productivity. Intensive
agriculture in the second half of the twentieth century was able to achieve high
productivity from high inputs but also with some important environmental impacts.
The next step in agricultural technology is based on the use of proximal or
remote sensors and computers that allow a more precise and/or efficient application
of inputs in the field. This is based on the improved knowledge of crop physiology
P.J. Zarco-Tejada (*) • L. Mateos

E. Fereres • F.J. Villalobos

Agriculture, DOI 10.1007/978-3-319-46116-8_33
504 P.J. Zarco-Tejada et al.
and the advances of agronomy during the twentieth century. At this point we see
that agriculture evolved from low input/low control to high input/low control
during the 1950s–1960s and is now turning to optimal input based on high control.
In this chapter we review some of the technologies that contribute to better control
in crop management.
33.2 Remote Sensing Principles
33.2.1 Definition of Remote Sensing
Remote sensing can be defined as the acquisition of information about an object

without making physical contact. Although it has been generally related to the
acquisition of images from satellite platforms, the term remote sensing is actually
broader and it relates to data collection from objects without physical contact. In
other words, remote sensing is both the acquisition of data from long distances
(i.e. 36,000 km from the Earth using a satellite) as well as that obtained at just a few
centimeters using a camera or an instrument that we place near the object
(i.e. a leaf).
33.2.2 Active and Passive Remote Sensing
The remote sensing of objects using different types of sensors is necessarily

conducted through a physical carrier that travels from the objects to the sensors
through an intervening medium. This carrier in remote sensing is the electromag-
netic radiation, and the medium is the atmosphere. A general classification of
remote sensing can be made depending on the nature of the radiation used to gather
information from the objects: (i) active remote sensing is when a signal is emitted
and received by the sensor used, and (ii) passive remote sensing uses naturally
occurring energy (e.g. reflected from solar radiation).
33.2.3 The Electromagnetic Spectrum and the Spectral

Bands
Electromagnetic radiation is energy propagated through space between electric and

magnetic fields. The electromagnetic spectrum is the extent of that energy ranging
from cosmic rays, gamma rays, X-rays to ultraviolet, visible, and infrared radiation
including microwave energy as a function of their wavelength (Fig. 33.1).
33 New Tools and Methods in Agronomy 505
Fig. 33.1 The electromag Wavelength Frequency

netic spectrum classified as (μm) (MHz)
a function of wavelengths
(μm) (MHz)
10–5 3·107 .4 750
10–2 3·104
.5 600
3 102
102 3
104 3·10–2 .6 500
.7 429
106 3·10–4
Our eyes detect radiation in the so-called “visible” spectral region, ranging
between 400 and 700 nm in wavelength. Nevertheless, there are sensors that can
detect and measure radiation coming from below and above the visible spectral
region, i.e. below 400 nm and above 700 nm wavelengths. Remote sensing of
objects is generally conducted with sensors working in wavelengths in the ultra-
violet (UV) below the visible region, and then in the reflected infrared, thermal
infrared and microwaves in the spectral region above the visible part of the
spectrum. It is important to clarify that above the visible region some sensors detect
reflected radiation while others detect the emitted radiation. The reflected infrared
region therefore is used to collect data in the infrared wavelengths reflected by the
objects, while the thermal infrared region is used to gather emitted radiation to
obtain information about the object’s temperature.
33.2.4 The Concept of Multispectral
The collection of reflected or emitted data by sensors needs to be conducted in the

so-called “spectral bands”. This is because each detector used to collect reflected or
emitted energy is sensitive to a different range and number of these specific spectral
bands. The width in wavelengths units and the number of these spectral bands
enable the acquisition of data from objects in a multispectral way. Therefore,
multispectral remote sensing is defined as the collection of reflected or emitted
energy from an object in multiple bands of the electromagnetic spectrum
Aircraft/satellite
Sensor
Bands 1 2 3 4
Inc
ide
ion
Scattering/
nt r
iat
absorption
adi
ad
atio
dr
cte
n
fle
Re
Absorption
Fig. 33.2 Multispectral remote sensing
(Fig. 33.2). The more the number of bands, and the narrower these bands are, the
more information can be gathered from the object.
33.2.5 Reflectance and Spectral Signatures
The energy measured by the sensor is a function of the total incoming radiation at
such particular moment. In other words, measuring the energy reflected by the same
object at two different times would give us a different amount if the energy coming
from the sun changes. For this reason, remote sensing methods rely on normalizing
such reflected energy to the incoming radiation at the time of the measurement.
Reflectance is defined as the ratio of the total amount of radiation reflected by a
surface to the total amount of radiation incident on the surface. Remote sensing
methods are based on this reflectance as a function of wavelength, developing the
so-called spectral signatures (Fig. 33.3). In Fig. 33.3 we can observe the reflectance
measured for three different objects: (i) green vegetation; (ii) dry vegetation; and
(iii) soil. The amount of reflected energy changes as a function of the wavelength,
which is function of the pigment absorbing and reflecting energy from each object
under study. By looking and analyzing these spectral signatures it is possible to
Green Vegetation
0.6 Dry Vegetation
Soil
REFLECTANCE
0.4
0.2
0.0
0.4 0.5 0.6 0.7 1.0 1.5 2.0 2.5
WAVELENGTH (μm)
Fig. 33.3 Spectral signatures from green vegetation, dry vegetation and soil
infer the status of the vegetation and soils by estimating the amount of photosyn-
thetic pigments (chlorophyll content, carotenoids, xanthophylls), the amount of
vegetation layers (leaf area index, canopy densities), and water content, among
others.
33.2.6 Remote Sensing Resolution
There are four definitions than are critical in remote sensing to understand the data
quality and image characteristics: (i) spatial resolution. (ii) spectral resolution, (iii)
radiometric resolution and (iv) temporal resolution.
Spatial resolution is the size of a pixel that is recorded in an image typically
corresponding to square areas. The minimum detail discernible in an image is
dependent on the spatial resolution of the sensor and refers to the size of the
smallest possible feature that can be detected. The spatial resolution of passive
sensors depends primarily on their Instantaneous Field of View (IFOV) of the
system comprised by the detector and the lens. The area on the ground represented
in the pixel is called the resolution cell and determines a sensor’s maximum spatial
resolution.
Spectral resolution is the wavelength width and the number of the different
frequency bands recorded by the detector, which determine the spectral signatures
used to assess the objects by remote sensing methods.
Radiometric resolution is defined as the number of different intensities of
radiation the sensor is able to distinguish. Typically, this ranges from 8 to 14 bits,
corresponding to 256 levels of the grayscale and up to 16,384 intensities in each
band. It also depends on the instrument noise.
Temporal resolution is the frequency of satellite or plane overpasses over the

same place. The temporal resolution is relevant in time-series studies in cases in
which collecting data over the same site is required to understand seasonal changes.
33.3 Sensors and Platforms for Remote Sensing
The sensors are generally classified as: (i) optical; and (ii) microwave sensors.
Optical sensors detect visible and infrared radiation in different sub-regions:
(i) near infrared, (ii) intermediate infrared and (iii) thermal infrared. There are two
types of radiation that can be measured from optical sensors: (i) visible/near
infrared (reflected); and (ii) thermal infrared (emitted).
In the Visible and Near Infrared region the sensors detect radiation of sunlight
reflected by the objects which is used to quantify land surface conditions such as the
distribution of plants, forests and farm fields, rivers, lakes, urban areas, etc. Obvi-
ously this technique can only be used during the daytime under clear sky conditions.
In the Thermal Infrared region the detected radiation is that emitted by the objects,
which is typically used to monitor the temperature of the land’s surface (see Chap. 3)
and is not restricted to the daytime period but is affected by cloudiness.
Microwave sensors may be active or passive and are designed to measure
radiation in the microwave spectral region, with longer wavelengths than those of
the visible and infrared regions. The observation is not affected by period (day or
night) or weather as microwaves penetrate the clouds.
The platforms generally used for remote sensing can be divided into the follow-
ing classes:
(a) proximal sensing platforms are used for acquiring high spatial resolution
imagery from towers, cherry pickers, trucks and mobile platforms in the field.
(b) airborne platforms (i.e. planes of different sizes and weights) are generally used
for acquiring imagery at typical altitudes between 300 m and a few kilometers
over the ground. They can be divided into manned and unmanned aerial
vehicles.
1. Manned vehicles are the traditional platforms for aerial photography and
remote sensing operations using heavy cameras working in the visible, near
infrared and thermal regions. Access to these sensors is limited, as they have
a high cost of operation, so they are generally used for research purposes but
sometimes for operational applications. Sensors that can be installed in these
platforms are generally very expensive, but they obtain very high quality
imagery which helps validating remote sensing models and methods.
2. Unmanned aerial vehicles (UAV) also known as Remotely Piloted Aircraft
(RPAS) or “drones” are new methods recently available which are the result of
transferring this technology from the military to civil applications. These
drones are small planes that can fly autonomously over desired areas at low
altitudes, therefore acquiring high resolution imagery using miniature sensors
carried on-board. Wingspan and weight for these platforms range between less
than a meter up to 5 m, and between under 2 kg up to several hundred
kilograms. The sensors used weigh between a few grams up to several kilo-
grams, and they are currently available to collect images from drones in the
visible, near infrared and thermal regions. Despite the legal limitations on the
use of drones for civil applications, they offer high flexibility in acquiring
imagery at low cost, higher spatial resolution and easy operation by end users.
(c) The satellite platforms follow typically elliptical orbits around the earth. The
time taken to complete one revolution of the orbit is called the orbital period.
The satellite traces a path on the earth surface as it moves across the sky. As the
earth below is rotating, the satellite traces a different path on the ground in each
subsequent cycle. Remote sensing satellites are launched into orbits such that
the satellite repeats its path after a fixed time interval. This time interval is
called the repeat cycle of the satellite.
When a satellite follows an orbit parallel to the equator in the same direction as
the earth’s rotation and with the same period of 24 h, the satellite will appear
stationary with respect to the earth surface. In other words, the satellite will be
positioned all the time over the same spot over the Earth. This orbit is called
geostationary. Satellites in the geostationary orbits are located at a high altitude
of 36,000 km. The geostationary orbits are commonly used by meteorological
satellites as they can monitor large areas continuously (i.e. acquiring one image
every 30 or 60 min to monitor weather).
A near polar orbit is one with the orbital plane inclined at a small angle with
respect to the earth’s rotation axis. A satellite following a properly designed near
polar orbit passes close to the poles and is able to cover nearly the whole earth
surface in a repeat cycle. Nevertheless, earth observation satellites usually follow
sun synchronous orbits, i.e. an orbit whose altitude is such that the satellite will
always pass over a location at a given latitude at the same local solar time. In this
way, the same solar illumination condition can be achieved for the images of a
given location taken by the satellite.
As a function of the orbit (distance to the Earth) and the type of sensors carried by
the satellite platform, the satellite imaging systems can be classified into low
(>1000 m), medium (100–1000 m), high (5–100 m) and very high resolution (<5 m).
33.4 Site-Specific Irrigation Management
33.4.1 Site-Specific Crop Management
Although the term precise irrigation usually refers to the application of precise
amounts of water to crops at precise locations and at precise times – but uniformly
across the field–, precision irrigation, as part of the precision agriculture concept,
differs markedly from this common meaning. Precision agriculture, or site-specific

crop management, is a farming management concept based on observing and
responding to intra field variability. Precision agriculture has been practiced for
site-specific nutrient or pesticide application, for varying seeding rate, for control of
traffic on fields, etc. Precision irrigation falls in the precision agriculture category
since it consists on the application of water to predefined zones in a volume and at a
time needed for optimum crop production, maximum profitability or other man-
agement objective at each particular zone in the field.
Precision agriculture involves a number of methods, technologies and equip-
ment. Site-specific (SS) irrigation has so far been developed commercially for
self-propelled center pivot and linear move irrigation systems, whereas site-
specific microirrigation (drip/trickle, microsprinkler) has not yet emerged from
academia or industry. Center pivot and linear move manufacturers have produced
commercial equipment to regulate water application in time and space (VRI,
variable rate irrigation). The irrigation machine is governed by a GPS controller
and the operator can enter irrigation prescriptions in the control panel directly or
remotely. Prescriptions are based on maps of soil and crop attributes. Moreover,
field-distributed sensors can collect and transmit data to support real-time man-
agement decisions.
33.4.2 Opportunities Provided by Site-Specific Irrigation
SS-VRI adapted to center pivot or linear move machines allows stopping irrigation
over roads, ponds, water courses, rocky outcrops, or any other landscape element
that doES not require irrigation. If various crops are grown under the same center
pivot or linear move system, irrigation can be scheduled according to the needs of
each crop.
Another situation for using SS irrigation is when runoff occurs. This is more
likely to happen in steep slopes, where soil infiltration capacity is low, and at the
distal end of center pivots, where the application rate is highest. SS-VRI allows
applying water at reduced rate at the zones where infiltration should be increased to
minimize runoff. The irrigation machine should then move at slower speed to
maintain the irrigation depth uniform.
The response of yield to evapotranspiration of most crops is a linear function
(Chap. 14). If irrigation ensures that the field does not suffer water scarcity at all,
then yield would be maximal and SS irrigation would not represent any yield
advantage. However, some indeterminate crops and trees yield maximum under
moderate deficit irrigation. For those crops, any deviation from optimal evapotrans-
piration will translate into yield loss. This deviation may be caused by variations in
soil water storage due to precipitation (rain, snow or irrigation) non-uniformity, to
variability of the soil water holding capacity, and to lateral flow during or after the
precipitation event.
In very arid environments, rainfall contribution to soil water storage is usually

negligible for irrigated crops. There, uniform irrigation designed to prevent surface
flow will result in uniform soil water storage as long as irrigation is scheduled for
the field zone where the critical soil water deficit (SWDc, Chap. 20) is lowest.
Therefore, in arid and semiarid environments, SS irrigation does not help in saving
water or improving yield. The situation is different in sub-humid or Mediterranean
environments, where the contribution of rainfall to the water consumption of
irrigated crops may be significant. Using SS-VRI, the irrigation depth can be
adjusted to each zone by applying less water where the water storage is greater.
The crop will keep using the rainfall water storage, where available, so total
irrigation will be reduced.
Evapotranspiration differences may occur across the field due to differences in
canopy cover. An appropriate way to estimate evapotranspiration for SS irrigation
schedules is computing spatially distributed crop coefficients based on multispec-
tral vegetation indices that are related to the fraction of intercepted radiation as the
NDVI (Normalized Difference Vegetation Index). The so derived evapotranspira-
tion would account for the spatial variation of canopy cover and would represent the
water needs of the non-stressed crop. The idea is only valid before full canopy
cover.
33.4.3 Definition of Management Zones
The technique most widely used by site-specific irrigation practitioners to define

field zones is electromagnetic induction survey of soil apparent electrical conduc-
tivity (EMI-ECa). In non-saline soils, EMI-ECa varies primarily with soil texture,
water content, and cation exchange capacity. Therefore, EMI-ECa maps are used in
combination with soil sampling to determine those spatially variable soil properties
efficiently. EMI-ECa mapping is usually complemented by topographic, yield, soil,
and multispectral vegetation indices maps to delimit management zones. Canopy
temperature of crops entering into crop water stress could be used to determine the
spatial variability of the critical soil water deficit. If the crop is not watered for some
time after heavy rainfall or uniform irrigation, canopy temperature will increase
first where SWDc is lowest.
Continuous monitoring of soil water content or potential allows introducing real
time management decisions. The location of the sampling points may be based on
EMI-ECa maps. Neutron probes provide accurate measurements of soil water
content, but are expensive and labor intensive. Capacitance sensors are inaccurate
for determining soil water content due to the multiple environmental effects that
affect their readings. Travel time sensors (e.g. TDR) are more accurate than
capacitance sensors and are becoming less expensive and more adaptable to field
conditions. Tensiometers are difficult to maintain in the field. However, resistance
blocks can be left installed in the field and connected to dataloggers and wireless
transmitters. Their problem is that they may lose contact with the soil matrix as it
dries. The development of robust wireless networks is facilitating the installation of
sensors in the field without interference with cropping operations.
An alternative to soil water measurement is sensing plant water status. Canopy
temperature has been used since the 1980s to determine crop water stress indices.
Stationary infrared thermometers installed at strategic locations pointing to the crop
canopy, or moving sensors mounted on the sprinkler lateral to scan the crop across
the field, can be connected to dataloggers and wireless transmitters to continuously
measure canopy temperature. Satellite and airborne high-resolution visible and
thermal infrared images provide snapshots of canopy cover and temperature over
the entire field. These are zenithal views that, under partial ground cover, integrate
soil and vegetation, making difficult discriminating plant stress from dry soil
effects. Infrared sensors mounted on center pivot laterals change their orientation
with respect to the sun and the crop rows, which affects temperature readings. This
is less of a problem if mounted on linear move laterals.
There are several handicaps for the adoption of remote sensing techniques in SS
irrigation. Satellite images are inexpensive, but their frequency and/or spatial
resolution might be insufficient. Aerial imagery may be expensive for the required
frequency. The apparatus involved in mounting sensors on the moving lateral has
been a handicap for commercial application. The other barrier for adoption of this
type of information is the difficulty for farmers to understand and process this
information in a timely manner.
33.5 Positioning and Automation
High precision positioning systems like GPS are key technologies for precision,
site-specific agriculture. The systems record the geographic coordinates of the field
and management zones and locate and navigate agricultural vehicles with accuracy
of few cm.
There are different levels of automated steering. Assisted steering systems
simply show drivers the path to follow in the field, thus the farmer still needs to
steer the wheel. Automated steering systems take full control of the steering wheel
along the row, allowing the driver to watch the machine in use (sprayer, seeder).
Intelligent guidance systems allow different guidance patterns adapted to the shape
of the field.
Management zone maps, automated steering, and variable rate technology are
used jointly to adjust machines to apply, for instance, seed or fertilizer according to
the spatial variations in plant needs.
Data sensors can be mounted on moving machines also. Grain yield monitoring
is becoming very popular. It consists of devices and sensors installed in the
harvester that calculate and record grain yield and machine position as it moves.
Yield maps can be useful for delineating management zones.
33.6 Crop Simulation Models, Expert Systems

and Information Technologies
A crop simulation model is an abstract, mathematical model that captures the

behavior of the crop in response to the environment and management throughout
the growing season. The model is converted to a computer program with equa-
tions that represent the response of the crop to management practices
(e.g. planting density) and the environment (meteorological, soil). Therefore the
model behaves similarly (and roughly) than the real crop, so we may perform
virtual experiments that may help in crop management or genotype evaluation.
The obvious advantage is that the model may be applied to a wide range of
climates or soils thereby expanding the limited knowledge obtained from field
experiments. However, models are always simplified representations of actual
systems, so we have to be very careful in selecting the right model and test it
against experimental data.
Crop simulation models may be classified according to different criteria:
(a) Empirical versus mechanistic: an empirical model relates yield to a set of
environmental or management variables. For instance in arid and semi-arid
areas yield may be calculated as a linear function of seasonal rainfall. On the
other hand a mechanistic model is based on the biological, chemical and
physical mechanisms related to plant growth, development and yield.
(b) Black-box versus comprehensive: This classification is parallel to the previous.
A black-box model simply relates (empirically) the inputs and outputs of a
system, without any information on how the system works. A comprehensive
model represents the system (e.g. crop) by equations that explain the function-
ing of the organizational level below (e.g. plant, organ). Take for instance
wheat production in a farm as affected by nitrogen. We could write a black-
box model at the entry of the farm by simply recording the amount of fertilizer
N that goes into the farm and the amount of wheat that goes out. Note that in this
case we have no idea of what’s going on in there but in the long run we may be
able to predict wheat yields of the farm as a function of purchased N fertilizer.
The comprehensive model would require entering the farm, collecting data
(soil, weather, management) and setting up a model of wheat growth as affected
by N and other factors. It is clear that a black-box model is cheap and easy to
develop but is not valid for other systems.
(c) Specificity: Some crop models have been built for a single crop species
(e.g. Oilcrop-Sun for sunflower) while others are valid for any species
(e.g. Wofost). The former are more detailed and use parameters to characterize
the different cultivars while the latter use different parameters for each species.
Most existing crop models have been built for annual species while only a few
have been developed for fruit trees (e.g. OliveCan for olives) because of the lower
importance, the larger complexity and the scarcity of good data sets for calibration
and validation.
Crop simulation models are not only an alternative to field experiments but are
the only choice for evaluating crop production under non existing conditions. For
instance predictions of the impact of global change in agriculture are based on crop
simulation models. Something similar happens with the evaluation of non-existing
genotypes that may be tested using a model in order to identify characters that
should be incorporated in breeding programs
Although primitive crop models may be traced to the 1960s with the work of de
Wit in The Netherlands and Loomis in the USA, the first large scale attempt at
developing general use crop models for the main crops species (wheat, maize)
occurred in the USA during the late 1970s and was led by Ritchie, Kiniry and Jones.
The primary interest for the USA was strategic, i.e. being able to predict crop yields
of their main rivals at the time (USSR, China). Pretty soon this effort was aban-
doned and crop modelling became a very useful tool in agronomy around the world.
Crop models may be integrated in a decision support system (DSS), a software
package including also other tools for input data management and output data analysis,
such as in DSSAT (Decision Support System for Agrotechnology Transfer). A more
sophisticated tool is an expert system which is a computer application with learning
capacity intended to operate like a human expert. In general an expert system includes
a data base and sometimes also a crop simulation model (e.g. model Gossym in expert
system COMAX for cotton). Available information technologies (powerful cell
phones, wireless sensors) will facilitate the development of very specific applications
for crop management incorporation crop models, DSS and expert systems.
Bibliography
Campbell, J. B., & Wynne, R. H. (2013). Introduction to remote sensing (5th ed.). New York:
Guilford Press.
Gebbers, R., & Adamchuk, V. I. (2010). Precision agriculture and food security. Science, 327,
828–831.
Jones, H. G., & Robin, A. (2010). Remote sensing of vegetation: Principles, techniques and
applications. Oxford: Oxford U Press.
Jones, J. W., Hoogenboom, G., Porter, C. H., Boote, K. J., Batchelor, W. D., Hunt, L. A., Wilkens,
P. W., Singh, U., Gijsman, A. J., & Ritchie, T. J. (2003). The DSSAT cropping system model.
European Journal of Agronomy, 18, 235–265.
Oliver, M., Bishop, T., & Marchant, B. (Eds.). (2013). Precision agriculture for sustainability and
environmental protection. Abingdon: Routledge.
Rees, W. G. (2013). Physical principles of remote sensing (3rd ed.). Cambridge: Cambridge
University Press.
Chapter 34
Cropping and Farming Systems
Helena Gómez-Macpherson, Francisco J. Villalobos, and Elias Fereres
Abstract Cropping systems can be based on a single crop (monoculture) or on

many (polyculture), including intercrops and rotations. Intercrops are rarely used in
western agriculture although they offer several advantages (better use of resources,
improved nutrient cycling) that are quantified by the Land Equivalent Ratio (LER).
Agroforestry systems are a case of intercrop in which trees provide protection to
soil, improve the crop nutrient balance and can provide some additional useful
products. Rotations involve crop diversification in time and have many advantages
over monoculture (control of weeds, pests and diseases, improved nutrition, risk
diversification). Farming systems have interlinked components of inputs and out-
puts through processes managed to achieve economic agricultural production in
order to meet enterprise and/or household requirements.
34.1 Introduction
During the twentieth century agriculture generally evolved from low input agricul-
ture to intensive systems with high inputs of energy, inorganic fertilizers and
pesticides. During the 1960s and 1970s, the Green Revolution exported success-
fully this type of agriculture to less developed countries, mainly in the tropics.
Despite its success questions have been raised on the sustainability of these
agricultural systems that seek maximum yield, and alternative systems (more
sustainable systems) have been proposed in which long-term yield stability is
pursued with minimal impact on the environment. A sustainable system must
have some of the characteristics of a mature ecosystem (e.g. diversity), but one
H. Gómez-Macpherson (*)

Agriculture, DOI 10.1007/978-3-319-46116-8_34
must not forget that the agroecosystem exports nutrients that have to be returned in
the form of inputs (fertilizers) to maintain long term soil fertility.
34.2 Types of Cropping Systems
Cropping system refers to the crops, their sequence and the management practices on
a given field. One type of cropping system is continuous monocropping
(or monoculture) in which the field is cultivated with the same species every year,
which is characteristic of large areas of North and South America (e.g. the US Corn
Belt) and sometimes are based on high inputs of energy and fertilizers. On the other
extreme we find multiple (mixed) cropping systems that have in common the
diversification of crops in time and/or space. The multiple cropping appears to be
the oldest form of agriculture, and in fact it remains common practice in many areas
of the tropics. In most developed countries, multiple cropping has almost disappeared
and crop rotations, i.e. two or more crops grown sequentially in the same plot, are
more common. In this case, the diversification is only performed over time.
Multiple-cropping variations are described by the number of crops per year and
the degree of crop overlap. Double cropping or triple cropping signifies systems
with two or three crops grown sequentially in a single year with no overlap in
cycles. For example, in the Indogangetic Plains, farmers may cultivate two crops of
rice in 1 year, the main one during monsoon season followed by another irrigated
with a shorter cycle, or rice followed by wheat and then by berseem in 1 year. In this
case, the production goal would be not so much to achieve high yields for any single
crop but to maximize yields per unit time (from kg/ha to kg/ha/day). Intercropping
indicates that two or more crops are grown with some overlap of their growing
cycles. Relay cropping describes the planting of a second crop before the first crop
is harvested. When a crop is allowed to regrow after harvest from the crowns or root
systems, the term ratoon cropping is used. This is the case for cereals with this
regrowth capacity (e.g. barley) or sugarcane.
34.3 Intercropping
Intercropping is the cultivation of two or more crops at the same time in the same
field in order to use resources more efficiently and space out labor demand.
Intercropping encourages biodiversity in the paddock and this tends to increase
stability. In Europe, intercropping is present in pastures (mixtures of clover and
grasses), in backyard vegetable gardens and in fruit orchards (alley cropping) but is
rarely found in annual crops-based systems. The limited extent of intercropping can
be explained, firstly, because the levels of soil fertility and the availability of
inorganic fertilizers are high making it difficult to find a productive advantage.
Secondly, intercropping hardly compensates for the additional management diffi-
culties and therefore the production costs increase. On the other hand, organic
34 Cropping and Farming Systems 517
production regulation in Europe is expected to increase interest in intercropping,

including grain production for feeding purposes. This new interest should not mean
necessarily returning to old systems but developing innovative technology-oriented
new organic farming.
In developing countries in the tropics, intercropping is relatively common,
generally combining a legume with a cereal. Legume crops are intercropped in
low fertile soils because of their N fixation ability. In the Sahel, sorghum or millet is
commonly sown with cowpea. The cereal grain is used for food, the straw for
feeding and the cowpea leaves and grain can be used for both food and fodder. In
wetter environments, maize may be combined with high value food legumes such
as groundnuts or green gram. Other combinations are: deep- and shallow-rooted
crops; tall and short crops, the last being able to grow in partial shade; climbing
crop on a tall crop, e.g. beans around maize; alley cropping; fast- and slow-growing
crops, so that the fast one is harvested before the slow one starts maturity.
Intercropping systems must be carefully designed in order to limit competition
for light, water and nutrients. Intercropping also poses challenges to fulfil increased
labor requirements or to mechanize operations, particularly harvesting. The fol-
lowing elements should be considered before establishing the crops in order to
minimize competition and maximize complementarity during their cycles:
• Spatial arrangement
– Mixed cropping: the plants of the different species are distributed randomly
in the field, e.g. vetch-oat association as a forage crop.
– Row intercropping: growing two or more crops together at the same time with
at least one crop planted in rows. Includes alley cropping and strip
intercropping (with strips wide enough to facilitate the use of machines but
close enough to interact).
– Relay intercropping: establish a second crop into a standing crop before the
harvest time of the standing crop.
• Plant density: it should be reduced compared to single cropping but less in the
most important crop.
• Crop cycle/sowing date: different cycles or sowing dates would result in differ-
ent peak demands of nutrients, water and light to reduce competition between
crops. A timely planting of each crop should be followed for an effective plan.
• Plant architecture: this is a key element when considering light competition and
the possibility to use one crop as a structure for other climbing crops.
• Proper management, i.e. adequate fertilization at optimal times, localized if
there is spatial arrangement in rows, effective weed and pest control, and
efficient harvesting with minimum damage to later crops.
The potential benefits from intercropping are the following:
– Reducing the risk from pests, extreme weather events and price fluctuations.
Apart from reducing risk by diversification, the population of natural biotic pest
control agents (predators, parasites) is usually increased in intercropping.
– More efficient use of resources (light, water and nutrients) in time and space due
to the different resource requirements of the components. For example, a
complex architecture cover can enhance the interception of light. The intensive
use of resources by the intercrop also reduces their availability to weeds.
– Improved nutrient cycling in the system: The combination of species with
different temporal patterns of nutrient absorption reduces leaching losses. More-
over, crops with deep root systems absorb nutrients from deeper layers. Some of
these nutrients then return to the soil surface after mineralization of the crop
residues, and can then be used by other crops with shallow root systems.
Intercropping advantages regarding pests and diseases are not clear in all cases.
The number of parasites and predators increases with the number of plant species,
but also the number of species of potentially harmful insects and fungi may
increase. The problem can be especially severe in the case of soil fungi. When
the host plant is always present, the survival of pathogens is ensured.
The Land Equivalent Ratio (LER) is used to determine the effectiveness of
intercropping systems. It is calculated as follows:
Y I1 Y I2 Y IK X
K
Y Ii
LER ¼ þ þ þ ¼ ð34:1Þ
Y P1 Y P2 Y PK 1
Y Pi
where K is the number of crops and YIi and YPi are the yields as intercrop and pure
stand, respectively, for crop i. There is an advantage in intercropping if the resulting
LER is above 1, which typically occurs when the soil resources (water and/or
nutrients) are limiting and the species differ in their pattern of root growth or
when one species is a legume. There is a disadvantage in using intercropping if
the LER value is below 1.
The advantage of intercropping depends on the relative importance of the
different components. For example, in Table 34.1, yield and LER of vetch (legume)
and oats (cereal) mixtures in the region of Castilla-La Mancha (Spain) are shown. In
this case LER is only greater than 1 when the proportion of oats is 20 % or lower,
because oats in small proportions serves to support vetch growth as it is a creeping
Table 34.1 Yield and LAR Dry matter yield (t/ha) LER
of intercrops of vetch and oats
Percent of seeds vetch:oats Vetch Oats Total
in the region of Castilla-La
Mancha (Spain) 100:0 3.1 0 3.1 –
90:10 3.2 1.0 4.2 1.19
80:20 2.9 1.8 4.2 1.13
70:30 2.2 1.8 4.0 0.97
60:40 1.7 2.6 4.2 0.95
0:100 0 6.6 6.6 –
Adapted from Caballero R, Garcia C (1996) Cultivo y utilizacion
de la asociación veza-cereal en Castilla-La Mancha. CSIC,
Madrid, Spain
plant, so both species are benefited. However, when the proportion of oats is high,
this species has a competitive advantage because of its greater height as compared
to vetch.
Example 34.1 Yields of tropical rainfed corn and beans are 800 kg/ha and
600 kg/ha, respectively, when intercropped and 1,200 and 800 kg/ha as pure
stands.
LER ¼ 800=1200 þ 600=800 ¼ 1:42
Example 34.2 Maize-bean-squash in Central America: beans climb up maize

stalks while squash plants are established in between capturing the light that
filters down through the canopy. When compared to pure stand crops in
Mexico, intercropped maize yields were considerably higher while bean and
squash yields suffered considerable yield reductions. Maize was the most
important crop, and the beans and squash were a bonus. The LER for the
whole mixture was considerably high (1.6).
Example 34.3 Maize and soybean were intercropped for silage in Canada.
Intercrops were more cost effective than pure stands, although its success
depended on seeding rate and spatial arrangement. The best performance was
observed using 67 % of pure stand recommended planting rate in both crops.
This system resulted in an LER of 1.14. Alternate rows of maize and soybean
had higher yield (LER ¼ 1.23) but with higher costs offsetting the yield
increase.
Example 34.4 In USA, alternated strips of maize, soybean and spring wheat
in a ridge-till system was tested. The strips width was adapted to the equip-
ment widths and herbicides were applied with a ground sprayer. Strips were
east-west oriented and followed a wheat–maize–soybean pattern, with soy-
beans on the north side of the maize. Wheat was harvested before maize had a
chance to shade it. On the other hand, maize rows next to soybean strip
profited of additional incident light.
Example 34.5 Backyard garden. When radish and carrot seeds are sown at
the same time, radishes germinate and grow quickly and are harvested when
carrots are just getting established. Lettuce plants tolerate shade and, there-
fore, are good to interplant among larger vegetables. Young tomato plants
may be planted among declining pea vines to replace them on the trellis.
Intercropping two vegetables with different architecture and nutritional
value such as beet and okra or pepper and onion is being practiced in
tropical Asia.
34.4 Agroforestry Systems
The practice of including trees in farming systems is very old and is still common in
many parts of the world. Agroforestry systems are now receiving special attention,
especially in tropical areas to increase sustainability. Agroforestry systems are
forms of intercropping.
Overall, an agroforestry system is more stable than other cropping systems.
Trees act as protectors of the soil and the crop from the direct effect of wind and
rain and improve the nutrient balance of the system. If trees (e.g. Acacia spp.) or
shrubs (e.g. Leucaena spp.) from the Leguminosae family are used, they contribute
to nitrogen supply. Trees can also provide food (seeds) and firewood.
The need to conserve the soil in many agricultural systems worldwide will likely
lead to a return to farming systems where trees and annuals that act as protective
covers from erosion are associated. In Europe, the European Silvoarable Agrofor-
estry For Europe (SAFE) project has described many different alley cropping or
agroforestry systems in the region. The “dehesas” in Spain and “montados” in
Portugal are good examples of agroforestry in which oaks trees, natural or intro-
duced pastures, crops and livestock are included. Other examples include: cereals
cropping between walnut, oak, olive or fig trees; grapevine rows planted with
walnut or olive trees; vegetables in peach or dual purpose (fruit and timber) walnut
groves; maize and soybean between rows of poplar trees; and fodder beet under
cherry trees. Cover crops are also very common in orchards, particularly in areas
prone to soil erosion.
In Kenya fruit trees are intercropped with all types of herbaceous crops
such as beans, peas, potatoes, maize, millet, exotic and indigenous vegetables
when they are still young as a way of attaining food security and income
before the trees mature. Banana may be intercropped with sweet potato and
beans to reduce the incidence of weevils and nematodes, and with Grevillea
robusta for wood.
34.5 Crop Rotations
A crop rotation is a sequence of crops over time, repeated cyclically or not. The
advantages of a crop rotation compared to monoculture are partly similar to
intercropping:
– Better use of resources (water and nutrients) or improving fertility if legumes are
included.
– Better control of weeds, pests and diseases.
– Risk diversification.
– Better distribution of the means of production in the farm.
The choice of crop rotation is to be based primarily on economic factors. This
has led to monoculture in many agricultural areas. The species and their order in the
rotation should be established based on the following criteria:
(a) Duration of the cycles and environmental requirements of the species: There is
a huge variability among species and within species in the crop cycle duration
and adaptability to climatic conditions. For temperate areas we can make the
following classification in terms of planting dates for the different species:
– Autumn-winter planting:
Winter cereals: wheat, barley, rye, oats, triticale.
Grain legumes: broad bean, pea, chickpea, lentil.
Oilseeds: rapeseed, safflower, flax
– Spring planting:
Grains: maize, sorghum, rice.
Oilseeds: soybean, sunflower.
Some species may belong to different categories depending on the climatic
characteristics of the area. Sugar beet is sown in autumn in mild winter areas
(e.g. South of Spain) and in the spring in colder areas like in most European
countries. Winter cereals with low vernalization requirements may also be sown
in early spring in cold areas or in late autumn in the Mediterranean region.
This classification should not be taken strictly as the trend over the last 20 years
has been to advance the date of sowing of spring crops. For instance, when
sunflower was introduced in Spain in the 1960s, it was regarded as a spring crop
with planting in April or May, while today it is planted in many areas in February
or March. Summer crops such as maize are planted in some areas of mild climates
(California, Spain) 2 months earlier than 30 years ago. This is also due to the
increased tolerance of this crop to suboptimal temperatures, as it is being increas-
ingly grown in the cool environments of the higher latitudes. The adoption of other
cultural techniques like plastic mulching may allow an advance on the planting
date some of species (e.g. cotton in the Guadalquivir Valley in southern Spain).
(b) Time required for preparing the sowing of the following crop: After harvesting
a crop a series of operations (residue management, primary tillage, seedbed
preparation) can significantly delay the planting of the next crop. This time may
be reduced by direct seeding the following crop (e.g. direct seeded wheat after
rice harvest in South Asia).
(c) Ecological characteristics (e.g. rooting depth) and management of different crops.
Traditionally it has been recommended to alternate closely sown (cereals) and
row crops which help control weed populations. From the point of view of the
control of pests and diseases we should avoid repeating the same (or similar) crop
in the same field in consecutive years. Crop rotation is a good tool to reduce the
incidence of pests and diseases, particularly soil borne pathogens, as the absence
of the host plant causes a reduction in the inoculum in the soil. Some crop species
(and weeds) of the Cruciferae family generate glucosinolates which have insec-
ticide and fungicide effect, therefore providing a cleaning effect on the soil. In
Australia, the introduction of canola in rotation with wheat has reduced the
incidence of take-all in wheat crops in no-tilled systems
(d) Use and conservation of resources: the cropping system should help to prevent
loss of water and soil nutrients. The current situation of European agriculture
can promote the adoption of more conservative cropping systems. Some crop
management practices can be very useful although a priori they are considered
negative. For example, in rainy areas in which nitrogen is the limiting factor,
keeping a clean fallow increases nitrate leaching. If weeds are left in the field,
they will capture N in organic form and reduce soil water content, which
reduces the amount of deep percolation and therefore N leaching. The same
objective can be achieved through the use of catch crops to “capture” the N
when nitrogen leaching risk is high.
The inclusion of clean fallow (uncultivated land free of weeds) for long periods
is justified only in areas with very low rainfall, as leaving the soil bare increases the
potential for soil erosion and N losses by leaching. In principle the rotation should
contribute to keeping the soil protected by a crop canopy or by residues during the
rainy period or when winds are strong.
The inclusion of legumes in the rotation improves N supply. The contribution is
more important when the legume is incorporated into the soil as green manure
which also helps to increase the organic matter content.
The effects of some crops over others in a rotation are sometimes unclear as even
the rotation of different cultivars of the same species (for example maize) may have a
positive effect. It is believed that an important part of the interaction is related to the
maintenance of a large microbial biomass in the soil, capable of rapid mineralization.
34.6 Designing Crop Rotations
The primary concern for choosing crops in a rotation is farmer’s profit. That implies
reducing costs and optimizing farm production means. In addition the rotation
design should be flexible enough to accommodate possible changes and be suitable
for the environmental and management (e.g. soil conservation) conditions. How-
ever, for increasing the sustainability of farming, several additional criteria should
be considered:
(a) Conservation of resources (water, soil, nutrients)
– Minimize the duration of periods without a crop
– In periods without crops use crop residues to protect the soil surface
– Include crops with deep rooting systems to use the N deep in the profile
– Include legumes to improve the N supply
– Maintain/increase levels of organic matter
– Switch between species with different nutrient requirements
(b) Control of weeds, pests and aerial diseases
– Alternate crops (the more they differ, the better) and avoid contiguity with the
same or similar crops. When we talk about different crops we do not only
mean in botanical terms but also in relation with management, like for
instance, winter versus spring crops or wide versus narrow row spacing.
(c) Control of soil fungi
– Increase the time between planting the same crop or similar (same type, same
family)
– Include cruciferous crops (containing glucosinolates which result in com-
pounds having fungicidal, insecticidal and herbicide effects such as
isothiocyanates)
34.7 Farming Systems
A farming system is defined as a population of farms that have similar resource

bases, enterprise patterns, household livelihoods and constraints. Farming systems
may be classified based on the following criteria:
– Available natural resource base, including water, land, grazing areas and forest;
climate, of which altitude is one important determinant; landscape, including
slope; farm size, tenure and organization; and,
– Dominant pattern of farm activities and household livelihoods, including field
crops, livestock, trees, aquaculture, hunting and gathering, processing and
off-farm activities; and taking into account the main technologies used, which
determine the intensity of production and integration of crops, livestock and
other activities.
In subsistence farming systems, farmers produce food for themselves and their
family and there is no profit, e.g. rice production in Bangladesh or rainfed sorghum-
cowpea in the Sahel, whereas in commercial farming, farmers would sell their crops
and animals to make a profit as in most European farming systems. In general,
intensive systems require high inputs of capital, labor or technology to achieve high
outputs or yields per hectare; the farms are usually small, for example protected
agriculture in Almeria or pig production in Denmark. Extensive systems are,
however, characterized by low use of inputs, large areas of land and low outputs
or yields per hectare, e.g. wheat-sunflower production in the rainfed areas of
Andalusia, Spain. Arable is the growing of crops, pastoral is the keeping of animals
and mixed is when farmers grow crops and rear animals. Sedentary is when the
settlement is permanent and the landscape farmed every year whereas nomadic
farmers move around looking for fresh pasture or new plots to cultivate. There are
extensive subsistence systems, e.g. nomadic pastoralism in Africa and Central Asia,
or intensive subsistence systems, e.g. rice-based farming systems in the Sahel.
Among the most complex, there are rainfed systems in humid tropics of high
resource potential, characterized by a crop activity (typically cereals, cassava,
banana, coffee, etc. at small scale or in plantations, and commercial horticulture),
often mixed with livestock production.
Farming system represents a resource management strategy to achieve economic
and sustain agricultural production in order to meet some household requirements.
Farming systems are not static as they have interlinked components of inputs and
outputs through processes (Fig. 34.1). System management should give the crop its
best chance of expressing its potential. For this, an understanding of the system is
required. Firstly, the inputs, processes and outputs, then, the influence of natural
(soil, slope, rain, temperature, sunshine, etc.) and human inputs (labor, machinery,
energy, political, etc.) on the processes and outputs. Their combined effects on the
SUBSISTENCE:
output consumed by
family
COMMERCIAL:
INPUTS output profit
PROCESSES invested in farm
Temperature To turn inputs into
Rain outputs
Soil OUTPUTS
Slope Activities needed to
Radiation … grow crops or rear Products produced:
animals: - Crop fruit/grain
- Crop straw
- Sowing
Labour - Weeding - Milk
Machinery - Irrigating - Livestock…
Energy Not harvested:
- Harvesting
Fertilizers - Losses
- Milking …
Pesticides… - Mulch…
External influence: market price, policies…
Fig. 34.1 Inputs, processes and outputs in a farming system

scale of production, methods of organization and the products should be understood

for optimizing the system, also in the mid and long term.
Understanding farm household decision-making is essential for targeting
research, fostering innovation and accelerating adoption of innovations. How to
manipulate natural inputs for using them optimally and to avoid any waste? Could
more of the inputs supporting diseases, weeds and pests be reallocated to grain? Are
diseases building up, should a changed crop rotation be considered to control pests
and diseases? Could the crop stubble be used to increase production of the next
crops? Could water inputs be used more efficiently for producing grain? Is the
management system environmentally sustainable or are soil and water resources
being gradually downgraded or overused? Is current irrigation management pro-
gressively increasing soil salinity? Are the yield targets too high for the location?
Would a lower yield target lead to a more efficient use of resources?
Bibliography
Dixon, J., Gulliver, A., & Gibbon, D. (2001). Farming systems and poverty: Improving farmers’
livelihoods in a changing world. Rome: FAO and World Bank. 412 pp.
Francis, A. (1986). Multiple cropping systems. New York: Macmillan Publishing Company.
383 pp.
Ouma, G., & Jeruto, P. (2010). Sustainable horticultural crop production through intercropping:
The case of fruits and vegetable crops: A review. Agriculture and Biology Journal of North
America, 1(5), 1098–1105.
Sullivan P., (2003). Intercropping principles and production practices. ATTRA publication,
Agronomy systems guide. National Center for Appropriate Technology, USA. http://el.
doccentre.info/eldoc1/k33_/intercropping-principles.pdf
Vandermeer, J. (1989). The ecology of intercropping. Cambridge, UK: Cambridge University
Press.
Chapter 35
Agronomy and the Sustainability of Crop
Production
Abstract Lessons learned since the discovery of agriculture suggest that good
agronomy as an integrative science is essential for improving the sustainability of
current agricultural systems. To meet the challenges of producing sufficient, nutri-
tious food for a growing population, future agronomists will have to combine
advances in plant breeding and biotechnology with new approaches to improve
the efficiency of nutrient and water use in agricultural production. There is signif-
icant potential in many areas to increase yields by bridging the gap between
potential and actual yields, but as average yields increase with time, such potential
diminishes. The threats of soil degradation and water scarcity will require wide-
spread adoption of conservation practices based on strong extension efforts and the
use of new IT technologies. Global change will have positive and negative out-
comes in the agriculture of different regions, but will introduce more uncertainty in
defining the best strategies to cope with climate variability. The most likely path to
the sustainable intensification of production would be through continuous, small
productivity improvements rather than through a few revolutionary discoveries, at
least in the medium term.
35.1 Introduction
Agriculture started with the domestication of cereals around 10,000 years ago
(10,000 BP). Today the same species (wheat, rice, maize) constitute the basis for
global food production. Much before 10,000 BP seeds from some grass species
were collected and processed to increase digestibility, as part of a diverse diet that
included fruits, animals and fish. Climate variations (colder, drier periods) probably
led to a reduction in the availability of natural food sources, making it difficult to
gather wild plants and to hunt animals in sufficient amounts. That explains why


Agriculture, DOI 10.1007/978-3-319-46116-8_35
humans were forced to move up the trophic chain as herbivores have greater
conversion efficiency than carnivores. It was found that productivity of natural
grass populations increased with some management operations (e.g. weeding).
While doing so these proto-farmers, unconsciously, started selecting plants with
favorable characteristics (larger seeds, lack of dehiscence, absence of dormancy),
and after harvest, some seeds were saved for planting subsequently in other fields.
This process of domestication occurred independently at about the same time in
several different areas of the world such as the Far East, Mesoamerica, and in the
Near East where wheat and barley originated. Rice cultivation in China began
11,500 years ago, while squash was domesticated in Central America about
9000 years ago. That was the start, and very rapidly a few species of cereals and
legumes achieved the desired agricultural characteristics. At the same time early
farmers probably observed the advantage of concentrating useful plants in fields
which could be protected from herbivores or neighbors and cleaned from other
competing plants. This also allowed for a more efficient harvest.
Most hunter-gatherers had a very varied diet of wild plants and animals although
in some cases they subsisted almost entirely on meat or on a few plant species. As
agriculture developed, some wild species were selected under domestication for
different purposes, leading to quite different crops. For instance, Brassica oleracea
has been selected for its leaves (cabbage), stems (kohlrabi), flower shoots (cauli-
flower) and buds (Brussels sprouts).
With agriculture started the development of modern civilization whereby
increases in food production led to technological development, because food
surpluses could be used to feed full-time craftspeople and inventors, leading also
to the diversification of human activities. It also led to social stratification, political
centralization and militarization by feeding full-time aristocrats, bureaucrats and
soldiers. These advantages enabled agricultural societies to eventually displace
most hunter–gatherers around the world towards marginal environments.
Early agriculture was rainfed so it could only thrive in areas where enough
rainfall could sustain grain production, which in the case of wheat and barley
represents a minimum of 200–300 mm/year. In the arid zones, where precipitation
was erratic and insufficient to sustain crop production, irrigated agriculture
appeared first in 6000 BP in Egypt and Mesopotamia by merely diverting water
from rivers to adjacent fields during periods of flood. This soon evolved into
sophisticated systems of water distribution which required a strong social organi-
zation for operation and maintenance. Interestingly, lack of knowledge about the
need to control salinity and excess water from irrigation through drainage led to the
decline of some ancient civilizations in the arid areas of the Near East that expanded
based on irrigated agriculture.
Continuous cropping of the same field soon showed declining yields due to the
loss in soil fertility from extraction and cultivation. This led to shifting cultivation
systems such as slash and burn agriculture, a primitive mode of rotation aimed at
concentrating mineral nutrients after many years of forest growth and then releasing
them by burning the vegetation, which allowed a few years of cultivation with
sufficient production. This form of agriculture could be made sustainable if the
35 Agronomy and the Sustainability of Crop Production 529
turnaround time for burning the forest is long enough to allow for building back the
natural soil fertility. However, population growth increased the pressure on land use
and slash and burn expanded in many world areas and was in the end responsible for
the deforestation and land degradation of many regions, when population pressure
led to unsustainably low ratios of forest to cropped land.
The different agricultural techniques evolved in parallel. Tillage started using
the ard which only cuts a small furrow (drill) in the soil and is therefore helpful for
sowing but not for weed control, incorporation of residues or clearing new land. The
ard appeared around 7000 BP in parallel with the domestication of cattle. In fact,
the most primitive form of planting must have used a stick to drill a hole in the
ground, place the seeds and covered them with soil, a practice that was used by most
indigenous societies. Moldboard plows appeared much later and were designed to
turn the soil for more effective weed control. The plow, pulled by man or animals,
became popular in Europe around 1500 AD allowing a more complete and deeper
soil disturbance, and the upturning of the soil which was the only way to control
aggressive weed invasions. The Europeans exported the plow to America, Asia and
Africa where it facilitated greatly the expansion of commercial agriculture with
limited human labor inputs. In some areas, particularly within the tropics and
subtropics, the use of the moldboard plow has become clearly unsustainable due
to enhanced soil erosion and land degradation problems.
Animal husbandry also evolved in parallel with crop agriculture. Domesticated
animals not only provided for food and clothing but contributed as draft power for
tillage, allowed the exploitation as pastures of lands which were unsuitable for crop
production, and contributed to nutrient cycling by redistributing nutrients within the
agricultural systems. Other domesticated animals had a more specific role like cats
as hunters of grain-eating rodents or dogs as guardians in rural areas. Production of
animals and their products by grazing pasture and range lands in an extensive
fashion is a practice that has ecological values such as contributing to the conser-
vation of biodiversity.
Giant steps forward in agricultural science and technology have taken place in
the last two centuries with the development of machinery, breeding of new culti-
vars, use of mineral fertilizers and pesticides, at an accelerated pace since the
middle of the twentieth century. Modernization of agriculture has led to the
separation of the different activities that once were all part of the life of farms,
and has transformed human society. Prior to mechanization, the manual labor of
most farming activities represented a physical effort that required large numbers of
farm workers which had very low productivity. Additionally, life as a farm worker
was not very pleasant and that was one of the incentives for introducing the
mechanization of many farm operations. As technology improved, more speciali-
zation was required so that farmers could concentrate on fewer activities for which
they had to develop the proper skills and afford the required machinery and
infrastructure. Rural societies thus experienced a revolution that changed how
farming was conducted and led to land consolidation, fewer farmers, and vast
migration movements from rural areas to cities in search of a better life. Before
mechanization, farmers exploited crops, pastures and forests using animals for
different uses (food, draft, transportation). At that time, life of most of world
population was based on agricultural activities, and even by 1950 more than 70 %
of the population lived in rural areas. Since the 1950s, the challenge of feeding a
population growing at unprecedented rates was more than met by an evolving
agriculture in what is now considered one of the most remarkable success stories
of mankind. In 1950, there were about 2500 million people in rural areas and about
750 million in the cities. By 2014, the total population had reached 7200 million
with almost 4000 million in urban dwellings and yet, agriculture produces now
25 % more calories per capita than in 1950. There continues to be, however, serious
limitations in food distribution and access, primarily in rural areas, as evidenced by
the persistent existence of extreme poverty and hunger in hundreds of millions of
persons in the Planet. Furthermore, there are concerns regarding the sustainability
of present agricultural systems as to whether the recent productivity increases have
been achieved at the cost of resource base degradation, with the ultimate conse-
quence of a decline in productivity in the future.
Although there is wide diversity among the different agricultural systems cur-
rently in existence, commercial agriculture has now been transformed into a set of
industries where crops or animals are grown that may provide inputs for each other
but operate in isolation. This new specialized agriculture has been very successful
in increasing productivity but its long term sustainability remains unclear. In this
chapter we will discuss some important issues which represent future threats and
challenges to agriculture and food production as it is presently carried out.
35.2 Climate, Soil and Water
Agriculture takes place outdoors and plants are primarily dependent on the weather
around them for growth and development, and on the soil for nutrient and water
supply. The climate of a location determines what species can be grown and, also,
the level of risk that a farmer faces if he selects crops that may be sensitive to the
anticipated climatic features. As the climate becomes more limiting, the risk of very
low yields or even crop failures increases (risk being the product of probability and
impact) and crop choice is reached as a compromise between profit expectations
and risks. While agriculture has always been pushing at the margins by approaching
the climatic limits of crop viability, farmers are risk avoiders and always try to
balance profitability against risk. An additional factor that must be considered is the
normal climate variability that agriculture must deal with every season. Some of the
variability can be explained by regional phenomena such as the warming of ocean
waters in the Pacific, an event called El Ni~no, which occurs with a periodicity of
several years and causes excess rainfall in some regions and drought in others. In
some areas such as Eastern Australia, predictive tools based on prior observations
of El Ni~no events coupled with barometric pressure oscillations (El Ni~no-Southern
Oscillation) have been developed to anticipate whether the upcoming season would
be wetter or drier than normal. This information is critical to design seasonal
strategies leading to the optimization of planting dates, and of investments in the

application of fertilizers and other inputs. Seasonal predictions are still in their
infancy in many world areas but improving their reliability is essential to greatly
reduce the risk levels in farming.
35.2.1 Climate Change
Imbedded in the climate variability that agriculture has experienced since its
invention, there is now a general consensus that the climate is currently warming
as CO2 concentration increases due to human activities (burning of fossil fuels,
deforestation). The average temperature of the Earth has increased about 0.8 C
since 1880, while the magnitude of future warming is uncertain depending on the
Global Circulation Model used and the future scenario of CO2 emissions. One may
expect a temperature increase of 1–4 C with CO2 concentrations between 500 and
700 ppm by the end of this century. The possible impact of this global change on
agriculture has been studied mainly using simulation models (Chap. 33) that predict
a general decrease in agricultural productivity. However most studies have not fully
considered the positive effects of elevated CO2 concentration on photosynthesis
and assume that agricultural systems do not adapt to change. Nevertheless, crop
management will adapt to and mitigate, at least partially, any possible effects of
environmental change. On the other hand a higher CO2 concentration leads to lower
canopy conductance and will increase photosynthesis in C3 plants. Some of the
main effects of global warming on crop performance include:
– Accelerated crop development with shortening of the growing cycle in annual
species. This would reduce yields but may be offset by earlier plantings and/or
by changing the cultivar (using longer cycle or responsive to photoperiod
varieties).
– Increased evaporative demand: many studies have concluded that ET will
increase with global change as reference ET (Chap. 10), which is calculated
with meteorological data only, will increase as air temperature and vapor
pressure deficit (VPD) increase. This is misleading as the increase in CO2
concentration will increase canopy resistance, i.e. stomatal aperture will be
reduced. In the end, the increase in VPD due to warming may be offset by
partial stomatal closure, so ET probably will hardly change.
– While global warming should lead to a global increase in precipitation, it is not
possible at this time to predict future changes in precipitation at the regional
level with any degree of certainty. Many models predict an increase in the
frequency of extreme events, droughts and floods, as the hydrologic cycle
intensifies due to global change. However, such predictions have not been
validated yet and should be taken with caution, although agriculture would be
much more vulnerable to an increase in the frequency of extreme events than to a
gradual change in temperature.
– Changes in biotic factors (changes in the incidence of insects, diseases, weeds)

may occur but the actual outcome is hard to predict. Some pests would be more
damaging than at present and others would be less, while new pests may emerge
in locations where they did not exist now. For instance milder winters may favor
survival of insects while warmer springs would promote growth of C4 weeds
which compete favorably with C3 crops.
Agriculture has contributed in the past to global change primarily through the
change in land use as agricultural lands greatly expanded since the 1800s until
1960. Since that time, the surface area devoted to crops has remained more or less
constant and its contribution to greenhouse gas emissions has been modest relative
to those from industry and transportation. While there are agricultural practices that
contribute to mitigation (e.g. minimum tillage, Chap. 18), the capacity of agricul-
ture to mitigate global warming by sequestering carbon is also modest relative to
options derived from changes in the energy and transportation sectors. Emphasis
should be placed on adapting agriculture to global change using the combination of
breeding agronomy management that has been so successful until now, while
at the same time, taking advantage of mitigation options that may also contribute to
the sustainability of agriculture (see below).
35.2.2 Soil Degradation
Over the centuries, the conversion of lands for use in crop production has included
the development of fragile areas which are prone to degradation. Exploitation of
soils without maintaining their fertility by restoring nutrient extraction and their
physical properties (Chap. 26) also leads to soil degradation. Exposure of bare soil
surfaces to rainfall and tillage operations enhance the rate of natural soil loss or
erosion of the surface layers which normally are the most fertile. A single soil
erosion episode represents an amount of soil loss that exceeds by orders of magni-
tude the rate of soil formation. Despite the advances in methods of Earth observa-
tion, there are no good statistics of the degree of soil degradation around the world
but estimates indicate that the problem is very relevant, requiring periodic moni-
toring to assess its severity in the different regions. Soil erosion will continue to be a
major threat to sustainability of agricultural systems around the world. The expan-
sion of conservation agriculture (Chap. 18) is helping in many areas to control
erosion but it requires the adaptation of agricultural techniques (machinery, culti-
vars, pest control) to site specific conditions and cannot be used in all agricultural
systems. There are soils that require periodic tillage to maintain some physical
properties and in some regions, crop residues needed to protect the soil surface as
part of conservation agriculture are used for animal feed and therefore are not
available for soil protection. Intensification may lead to the production of enough
crop residues to both uses. Soil salinization is another threat to sustainability that
affects possibly up to 15–20 % of world irrigated area. Again new monitoring
methods can reduce the risk and help to introduce salinity control measures to
prevent the problem (Chap. 22).
The maintenance of soil fertility in the long term is essential to ensure the
sustainability of agriculture. This is particularly important concerning phospho-
rus, as sources for P fertilizers are limited (Chap. 26). Efforts here should focus on
P recycling and on increasing the availability of soil P to plants. In the case of N,
the availability of N fertilizers will depend on energy prices so the inclusion of
legumes in crop rotations would be a partial solution when needed. Nevertheless,
the use of synthetic N fertilizers in agriculture is extremely efficient in energy
terms. If N concentration in grain is around 2 %, each additional kilogram of N
added to the crop will support a yield increase of 50 kg. The average energy
required for producing the N fertilizer is 77 MJ/kg (Chapter 7) and since the
energy content of grain is around 18 MJ/kg, the marginal efficiency would be
900/77 ¼ 11.7. Innovative approaches for improving the efficiency of N fertilizer
use will reduce N fertilization rates and the consequent non-source pollution
which affects surrounding ecosystems and water quality in many intensive pro-
duction areas.
35.2.3 Water Scarcity
Irrigated agriculture expanded greatly in the second half of the twentieth century,
increasing from 120 to about 300 million ha. In fact, given that the productivity of
irrigated systems is about 2.75 times more than that of rainfed systems on a
worldwide basis, today the production of sufficient food relies significantly on
irrigated agriculture. Irrigation expansion has come at a cost from the environmen-
tal point of view. On the one hand, the construction of reservoirs for irrigation has
changed the natural environment and had an impact on river ecology. On the other
hand, the return flows from irrigated lands constitute a major source of non-point
pollution, which is unavoidable to some extent if irrigated agriculture is to be
sustainable. This is because the maintenance of salt balance through drainage is
essential to prevent salinization of large irrigated areas. Additionally, irrigated area
expansion and production intensification require large amounts of water, to the
point that irrigation is the primary consumer of the water diverted by man for
various uses. More of two thirds of diverted water is consumed in irrigation
worldwide. Contrary to other uses (for example, domestic) where water used can
be recovered and reused after appropriate treatment, much of the water used in
irrigation is evaporated and thus leaves the basin. While a fraction of the irrigation
water can be reused, as irrigation becomes more efficient, such fraction diminishes
and the ET process dominates irrigation water use. With efficiency increases, due
attention must be paid to the maintenance of salt balance in areas of low rainfall
and/or where saline waters are used for irrigation.
An emerging problem that threatens the sustainability of irrigation in some areas
is the excessive use of groundwater beyond the long-term supply. Groundwater
usage may exceed aquifer recharge in droughty years provided that the excess
extraction is eventually replenished in the long run. However, long-term decline in
water table depth as it is occurring now in some regions of China and India, among
others, is an indication of unsustainable use. In extreme cases, land subsidence can
reduce aquifer capacity permanently or cause sea water intrusion in coastal areas
with the permanent deterioration of water quality. Better assessment of groundwa-
ter resources combined with recharge programs and wise and strict resource
management can bring solutions to this problem.
At present, irrigation is under scrutiny by the other sectors of society that
perceive that its share of water usage is too high. This is particularly critical in
areas or times of water scarcity, where competition with other uses becomes fierce
and urban and other demands have higher priority. Thus, while there is a need to
expand irrigation as one option for production intensification to meet future food
demands, competition for scarce water with other sectors, including the environ-
ment, is going to restrict such expansion forcing irrigated agriculture to do more
with less water. Although many advanced technologies are available for improving
irrigation management, widespread dissemination has been limited so far. The time
has come for many irrigated areas to promote on a large scale the adoption of
efficient irrigation practices in order to meet both increased productivity needs and
societal goals. Independent certification of efficient use of water in food production
with appropriate indicators will be welcomed by consumers and the rest of the
society.
Promoting the efficient use of water in rainfed agriculture is also a very prom-
ising goal for production intensification in the future. The approaches should focus
on other factors co-limiting yields (such as nutrients) and on accepting more risks,
abandoning the conservative approaches to rainfed farming that avoided risk but
that had little reward on good years. Acceptance of more risk in rainfed farming
requires new tools such as reliable seasonal forecasts, and advisory services that
will assess risks quantitatively and will offer flexible options adapted to local
conditions. In its simplest view, risk equals the product of probability by impact,
and the avoidance of extreme events that could impact the viability of farming
irreversibly causing famine has dominated past rainfed strategies. In this regard, the
resilience of the agricultural system, that is its capacity to recover after a perturba-
tion, is critical for the sustainability of the system. As new technologies and policies
enhance the resilience of rainfed systems, accepting more risk will lead to produc-
tivity increases in the future.
35.3 The Role of Plant Breeding
The development of modern plant breeding technologies after 1950 has produced
new cultivars which are highly productive and widely adapted. The major plant
feature that has been improved in the major crops is its harvest index whereby
current varieties have HI values that are 50 % greater than those of 50 years ago.
The success of the recent agricultural intensification has often been attributed to the
new varieties without recognizing that varieties or agronomic inputs produce
nothing in isolation. It was the combination of new varieties and new agronomy
together with adequate management what has enhanced the productivity of agri-
cultural systems until now.
Plant breeding techniques are nowadays more powerful and more efficient due to
genetic engineering which has led to the production of new cultivars labeled
‘transgenic’ crops (Genetically Modified Organisms, GMO). Transgenic crops
have been highly successful so far by addressing crop features than are related
only to a few genes. For example, the quality of the seed may be improved
(e.g. yellow rice) or the plant may acquire insecticidal properties (e.g. BT maize
or cotton) or resistance to a given herbicide (e.g. resistance to Roundup in soybean).
The primary goals were to reduce production costs (by applying less pesticides)
and, by reducing/eliminating usage of some pesticides, to contribute to improved
human health and to the environment. The improvements in farm profitability have
been such that transgenic soybeans, maize and cotton have been widely adopted in
less than 20 years, not only in the USA where more than 90 % of the three crops are
now transgenic, but also in some developing countries, as India or China. Plant
breeding efforts to produce transgenics are now being extended to other crops to
address biotic stresses or crop quality problems.
By contrast, the promises of improving plants against abiotic stress (drought/
salinity) using GMOs have not been fulfilled so far. This is firstly due to the
complex nature of the problem. What is drought? Is the pattern of water deficit
the same every year? Should we look for plants that are “water savers” or “water
expenders”? The former would grow slowly thus allowing more soil evaporation to
occur but would generally have more water for completing seed growth, thereby
ensuring a high HI. On the contrary a “water expender” leads to higher biomass
production and probably higher yield in good years at the expense of lower HI and
yield in bad years. Thus, the best cultivar for rainfed conditions depends on local
conditions (climate, soil) and changes from year to year. Furthermore, the tight
relationship between assimilation and transpiration (Chap. 14) must be considered.
Water use efficiency is mostly dependent on the evaporative demand (air VPD) so
little can be achieved by breeding for high WUE under specific conditions. Breed-
ing for high WUE could result in cactus-like cultivars that would keep their stomata
closed most of the time! Breeding efforts should be directed instead at manipulating
development to fit the most probable drought patterns and to tuning stomatal
aperture to periods of low VPD.
Despite the success of the first transgenic crops, there are concerns on the use of
this technology mostly related to perceived risks in food safety and the environ-
ment, and to the loss of autonomy of farmers for seed production. The risk for
humans is unfounded and unfair as there are strict regulations regarding food safety
and environmental impact assessment during the breeding process. Additionally,
the improved GM varieties are allowing an important reduction in pesticide use thus
reducing a potential toxic effect. The other concerns deal with broad social issues
and intellectual property rights and is beyond the scope of this book. Is agricultural
technology such as transgenic crops which are in the hands of a few private
companies a real menace to small farmers around the world?
Plant breeding has been extremely effective not only in contributing to increased
productivity, but in adapting crops to new environments. This will be even more
important as global warming continues and crops will have to be adapted to warmer
environments or to cold areas of the higher latitudes that until now have not been
suitable for agriculture. Every major crop species has many thousands of different
varieties offering wide adaptation that can be tested and adapted to specific envi-
ronments through conventional and modern plant breeding combined with new
agronomy and management, thus, as in the recent past, crop adaptation will be a
very important target for the future of agriculture.
35.4 Alternative Agricultural Systems: Organic Farming
The intensification of agricultural production of recent decades with the extensive

pesticide use and the episodes of environmental non-point pollution have given way
to alternative movements that question mainstream agricultural practices, viewing
them as unsustainable and unhealthy. As a result, other forms of agriculture have
been proposed, some based on avoiding the use of synthetic chemical inputs and
others that combine different practices using extensively traditional knowledge.
These alternative movements have been met with positive views from some urban
societies around the world that perceive ‘industrial’ agriculture as a threat to human
well-being and to the environment.
The most popular alternative agriculture system is organic farming based on
using only organic fertilizers such as manure and plant protection methods that
forbid the use of synthetic pesticides and are founded on biological pest control.
Eliminating pesticide use has been welcomed by consumers and reduces the
environmental impact of agriculture but organic farming has also established a
set of rules without scientific basis, particularly those related to soil fertility, which
are solely based on the naı̈ve idea that natural is good and synthetic is bad.
Molecules such as nitrate, are exactly the same independent of the origin of the
fertilizer, so they produce the same benefits to the crop or may lead to the same
environmental problem (groundwater pollution). Thus when their systems are
based on following a set of strict rules, organic farmers may be condemned to
low yields/income if organic fertilizers are scarce and/or expensive. Often, addi-
tional land is needed to fix the N needed in the soil through the use of cover
crops. While organic agriculture has been very successful in finding a market
niche among the urbanites of affluent societies, the feasibility of expanding organic
farming beyond a relatively small share of world agricultural production is highly
questionable. Global N fertilizer production in 2010 was around 100 Mt N. If we
eliminated completely synthetic fertilizers we would require legumes incorporated
into the soil as green manure. Assuming an average input of 100 kg N/ha/year, that
would take 109 ha which is clearly impossible to achieve as total arable land is only
1.5 109 ha. In other words, green fertilization would reduce current world produc-
tive arable land to one third of the current value.
35.5 Agriculture as an Energy Source
Primary production is an inexhaustible source of energy and therefore has been

used by man since long ago. It was during the energy crisis of the late 1970s when
agriculture was first considered as a potential source of energy, either through
novel, energy crops or using some of the main crops for converting biomass and
grain into fuels. Since that time, most of the energy crops have not fulfilled their
initial promises (although newly tested C4 species such as Miscanthus might be a
viable option) and the focus has shifted to ethanol production from sugarcane and
maize, with some attention paid to converting edible oils into biodiesel. The
contribution of fossil fuels to global warming and the high prices of oil in the
recent past have fostered policies for promoting the use of biofuels produced in
agricultural lands, particularly in South (sugarcane) and North America (maize).
Globally, while the use of biofuels reduces greenhouse gas emissions, the com-
petition between food and energy production is the subject of debate in ethical and
environmental terms. For instance, the incentives for producing biofuels have
contributed to the expansion of oil palm production at the expense of food crops or
of the maintenance of tropical forests thus increasing deforestation. This debate is
sided by proposals to use only crop residues as the energy source. This promise of
“second generation” biofuels based on the use of residues by conversion of
cellulose to sugars, which would then turn into alcohol has been achieved in
technical terms, although production costs are still high relative to those of
biofuels from sugarcane or maize. The claim that by using only crop residues
there is no competition with food production is not valid for two reasons: first,
crop residues do have an important role in soil conservation and in maintaining
soil organic matter (Chap. 18), and as animal fodder in many agricultural systems
and second, if sugars can be produced, then they could also be used for food
production.
The relative capacity of agriculture as a potential energy producer may be
quantified by comparing the energy contained in food products, against the energy
burnt in fossil fuels. The total global consumption in 2010 of liquid fossil fuels
(gasoline, refined fuel oils, etc) which is mostly used in transportation was around
70 million barrels/day which is equivalent to 135 1012 MJ/year. For the same year,
global agricultural production was 3866 Mt of dry matter (Table 35.1) which
corresponds to a total energy of 71 1012 MJ which is less than 50 % of the energy
of liquid fuels. The EU has established the goal of supplying 10 % of fuel as
biofuels by 2020. If that rule is applied globally it would require 13.5 1012 MJ
which is equivalent to almost 20 % of agricultural production.
Table 35.1 Global crop production in 2012 classified by crop type and equivalent energy
captured
Crop production Energy content Equivalent energy
Crop type Mt dry matter MJ/kg EJ
Grain 2276 17 38.70
Oil 450 27 12.15
Legumes 317 19 6.03
Sugar 307 17 5.22
Starch 230 17 3.91
Fruits 157 17 2.67
Vegs 74 17 1.27
Non food 38 17 0.65
Other 15 17 0.26
Total 3866 70.85
35.6 The Role of Research, Extension and Information/

Communication Technologies
The returns on past investments in agricultural research have been so high that some
have termed agricultural research as the best business of the public sector ever.
Modern agricultural research started in the last decades of the nineteenth century,
primarily in Germany, USA and England. After the Second World War, in view of
the need to produce more food for a growing population, there was an initiative led
by private foundations and some countries to develop a system of international
agricultural research which eventually became the Consultative Group of Interna-
tional Agricultural Research (CGIAR) with research centers located in developing
countries. The CGIAR developed the first cultivars of dwarf wheat and rice that
were more productive than existing tall cultivars, leading what was later called ‘the
Green Revolution’. All countries have since developed their agricultural research
systems which have contributed to the sustained increases in food production
worldwide since 1950.
Along with agricultural research, some countries such as the USA developed in
parallel a system for disseminating the new results among farmers to promote
adoption of new techniques as they were developed by researchers. Agricultural
extension has also been very successful and there are many examples of successful
adoption of new techniques that were experimented locally and tested by extension.
Many of the newly developed techniques require adaptation to local conditions as a
prerequisite for adoption by farmers. Without a good extension system, farmers
hesitate in adopting new ideas that have not been adapted and tested locally, and
progress is slower. Also, being extension part of the public sector, they are
independent of private corporations and free of biases towards certain varieties or
products. Agricultural extension started in the USA before the end of the nineteenth
century and has been largely responsible for the expansion and productivity
increases of US agriculture. Other countries have created effective extension
systems but many developing countries have not invested sufficiently in agricul-
tural extension, and this is limiting the rate of adoption of effective solutions that
increase productivity and sustainability. One limitation is the huge number of small
farmers that exist in many countries which will require a very large extension force
to carry out the work in the field, if extension is to be conducted in the way it has
been until now (face to face). However, new communication technologies such as
cell phones which are readily available in most areas could serve as innovative
ways to reach the large populations of small farmers effectively.
In general, communication technologies have accelerated the access to vast
amounts of information but cannot guarantee its quality. Information delivered by
private companies is often biased towards the benefits of their products and
sometimes it escapes regulations on false advertising. It is common to see web
pages where companies mention “studies performed at different universities”
(without more detail) to support their products. Public research/extension systems
will be required to address the needs of farmers and the whole society in particular
providing assessments concerning the long term or large scale effects on agricul-
tural systems (e.g. soil erosion).
Given the predictions of increase in global population and economic develop-
ment, it is estimated that 70 % more food will have to be produced by 2050 (see
below). The magnitude of this challenge cannot be underestimated given the
current productivity trends of the major crops. Agricultural research will play an
important role in meeting this challenge as it has done in the past, provided that
governments around the world realize the difficulties ahead and invest sufficient
resources to tackle the research issues related to increasing production in a sustain-
able fashion. The associated extension efforts, which will be badly needed, will
increasingly be based on the use of crop simulation models and the development of
decision support systems tailored to the specific needs of the farmers and commu-
nicated through the web.
Box 35.1 Visualizing the Future

A farmer in 2050 is planning to sow wheat by November 1. By October 15 a
sampling robot is sent to the different fields of the farm where it automatically
samples the soil in different locations and produces maps of nutrient
(nitrate, P, K) and soil water content, which the farmer checks against similar
observations obtained 2 weeks ago from a satellite service that he subscribes.
The robot also takes some samples that are packed and sent to the regional
research center to test for soil pathogens or insects. On the same day, his
drone flies over the farm and collects visual and NIR images to map the weed
spots in the fields to be sown. Then the farmer looks at a DSS that shows the
estimated soil water content in the different fields. With that data and the local
rainfall forecast for the next 2 weeks, the system connects with the web sites
of seed companies, collects information on the different cultivars available,
(continued)

runs a simulation model of the crop based on a reliable seasonal weather
forecast and compares which are the best options, considering seed price,
expected yields, product prices and local availability. The farmer buys online
the seed required.
The same DSS builds also a map of recommended application of N, P, K
and contact herbicide and calculates the quantities to be ordered by checking
the actual stocks. The farmer compares online the prices and conditions of
different suppliers and confirms the order. According to weather forecasts dry
conditions are expected by October 22 and 23 with rainfall afterwards. These
are appropriate for applying the N and the herbicide. By October 20 an email
is received from the regional research center advising the use of an insecticide
at a given rate along with the seed. The farmer confirms online the use of the
insecticide which is registered on an external database of pesticide use.
On October 22 the robot fertilizer-sprayer goes to the fields and applies
urea at variable rates depending on need. It also sprays herbicide only on the
spots where weeds had been previously detected. After the job is done the
farmer confirms online the amounts of N and herbicide used in each field.
This information goes to the external databases for subsequent N fertilizer
and pesticide use.
On November 1 the robot seed drill is sent to the fields to sow and to apply
localized P and K fertilizers at variable rates. The planter will follow always
the same path as all other machinery to avoid compaction due to traffic.
35.7 Food Security and Food Safety
Following a sharp increase in food prices in 2008, concerns for food security,
understood as a situation where all humans will have access to sufficient and
nutritious food, have increased around the world. Food security is now high in
the agenda of many countries that are planning for an uncertain future where, at the
same time that global food trade is reaching historical levels, food sovereignty
issues related to the capacity of each country to be self-sufficient in food production
are increasingly important given the current political climate. Food trade is
balancing supply and demand in an effective manner and is the major instrument
now to cope with instability in production caused by extreme weather events and by
changes in food demand due to diet changes or other features of economic
development.
Food safety refers to health issues from the standpoint of ensuring that market-
able foods are both healthy and nutritious. Health-related problems in food pro-
ductions appear periodically (for example, the mad cow disease caused by dubious
animal feeding practices) and attract substantial attention from a society that is
more and more distant from agriculture and food production processes.
Periodically, episodes of food contamination by chemical or biological agents

occur in many countries and generate alarms regarding food safety. One important
source of contamination is the use of untreated waste water for irrigation that still
takes place in many world areas and that must be avoided by appropriate water
treatment. Alarms due to food contamination cause great concern among con-
sumers and this is rightly forcing more control and regulation of food production
processes from farm to fork. Agronomists must ensure that products leaving the
farm are always safe for consumption, the major issue being inadequate pesticide
usage. Another important goal is to enhance the nutritional qualities of the food
produced. Content in terms of protein, essential amino acids, vitamins, and other
nutritional factors must be enhanced where possible by good agronomy. Other
interesting aspect refers to the positive interactions between nutrition and health
of certain agricultural products such as red wine, nuts, and olive oil among others,
that have proven health-related benefits but where the content of the chemical
products responsible for those benefits depends in part on the growing conditions.
Finally, given the increasing importance of gastronomy in affluent societies, agron-
omists should focus more on issues related to ensuring product quality from the
gastronomic viewpoint.
Although predictions vary, it is estimated that agricultural production should
increase by 70 % to meet the demand of nine billion people expected in 2050. Is the
world going to provide food security for all by 2050? First of all it is important to
consider that not all agricultural products are used directly for food. Around 10 % is
devoted to industrial crops including biofuels. The remaining 90 % is shared
between food (65 %) and animal feed (35 %), which results in an overall efficiency
of crop production for food of 0.65. This low efficiency is due to the low conversion
efficiency of animals mainly for meat production. Here, there are ample differences
in efficiencies among animals, chickens being the most efficient and cows the least
(Table 35.2). However, ruminants exploit rangelands (which occupy more land that
is used in agriculture) that otherwise would not be used for food production and this
must be taken into consideration when addressing meat production in global food
assessments. Calls have been made to reduce meat consumption in the developed
Table 35.2 Distribution of uses of edible crops and all crops circa 2012. The efficiency of
conversion for energy is taken 1 for human as direct consumption. Using this Table a general
efficiency of global crop production to food of 0.65 can be estimated as the weighted average of the
efficiencies taken the fraction of use as weighing factors
Edible crops Total crops Efficiency
Fraction used Fraction used Fraction
Humans 0.65 0.5915 1
Pork 0.12 0.1092 0.1
Dairy 0.09 0.0819 0.4
Beef 0.05 0.0455 0.03
Chicken 0.05 0.0455 0.12
Eggs 0.04 0.0364 0.22
countries and this could have an impact on future food security. For instance if feed
for meat production was reduced by half, the overall efficiency would increase from
0.65 to 0.74, a 14 % increase in calories available to humans. Such a drastic change
would be difficult to achieve as it is doubtful that it would free as many calories for
humans as computed above, as animal feed includes residues and other non-food
components, in addition to the consumption of pastures. On the other hand there are
clear health-related advantages of reducing the amount of animal products in
human diets, particularly in countries of high consumption where obesity is a
growing problem.
Another area where improvements will contribute to future food security is
reducing food waste. It is estimated that up to one third of global food production
is wasted before it can be consumed by humans. The nature of waste varies in
different food chains but generally speaking, food waste in poor areas is primarily
due to post-harvest losses caused by pests. By contrast, in the affluent countries the
majority of food losses occur at the consumers’ end of the chain. Although efforts
are being made to reduce waste, much of it is related to social and cultural factors
which, as in the meat consumption patterns, are difficult to change.
How can then agronomy contribute to food security in the future? We must make
current agricultural systems more sustainable without losing sight of the need to
intensify production in existing farmlands. The option of expanding agriculture has
significant ecological limitations and is not going to be sustainable as most of the
best lands have already been put in production. Thus, the sustainable intensification
of production by introducing new techniques adapted to local conditions should
continue that path of increased productivity. Ample opportunities exist around the
world for increasing both agricultural productivity and sustainability by using good
agronomy and appropriate crop management practices.
Bibliography
Fischer, R. A., Byerlee, D., & Edmeades, G. O. (2014). Crop yields and global food security: Will
yield increase continue to feed the world? (ACIAR Monograph No. 158). Canberra: Australian
Centre for International Agricultural Research.
Fresco, L. O. (2016). Hamburgers in Paradise: The stories behind the food we eat. Princeton:
Princeton University Press.
Appendix
Practical Project for an Undergraduate Course on Principles

of Agronomy for Sustainable Agriculture
Teaching the Course
The class of Principles of Agronomy for Sustainable Agriculture (PASA) may

require between 6 and 9 credits according to the European system (1 credit ¼ 25 h
of student’s work), depending on the actual background of students or the specific
interests in some of the topics (e.g. irrigation chapters may be skipped in humid
environments). This would be equivalent to between 60 and 90 classroom hours
with basic lectures and practical work split 50 %. Apart from that, the students are
grouped in couples that undertake a specific practical project that includes both
individual and joint work. Each practical project has a specific location (with its
climate characterized by a specific weather station) and a crop rotation (e.g. barley/
maize) so that each student has a specific crop species assigned.
At the end of the course each group submits a written report on the project results
that is evaluated by the teacher after each student gives an oral presentation (10 min
plus 5 min for questions). At this time the teacher may ask specific questions to
check the ability of the students to support their calculations. The quality of the
presentation and the report and the answers to questions are major determinants of
the final grade of the student.
An example of a proposed practical project used in the PASA course at the
University of Cordoba is shown below.

Agriculture, DOI 10.1007/978-3-319-46116-8
544 Appendix
Practical Project. Principles of Agronomy. 2015/2016

(Revised January 2016)
We will work in groups of two students. Each one will have a different crop and the
two crops will make the crop rotation. Only one report will be presented by each
group.
Weather station...................................................
Year 2015
Crop species.......................................................
Soil of medium texture, 1 m depth
Irrigation is performed by sprinklers with 12 12 m spacing, discharge rate
0.3 l/s, and application efficiency 0.85. Irrigation water has a salinity of
CE ¼ 1.5 dS/m and the water is pumped from a small dam using a pump with
diesel engine
1. Climate. Download the available weather data from the link shown in the web
page in class. Take only complete years starting with 2015 and going backwards
at least to the last 10 years.
Calculate mean monthly values of:
Maximum temperature
Minimum temperature
Solar radiation
Vapor pressure deficit
Wind speed
Total rainfall
Number of rainy days (consider only days with rainfall of 0.5 mm or more)
2. Calculate also for each month the mean values of:
Daylength
Maximum solar radiation (clear sky conditions)
Net radiation over grass
ET0 using the Penman-Monteith-FAO equation
ET0 using the Hargreaves equation
Effective rainfall (FAO method) (use monthly rainfall totals)
3. Productivity. Calculate for 2015 and the two crops:
(a) Thermal time from sowing to harvest. Assume that crop duration is equal to
that defined by the four stages of the FAO method of Kc.
(b) Intercepted PAR: the fraction of intercepted PAR is calculated for each stage
using:
Stage A: f ¼ 0.1
Stages C and D: f ¼ Kc 0.3
Stage B: Interpolate between the values of stages A and C
(c) Potential yield. Compare this value with typical yields for this crop in this
region.
Appendix 545
4. Fertilizer program. Calculate:

(a) Average fertilizer requirements of P and K for the crop rotation. All crop
residues are left in the field. Assume that soil test levels for P and K are
above the threshold levels and below maintenance levels, i.e. we need to
apply fertilizers to compensate for nutrient exports. Assume that the average
yield of your crop is 80 % of the value calculated in 3c.
(b) Calculate the N fertilizer requirements for the two crops.
(c) Calculate the total cost of the fertilizer program (not including the
application cost).
5. Irrigation Schedule
Calculate the irrigation programs for the two crops (dates and amounts) in
2015. Assume that the soil water deficit is zero at sowing.
6. Salinity: calculate the amount of irrigation that should be added to that
calculated in step 5, to achieve maximum yields.
7. Sowing: Calculate the seed rates for the two crops.
8. Frost: Calculate the probability of frost after March 1.
9. Crop calendar and energy requirements.
For the two crops establish the crop calendar. Choose a soil conservation
system (tillage or no tillage) and indicate dates and operations to be performed
(sowing, tillage, application of herbicides, harvest). Calculate the energy
requirements of your farm with the current rotation.
Index
A Arable area, 2, 6, 426, 488, 537

Acclimation, 173, 194, 203, 444, 445, 462 Ard, 237, 529
Acidification, 11, 19, 343, 413, 421 Arthropod, 210, 211
Acidity, 19, 21, 144, 205, 315, 326, 329, 333, Assimilate supply, 183
344, 351 Attainable yield, 6
Active ingredient, 472, 478, 484 Autonomy, 2, 535
Actual yield, 6, 7, 205, 213, 292 Available water, 270, 272, 273, 278, 287
Adaptation, 147, 149, 172, 187, 429, 460, 462,
469, 521, 532, 536, 538
Adiabatic, 56 B
Advection, 59, 61, 82, 122, 447 Bacillus, 350, 467
Aflatoxin, 492 Bean, 22, 38, 89, 124, 135, 136, 152, 155, 156,
Agricultural extension, 11, 538, 539 182, 184, 185, 219, 220, 222, 224, 272,
Agroforestry, 426, 520 298, 299, 305, 307, 318, 319, 344, 345,
Air temperature (Ta), 30, 31, 35, 36, 46, 55–67, 362, 364, 365, 374, 376, 394, 420, 421,
79, 114, 121, 122, 199, 200, 209, 226, 495, 497, 499, 517, 519–521
349, 431, 435–438, 443, 447, 456, Biofuel, 537, 541
493, 531 Biological control, 468, 536
Albedo, 34–36, 80, 81, 84, 432 Biomass production, 152, 165, 166, 170, 177,
Alfalfa, 89, 124, 127, 134, 157, 182, 187, 222, 179, 181, 185, 187, 200, 282, 284, 292,
249, 271, 272, 298, 299, 317, 334, 335, 355, 371, 378, 392, 535
344–346, 363, 390, 394, 447, 498 Black body, 28, 35
Alkaline soil, 20, 349, 421 Boron, 312, 322, 335, 398, 414, 420–422
Allelopathy, 463 Buffer zone, 477
Allowable depletion, 270, 272–273, 278 Buoyancy, 47, 60
Aluminum toxicity, 208, 315 Burning, 230, 235, 455, 456, 459, 467, 491,
Ammonia, 87, 110, 325, 326, 328–330, 332, 528, 529, 531
336, 337, 342, 343, 351, 353, 371
Animal power, 230
Anoxia, 235 C
Anthesis, 147, 183, 186, 197, 209, 220 C3, 170–174, 176–179, 183, 185, 186, 200,
Anthropocory, 460 201, 293, 355, 461, 531, 532
Apical dominance, 149 C4, 170–174, 177, 178, 183, 185, 186, 200,
Application efficiency, 131, 259, 262 201, 293, 322, 355, 460, 461, 463,
Aquaporin, 193 532, 537

Agriculture, DOI 10.1007/978-3-319-46116-8
548 Index
Calcium, 16, 20, 21, 25, 311–313, 315, 322, Crop cover, 82
324, 326–334, 383, 384, 387, 389, 390, Crop growth rate (CGR), 160, 179, 426
397–399, 410–414, 422 Crop management, 120, 342, 344, 356, 370,
Canopy architecture, 186, 197 467, 504, 509–510, 513, 514, 522,
Canopy photosynthesis, 108, 175, 177, 531, 542
178, 200 Crop photosynthesis, 152, 175–178, 180,
Capillary rise, 109 186, 187
Carbohydrates, 174, 181, 184, 187, 200, 206, Crop residue (CR), 2, 7, 23, 89, 102, 111, 184,
433, 463 234, 246, 248, 249, 336, 371, 432, 433,
Carbon sequestration, 532 467, 488, 489, 518, 523, 532, 537
Cardinal temperatures, 146 Crop rotation, 247, 248, 258, 281, 376, 467,
Catch crop, 359, 522 468, 479, 516, 521–523, 525, 533
Cation exchange capacity (CEC), 21, 22, 316, Crop temperature, 55, 59, 425, 431, 436–437
351, 353, 354, 388, 390 Cultivation, 1, 3, 6, 22, 117, 125, 137, 149, 171,
Cattle, 7, 351, 421, 529 189, 221, 229, 250, 252, 255, 258, 265,
Center pivot, 117, 264, 274, 510, 512 301, 348, 372, 516, 524, 528
Chelate, 324, 333–335, 413, 420 Curve number, 101–103, 359
Chemical composition, 326
Chilling injury, 173, 444
Clay mineral, 17, 18, 384, 387, 390 D
Climacteric fruit, 496 Dairy cattle, 338, 541
Climate change, 429–430, 531–532 DAP. See Di-ammonium phosphate (DAP)
Cloudiness, 30, 32, 36, 121, 130, 437, 508 Dark respiration, 172, 175
Compensation point, 172, 174 Daylength, 31, 32, 35, 40, 438
Competition, 149, 159–167, 221, 231, 251, Decision making, 8–12, 525
252, 256, 411, 429, 430, 463, 517, Decision support, 514, 539
534, 537 Decomposers, 7
Conservation agriculture (CA), 246–248, 532 Deep percolation (DP), 97–101, 109, 131, 259,
Conservation tillage, 242, 246–250, 252 267, 270, 278, 283, 358, 522
Continuous cropping, 528 Deficit irrigation (DI), 183, 275, 281–294, 510
Conventional tillage, 230, 234–235, 246, 247, Deforestation, 529, 531, 537
249, 251 Denitrification, 210, 235, 336, 343, 350–352,
Copper, 20, 322, 335, 398, 414, 420–422 370, 371, 373, 374
Corn, 46, 48, 134, 194, 220, 244, 245, 264, 275, Deposition, 60, 66, 235, 337, 348, 371, 373,
361, 365, 388, 390, 446, 467, 474, 519 430, 447, 448
Corn Belt, 516 Desalination, 86, 88
Cosine law, 28, 31 Determinate crops, 183
Cotton, 22, 31, 36, 38, 82, 89, 124, 126, 137, Devernalization, 148
146, 156, 162, 173, 180, 182, 183, 195, Dew formation, 60, 430
199, 200, 210–213, 219, 220, 222, 224, Dew point, 57, 58, 441, 447
232, 236, 250, 253, 265, 272, 288, 298, Di-ammonium phosphate (DAP), 328, 332
299, 302, 319, 345, 362, 383, 394, 446, Diffuse radiation, 30, 178, 179, 431
467, 491, 492, 498, 514, 521, 535 Diffusion, 44, 71, 170, 199, 351, 393, 408, 479
Coupling, 146, 198–201 Disease, 4, 6, 8, 9, 11, 23, 24, 118, 149, 162,
Cover crop, 110, 125, 127, 137, 244, 248, 251, 167, 172, 181, 191, 199, 205, 210–213,
252, 379, 387, 433, 454, 455, 520, 536 221, 235, 246, 248, 258, 265, 355, 430,
Crassulacean acid metabolism (CAM), 433, 436, 441, 467, 468, 474–476,
171, 172 479–483, 488, 490, 492, 518, 521–523,
Critical concentration, 383 525, 532, 540
Critical soil water deficit, 270, 273–277, 511 Domestication, 527–529
Crop area, 132 Dominance, 2, 4, 16, 21, 149, 152, 313, 414,
Crop coefficient, 120, 122–129, 133–134, 178, 523, 533, 534
179, 279, 438, 511 Dormancy, 460–462, 528
Index 549
Drift, 426, 427, 429, 473–477, 483, 484 Flowering, 82, 142–144, 147–149, 183, 186,
Drip irrigation, 88, 265, 279, 301, 302, 198, 199, 271, 284–286, 288, 363, 364,
407, 408 383, 438, 445–447, 463, 473, 494
Drone, 508, 539 Flux density, 28, 29, 37, 73
Droplet diameter, 455, 473, 476, 483, 484 Food demand, 534, 540
Food supply, 187
Forage crop, 38, 89, 127, 134, 157, 244, 249,
E 299, 317, 363, 446, 489, 494–495,
Economic injury level, 464 498, 517
Ecosystem, 2, 3, 15, 16, 55, 79, 91, 107, 145, Forecasting, 479, 534, 539, 540
170, 211, 229, 231, 382, 515, 533 Formulation, 113, 330, 394, 472–476, 479
El Ni~
no-Southern Oscillation, 530 Fossil fuel, 237, 246, 531, 537
Electrical conductivity (EC), 22, 296, 297, Frost, 56, 147, 148, 157, 181, 206, 220, 258,
299–302, 304–309, 311, 314, 358, 411, 264, 265, 426, 430, 443–456
415, 418, 511 Frost risk, 147, 430, 444, 451, 454
Elemental composition, 20, 327 Frost, black, 448
Embodied energy, 87, 88 Frost, white, 447, 448
Emissivity, 28, 35 Fuel, 85, 86, 88, 237, 246, 330, 455, 456, 489,
Energy balance, 69, 75, 79–90, 109, 113, 171, 531, 537, 541
232, 452 Fuel requirement, 85, 86, 237
Energy content, 87, 89, 533, 538 Fumigants, 466, 473, 475, 476, 482
Energy efficiency, 7, 171 Fungi, 210–212, 218, 248, 346, 347, 350, 466,
Energy use, 171, 238 467, 471, 472, 479, 483, 492, 518, 523
Environmental impact, 3, 342, 393, 503, Fungicide, 210, 218, 334, 471–473, 476, 477,
535, 536 479–484, 522, 523
Equilibrium evaporation, 115, 440 Furrow, 225, 248, 253, 261, 262, 265, 301–303,
Equilibrium Moisture Content (EMC), 431, 529
493, 494 Fusion, 456
Erosion, 11, 17–19, 24, 230, 237, 241–246,
248–252, 262, 264, 311, 382, 399, 426,
429, 432, 520, 522, 529, 532, 539 G
Es, 97, 108, 110–112, 126, 203, 232, 292–294 GE. See Genetic engineering (GE)
Essential nutrients, 20, 322, 355, 382–384 Genetic engineering (GE), 497–499, 501, 535
ET0. See Reference ET (ET0) Genetically modified organisms (GMO),
Ethanol, 87, 537 467, 535
Eutrophication, 382, 397 Genotype, 2, 148, 149, 160, 166, 195, 462, 467,
Evaporation, soil, 76, 111–113, 122, 126, 127, 468, 479, 480, 513, 514
131, 202, 203, 232, 252, 265, 296, 433, Germination, 15, 69, 143, 149, 212, 217, 218,
455, 535 221, 225, 226, 229, 258, 259, 264, 265,
Evapotranspiration (ET), 59, 60, 105, 107–132, 337, 431, 433, 446, 447, 460–462, 466,
153, 202, 259, 273, 430, 438, 510, 511 467, 469, 494
Exponential growth, 44, 175, 271, 355 Germplasm, 2
Extinction coefficient, 37, 38, 175, 176 Global Circulation Model, 531
Global Positioning System (GPS), 252,
510, 512
F Global radiation, 30, 32, 34, 36, 116, 129, 171,
Farming systems, 2–6, 339, 488, 515–525 185, 490
Fat, 174, 184, 288, 497–499, 501 Grain legume, 149, 180, 184, 521
Feedback, 163, 187, 242 Green manure, 23, 336, 339, 522, 536
Field capacity (FC), 71, 72, 93, 97–100, 103, Green Revolution, 515, 538
184, 197, 230, 234, 237, 270, 296, 297, Greenhouse, 49, 80, 115, 128–129, 171, 179,
305, 346, 360, 408 226, 246, 265, 266, 342, 430, 439–441,
Flooded soil, 262, 286, 467 474, 475, 532, 537
550 Index
Greenhouse effect, 342 Insecticide, 10, 218, 225, 471–473, 475–477,

Greenhouse gas, 246, 532, 537 479, 482, 484, 522, 540
Gross energy, 89, 496 Intensive cropping systems, 515, 523, 524
Growth duration, 379 Ion exchange, 16, 21, 316, 331, 351, 389, 390,
Growth respiration (Rg), 174 397, 511
Gypsum, 22, 312, 313, 315, 316, 333 Iron, 20, 322, 329, 331, 334, 335, 343, 382,
384, 398, 411, 413, 414, 420–422
Irradiance, 37, 81, 171–173, 175, 176, 431, 440
H Irrigation adequacy, 260
Hardening, 225, 445 Irrigation efficiency (IE), 259, 260
Harrow, 225, 234, 235, 238, 248, 251, 466 Irrigation scheduling, 97, 123, 132, 269,
Harvest, 9, 85, 124, 143, 160, 194, 206, 218, 359, 511
234, 246, 265, 273, 282, 344, 371, 394, Irrigation uniformity, 261, 283
430, 460, 477, 487, 516, 528
Harvest index (HI), 160, 164, 180–184,
202, 203, 209, 210, 220, 282, 284, L
287, 292, 344, 355, 371, 374, 378, 430, Labor, 3, 6–8, 87–89, 238, 274, 278, 408, 488,
490, 534 491, 511, 516, 517, 524, 529
Hay, 103, 124, 134, 157, 272, 298, 317, 345, Ladybug, 468
363, 364, 394, 492, 494, 495, 498 Land equivalent ratio (LER), 518, 519
Hedgerow, 50, 265, 484 Lapse rate, 56
Herbicide, 224, 225, 230, 231, 238, 246, 248, Latent heat, 43, 66–67, 82, 109, 110, 113,
250, 251, 455, 460, 462, 465, 467, 469, 434–438, 452
471–485, 519, 523, 535, 540 Leaching, 24, 25, 259, 266, 275, 278, 296, 301,
Herbicide resistance, 467, 535 303–311, 314–316, 329, 330, 335, 336,
HI. See Harvest Index (HI) 338, 349, 357–360, 370, 371, 373, 382,
Horizontal resistance, 468 399, 413, 418, 477, 518, 522
Horticultural crops, 135, 148, 153, 167, 220, Leaching fraction (LF), 303–310, 418
226, 273, 298, 318, 364, 433, 444, Leaching Index (LI), 359, 360
490, 498 Leaching requirement (LR), 259, 306,
Humidity, absolute, 58, 110 308–311, 418
Humidity, relative, 57, 58, 61, 132, 133, 326, Leaf area index (LAI), 37, 117, 123, 127, 153,
435, 437, 479, 493, 494 175–178, 193, 196, 197, 206, 208, 236,
Hydraulic conductivity, 19, 95, 98, 110, 126, 468, 476, 483, 484, 507
191, 193, 194, 247, 267, 297, 311, 312, Leghemoglobin, 344
358, 444 Lepidoptera, 211, 467, 482
Hydrological group, 103 Lime, 242, 333, 337, 349
Hygroscopicity, 326, 329, 330 Lime requirement, 333
Limiting factor, 3, 6, 91, 160, 226, 232, 355,
357, 372, 522
I Lipid, 193, 218, 480
Imbibition, 148, 218, 221 Lodging, 149, 164, 167, 181, 426, 491
Immobilization, 25, 248, 337, 346–349, Long wave radiation, 35, 60, 80, 85, 121, 432,
384, 477 446, 452, 453, 455
Impact density, 476, 483, 484 Low energy precision application (LEPA), 264
Imposed evaporation, 115 Lv, 153, 193
Indeterminate, 181–183, 286, 355, 510 Lysimeter, 108
Infiltration, 16, 18, 19, 22, 98, 99, 101, 102,
231, 232, 234, 235, 242, 246, 247, 251,
252, 257, 258, 261–263, 265, 311–313, M
315, 433, 510 Machinery, 7, 10, 85, 89, 167, 188, 207, 220,
Infiltration rate, 16, 18, 19, 22, 98, 101, 102, 225, 229, 231, 233, 237, 246, 329, 399,
231, 232, 234, 261, 263, 311 430, 466, 488, 524, 529, 532, 540
Index 551
Magnesium, 20, 21, 25, 311, 313, 322, Nitrobacter, 349

327–329, 333, 334, 383, 390, 397, 398, Nitrogen (N) fixation, 431
412–414, 421, 422 Nitrogen cycle, 342
Maintenance respiration (Rm), 174, 175 Nitrogen fertilizer, 328, 342, 349
Manganese, 20, 322, 335, 398, 413, 414, Nitrogen mineralization, 345, 347, 373, 377
420–422 Nitrogen supply, 183, 199, 206, 207, 335, 349,
Manure, 23, 335–339, 350–352, 371, 373, 399, 355, 370, 373, 374, 376, 382, 391, 392,
420, 522, 536 520, 522, 523
Mechanization, 8, 241, 488, 529 Nitrogenase, 343, 344
Meristem, 142, 149, 162, 478, 480, 494 Nitrous oxide (N2O), 35, 343, 350, 351
Methane (CH4), 35 No till, 232, 238, 244, 246, 251, 352, 354
Microclimate, 108, 247, 427, 429–430 Nursery, 143, 226, 338, 434
Micronutrient, 322, 324, 327, 334–336, 389, Nut crops, 496
398, 410, 411, 413, 419–422 Nutrient cycling, 15, 248, 518, 529
Microsprayer, 265 Nutrient extraction, 388, 389, 532
Microsprinkler, 265, 510
Mineralization, 22–25, 335, 336, 338, 343,
345–349, 357, 359, 371–373, 377, 378, O
385, 431, 518, 522 Oilseed, 38, 170, 181, 184, 186, 521
Minimum tillage, 248, 251, 532 Olive, 10, 41, 50, 60, 67, 73–75, 84, 85, 112,
Mixed cropping, 516, 517 115–117, 124, 127, 128, 135, 143, 144,
Mobilization, 331, 334 158, 182, 185, 201, 226, 251, 252, 272,
Model, 4, 6, 38, 74, 99, 112, 116, 117, 123, 127, 291, 298, 299, 320, 335, 345, 366, 380,
128, 132–134, 144, 146, 178, 179, 383, 394, 398, 420, 421, 438, 446, 447,
201–204, 232, 252, 291, 293, 297, 314, 489, 496, 500, 513, 520, 541
355, 370, 373, 394, 395, 436–438, 441, Onions, 136, 154, 167, 182, 223, 224, 273,
452–453, 469, 508, 513–514, 531, 299, 319, 365, 465, 489, 495,
539, 540 501, 520
Moldboard plow, 229, 231, 234, 244, 245, 529 Optimization, 282, 291–294, 531
Molybdenum, 20, 322, 335, 398, 420–422 Opuntia, 468
Mono-ammonium phosphate (MAP), 328, 332, Organic agriculture, 536
412, 415–418 Organic farming, 8, 324, 465, 517, 536–537
Monoculture, 90, 248, 374, 516, 521 Orthodox seeds, 494
Mortality, 3, 159, 160, 163 Osmotic adjustment, 194, 313
Mowing, 251, 467, 489, 494, 495 Osmotic potential, 94, 190, 296, 297
Mulch, 80, 247, 248, 251, 432–433, 439, 454,
467, 521
Multiline, 468 P
Multiple cropping, 516 Partitioning, 63, 79, 81, 85, 164, 181–183, 188,
Mycotoxins, 492 290, 455
Pasture, 1, 4, 7, 38, 103, 127, 135, 157, 159,
187, 244, 391, 421, 463, 494, 516, 520,
N 524, 529, 542
N balance, 342–368, 371–376, 414 Penman, 113–116, 118, 120–122, 435, 436
N fixation, 343–347, 350, 517 PEP carboxylase, 172
Net primary production, 7 Percolation Index (PI), 359, 360
Net radiation, 31, 36, 43, 76, 80–82, 85, 109, Persistence, 4, 460–462, 465, 472, 477,
122, 133, 171, 429, 432, 436–440, 452 479, 481
Nitrate reduction, 343 Pest, 6, 8, 9, 11, 24, 149, 167, 205, 210–213,
Nitrification, 20, 326, 329, 330, 336, 343, 347, 218, 221, 246, 248, 426, 427, 429, 440,
349–351, 359, 415, 416 460, 464–469, 473–475, 482, 490, 492,
Nitrification inhibitors, 330, 350 517, 518, 521–523, 525, 532, 536, 542
Nitrite, 343 Pesticide drift, 473
552 Index
Pesticides, 3, 5, 7, 8, 79, 86, 87, 90, 218, 220, R

225, 238, 258, 266, 274, 427, 460, Radiation balance, 27–41, 80
471–477, 483, 484, 491, 510, 515, 529, Radiation use efficiency, 152, 170–188,
535, 536, 540, 541 206–208, 210, 211, 355
pH, 16, 19–22, 208, 312, 313, 315, 326, 329, Rapeseed, 89, 124, 137, 182, 219, 220, 224,
333–336, 344, 349–351, 353, 384, 385, 248, 272, 298, 320, 345, 362, 394, 492,
388–390, 398, 411, 413–416, 418, 420, 494, 495, 498, 521
421, 495 Recalcitrant seeds, 494
Phenology, 37, 143–148, 150, 152, 445, 448 Redox potential, 20
Phenophase, 144, 145, 152 Reference ET (ET0), 110–113, 115, 116,
Phenostage, 144, 146 120–123, 126, 129, 130, 132–134, 201,
Phenotype, 142 203, 273, 279, 289, 379, 380, 438, 531
Photoperiod, 142–144, 147–149, 152, 531 Reference evapotranspiration, 129
Photorespiration, 171 Regulated deficit irrigation (RDI), 289, 290
Photosynthesis type, 461 Relative humidity, 57, 58, 61, 132, 133, 326,
Photosynthetically Active Radiation (PAR), 435, 437, 479, 493, 494
29, 30, 37, 38, 41, 152, 153, 173, Remote sensing, 39, 504–509, 512
178–180, 183–186, 440 Renovation rate (RR), 49, 440, 441
Phyllochron, 150 Reproduction, 3, 147, 149, 463
Phytochrome, 149, 163, 461 Return flows, 267, 410, 533
Plant available water (PAW), 270, Rhizobium, 343, 344
272–274, 278 Ridge tillage, 250
Plant breeding, 142, 170, 534–536 Roller, 225, 230, 238
Plant density, 159–167, 463, 517 Root depth, 134–137, 218, 271–273
Plastic, black, 81, 432, 433 Root length density, 153, 191, 193, 408
Plastic, transparent, 432, 433, 455, 466 Root nodules, 343, 344
Plasticity, 142, 160, 166 Roughness, 45, 46, 48, 66, 101, 120, 232, 247,
Plowing, 229, 231, 234–238, 243–246, 430, 436, 455
352, 529 Rubisco, 171, 172, 206
Pollination, 183, 209, 285, 286, 437, 445, 462 RUE. See Radiation use efficiency (RUE)
Pollution, 3, 8, 230, 342, 358, 373, 374, 382, Runoff, 97, 98, 101–104, 109, 131, 212, 242,
455, 533, 536 251, 253, 259, 262–265, 267, 270, 283,
Population, 2, 3, 6, 159–161, 163, 212, 311, 399, 510
256, 348, 384, 427, 460, 462, 464–467,
469, 488, 517, 522, 523, 528–530,
538, 539 S
Potassium, 20, 21, 25, 87, 88, 225, 250, Saline soil, 22, 297, 313–315, 511
322, 323, 327, 328, 331–338, 380, Salinity, 22, 150, 152, 205, 206, 258, 267,
382–404, 408, 411–418, 420, 423, 424, 296–312, 525, 528, 533, 535
539, 540 Salinization, 11, 296, 301, 304, 313, 426,
Potential production, 6, 180–182 532, 533
Potential yield, 6, 184, 204 Satellite, 504, 508, 509, 512, 539
Primary production, 4, 7, 537 Seed bank, 231, 234, 461, 462, 466
Primary tillage, 87, 521 Seed bed, 225, 467
Production value, 200, 282 Seed dormancy, 461, 462
Protein, 172–174, 184, 186, 193, 198, 206, 288, Seed drill, 220, 225, 230, 540
342, 343, 347, 355, 445, 478–480, 483, Seed longevity, 494
497–499, 501, 541 Seed treatment, 218, 474
Seed viability, 218, 221
Seeding rate, 220–224, 510, 519
Q Seedling establishment, 259, 344
Q10, 347 Selection, 148, 186, 220, 258, 332, 334, 444,
Quantum efficiency, 173 453–454, 462, 467, 468, 479
Index 553
Sensible heat, 43, 55, 59, 61, 62, 65–66, 80–82, 319, 345, 363, 390, 394, 447, 467, 495,
84, 109, 110, 113, 122, 440, 452 497, 519–521, 535
Shelterbelt, 425, 427 Spatial variability, 167, 313, 465, 511
Silage, 80, 361, 363, 446, 494, 495, 498, 519 Specific heat, 56, 66, 70, 72, 80, 437, 440,
Slow release fertilizer, 328, 330, 335–336, 359 453, 456
Sodium adsorption ratio (SAR), 311, 312, 316 Spore, 79, 212, 479, 483, 494
Soil acidity, 21, 315, 333, 344 Spray, 87, 257, 263, 264, 322, 325, 330,
Soil aeration, 19, 21, 22, 349, 414, 461, 332–334, 427, 455, 472–477, 479,
490, 494 484–485, 540
Soil aggregate, 19, 21, 45, 231, 233 Spring wheat, 519
Soil bulk density, 70, 92, 207, 236, 316, 396 Sprinkler irrigation, 88, 117, 257–259, 261,
Soil evaporation, 76, 111–113, 122, 126, 127, 263–265, 274, 278, 294, 298, 300, 302,
131, 202, 203, 232, 252, 265, 296, 433, 303, 312–316, 407, 427, 456, 467, 512
455, 535 Squash, 136, 154, 155, 299, 319, 365, 499,
Soil fertility, 16, 20, 24, 198, 204, 248, 516, 519, 528
528, 529, 533, 536 Stability, 4, 18–21, 56, 60, 66, 242, 246, 256,
Soil formation, 16–18, 532 311, 335, 413, 426, 428, 515, 516
Soil horizon, 429 Stage 1, 110–113, 126, 247, 351
Soil management, 18, 19, 23, 24, 242, 243, 250, Stage 2, 111, 112, 126, 232, 349, 492
251, 300, 359, 391, 454–455 Steady state, 57, 300, 310, 373
Soil microorganism, 23, 79, 342, 384, 385 Stefan-Boltzmann, 28, 35
Soil mineral, 72, 348, 384 Stomata, 67, 108, 115, 127, 170, 187, 189, 190,
Soil organic matter (SOM), 20, 22, 23, 89, 241, 193–199, 202, 203, 207, 211, 384, 437,
243, 251, 336, 372, 387, 537 531, 535
Soil porosity, 19, 231, 232, 234 Strategy, 4, 5, 9, 10, 20, 143, 149, 183, 195,
Soil structure, 15, 18, 19, 218, 221, 229, 242, 220, 231, 275–278, 285–291, 293, 314,
246, 297, 311–313, 336 333, 334, 351, 371, 372, 374–377, 384,
Soil temperature, 62, 69–77, 79, 84, 85, 226, 393–397, 399, 400, 460, 469, 512, 514,
347, 349, 351, 431–434, 462 524, 530
Soil texture, 16, 18, 19, 74, 94, 232, 242–244, Strip cropping, 251, 517
258, 372, 395, 511 Strip tillage, 250
Soil type, 17, 72, 76, 101, 110, 234, 246, 252, Stubble, 83, 84, 122, 133, 230, 235, 241, 248,
257, 265, 266, 273, 351, 352, 387 467, 494, 525
Soil water deficit (SWD), 97, 105, 259, 270, Subirrigation, 257
273–278, 286, 297, 511 Subsistence agriculture, 7, 8, 491
Soil water deficit, critical (SWDc), 270, Subsoiler, 234
273–277, 511 Subsurface drip, 265
Soil-borne disease, 467 Subsurface drip irrigation (SDI), 265, 266
Solar constant, 28–30 Sulfur, 315, 322, 327, 330, 334, 336, 422, 471,
Solar radiation, 3, 6, 27–36, 38, 40, 58, 64, 71, 479, 482
83, 85, 116, 117, 120, 121, 127, 129, Sunflower, 38, 89, 104, 125, 137, 146, 147,
171, 172, 177–179, 184, 429, 437, 440, 150, 156, 161, 163, 164, 166, 173,
441, 504 181–183, 186, 188, 193, 195, 197, 201,
Solarization, 466, 467 206, 207, 218, 219, 222, 224, 231, 235,
Solid set, 264, 265, 278 248, 271, 272, 288, 298, 299, 320, 335,
Solid set sprinkler, 264 345, 356, 357, 362, 394, 447, 492, 493,
Sorghum, 38, 89, 124, 134, 148, 157, 171, 182, 495, 498, 513, 521, 524
194, 219, 223, 272, 276, 277, 287, 288, Superphosphate, 328, 331, 333, 334
298, 299, 301, 303, 317, 345, 361, 394, Surface irrigation, 88, 256, 257, 261–263, 274,
461, 463, 495, 497, 498, 517, 521, 523 300, 312, 314, 456
Soybean, 38, 89, 109, 124, 136, 144, 146, 148, Sustainability, 4, 8, 9, 241, 296, 376, 384, 469,
156, 164, 182, 188, 195, 212, 213, 219, 515, 520, 523, 527–542
220, 222, 244, 245, 272, 286, 298, 299, Sustainable agriculture, 8, 18
554 Index
Swathing, 492 U
Symbiosis, 342, 344 Uncertainty, 8, 9, 12, 252, 291, 351, 393
Systemic, 472, 476, 478, 483, 484 Universal soil loss equation (USLE), 242–245
Urea, 87, 324, 328, 330, 334–336, 347, 351,
353, 412, 413, 422–424, 480, 540
T Urea formaldehyde (UF), 328, 336
Ta. See Air temperature (Ta)
Tactics, 10
Technology transfer, 11–12, 514 V
Temperature, base, 146, 148, 150, 218–220 Vapor pressure, 35, 36, 57, 58, 60–67, 109,
Temperature, dewpoint, 57, 58, 441 114, 122, 128, 190, 191, 195, 199, 200,
Thermal admittance, 74, 80, 453–455 220, 430, 440, 496, 531
Thermal blankets, 455 Vapor pressure deficit (VPD), 57, 58, 61, 62,
Thermal conductivity, 70, 71, 453 114–117, 122, 127, 195, 200–203, 220,
Thermal diffusivity, 71, 232 293, 380, 435, 496, 531, 535
Thermal radiation, 27, 189, 455, 456 Vaporization, 476
Thermal time (TT), 146, 149, 203, 218, Variability, 9, 51, 148, 161–163, 167, 181, 297,
219, 271 313, 377, 454, 462, 465, 469, 510, 511,
Thickener, 476 521, 530, 531
Tillage implements, 232 Variable rate irrigation (VRI), 264, 510, 511
Tillage, primary, 87, 521 Vegetable crops, 265, 287, 299, 433, 444
Tillage, secondary, 87, 234, 235, 461 Vernalization, 148, 220, 521
Tillering, 149, 162, 164, 197, 206, 207, 285, Vertical resistance, 468
383, 445, 447 Vivianite, 334
Tolerance, 181, 243, 245, 297, 298, 384, 444, Volatilization, 329, 330, 332, 337, 351, 353,
468, 521 371, 373, 374, 476, 477
Topography, 101, 248, 257, 258, 264, 266, Volume application rate (VAR), 473, 483–485
454, 511 VPD. See Vapor pressure deficit (VPD)
Total dissolved solids (TDS), 415 VRI. See Variable rate irrigation (VRI)
Toxicity, 20, 22, 208, 297, 298, 300, 312, 313,
315, 316, 320, 334, 414, 421, 466, 472,
478, 481, 482 W
Toxin, 212, 463, 467 Water content, 18–20, 35, 70–72, 74, 75, 80,
Tractor, 86, 87, 230, 232, 237, 238, 241, 89, 92–95, 97–101, 103, 108–110, 122,
252, 377 127, 131, 132, 162, 184, 191, 192, 197,
Transgenic, 535, 536 204, 207, 210, 218, 219, 221, 224, 225,
Translocation, 183, 355, 377, 472, 478, 481 230–237, 246, 252, 257, 269, 270, 297,
Transplant, 226–227 302, 305, 314, 315, 332, 337, 346–351,
Transport, 17, 25, 43–53, 60, 66, 67, 79, 82, 359, 360, 374, 378, 391, 408, 453, 461,
86–88, 110, 122, 172, 174, 181, 190, 478, 488, 490–496, 507, 511, 522, 539
194, 198, 211, 218, 232, 242, 334, 343, Water holding capacity, 275, 510
382, 384, 398, 410, 432, 447, 460, 473, Water potential, 19, 22, 92–95, 98, 189–191,
478, 482, 488–490 194, 195, 269, 273, 297, 302, 303, 445,
Transportation, 530, 532, 537 492–494
Tree-Row-Volume (TRV), 484 Water supply, 67, 131, 172, 183, 191, 194, 203,
Trophic chain, 2, 3, 7, 479, 528 207, 221, 224, 225, 257, 258, 275, 282,
Tuber crop, 87, 89, 136, 155, 164, 174, 187, 286, 288, 291, 309, 357, 391, 456, 530
231, 287, 319, 367, 394, 461, 463, 491, Water table, 86, 94, 97, 102, 109, 257, 258,
495, 501 266, 267, 296, 301, 534
Turbulence, 44, 48–49, 59, 60, 66, 79, 80, 128, Water use, 3, 56, 172, 183, 190, 195, 202, 203,
427–428, 430, 437, 477 207, 220, 256, 259, 293, 379, 430, 435,
Turbulent transport, 43–53, 60 436, 533–535
Index 555
Waterlogging, 210, 218, 235, 246, 311, Winter wheat, 148, 157, 224, 272, 298,
347, 349 345, 394
Water-use efficiency (WUE), 195, 200–203,
220, 293, 379, 430, 535
Weathering, 296, 304, 387 Y
Weed control, 188, 212, 231, 232, 234, 246, Yield component, 147
248, 250, 432, 433, 459, 464–467, 469, Yield gap, 6, 7
474–476, 529 Yield loss, 148, 211–213, 235, 253, 289, 291,
Wet bulb, 408 297, 393, 463, 464, 490–491, 510
Wilting point, 71, 72, 93, 120, 193, 197, 270, Yield stability, 515
272, 297, 453 Yield variability, 8, 162, 167
Wind, 17, 44, 53, 75–76, 80, 110, 164, 179, Yield, dry, 374
205, 209, 242, 258, 282, 296, 425, 446, Yield, fresh, 180
460, 473, 520
Wind rose, 50, 53
Windbreak, 80, 425–430, 434, 439, 454 Z
Windbreak effectiveness, 427, 428 Zenith angle, 28, 30, 37, 39, 431
Windbreak porosity, 427, 428 Zero till, 247
Windbreak, scattered, 430 Zinc, 20, 322, 335, 382, 398, 413, 420–422

Villalobos Fereres 2016 Copy Corrected 130222

Uploaded by

Copyright:

Available Formats

Villalobos Fereres 2016 Copy Corrected 130222

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Villalobos Fereres 2016 Copy Corrected 130222

Uploaded by

Copyright:

Available Formats

Francisco J.

Villalobos · Elias Fereres

Principles of Agronomy for

ISBN 978-3-319-46115-1 ISBN 978-3-319-46116-8 (eBook)

Library of Congress Control Number: 2016958826

© Springer International Publishing AG 2016

Printed on acid-free paper

This Springer imprint is published by Springer Nature

Cordoba, Spain Francisco J. Villalobos

1 Agriculture and Agricultural Systems . . . . . . . . . . . . . . . . . . . . . . . 1

Part I The Crop Environment

10 Calculation of Evapotranspiration and Crop Water

Part II Crop Productivity

Part III Crop Management

24 Nitrogen Fertilization I: The Nitrogen Balance . . . . . . . . . . . . . . . . 341

Elias Fereres and Francisco J. Villalobos

E. Fereres (*) • F.J. Villalobos

© Springer International Publishing AG 2016 1

by the disappearance of arable land due mostly to urbanization and to irreversible

Box 1.1 Ecology and Ecosystems

1.2 Characteristics of Agricultural Systems

more correct to speak of the productivity of water or nutrients, expressed as kg/m3

1.3 Management of Agricultural Systems

Example 1.1 The response of a wheat crop to N fertilizer in a rainfed Medi-

on economic criteria. Nevertheless, there are many facets to the management of

1.4 Types of Agricultural Systems

Agricultural systems can be classiﬁed according to various criteria. An ecologically

1.5 Decision Making in Agriculture

Example 1.2 The result of two different nitrogen fertilization strategies

1.6 Sources of Information for Decision Making

1.6.1 Farmer’s Experience

it presents difﬁculties for adopting innovations or to adapt to new situations (crop or

1.6.2 Research, Experimentation and Technology Transfer

1.6.3 Commercial Information

Antonio Delgado and José A. Gómez

© Springer International Publishing AG 2016 15

2.2 Dynamics of Soil Formation and Soil Loss

SOIL FORMATION PROCESSES

Oxides or alumina & iron

Silt Fine parent

Fig. 2.2 Description of key processes in soil formation

2.3 Soil Physical Properties: Texture and Structure

Soil physical properties Soil processes & management

• Hydraulic conductivity • Infiltration, runoff, waterlogging

2.4 Soil Chemical Properties

2.4.2 Redox Status

2.4.3 Ion Retention in Soils

Ions can be retained in soils by precipitation and adsorption processes. Precipita-

2.4.3.1 Exchange Capacity

2.4.4 Salinity and Sodicity

Salinity is deﬁned as a high concentration of soluble salts (more soluble than

2.5 Soil Biological Properties

Roots Nematodes 1 Arthropods 1 Arthropods 2

Fungi Nematodes 2 Nematodes 3

Fig. 2.4 Soil food web

Table 2.2 Some soil biological properties

Understanding soil biological properties is important for soil management but

2.6 Nutrient Cycles and Balances in the Soil

Desorption Labile pool

Chemical (e.g. adsorption/desorption or precipitation/solubilization) and biological

hs ¼ arccos ½tg42:9 tgð23:45Þ ¼ 66:23∘