2.

Climate change

As we have seen, primary food production exerts many different kinds

of environmental impact, from land-use change, through the
consumption of resource inputs, to the generation of agricultural wastes.
Most of the activities associated with primary food production are also a
source of greenhouse gas emissions. This section reviews the challenge
that climate change poses for food production, then goes on to examine
the contribution that agriculture makes to global warming. It is
important to recognise that all components of the agri-food system, from
the agri-technologies supply industry, through primary production,
processing, manufacturing, distribution and retail, right down to the
domestic arena of the storage, cooking and disposal of food, are
responsible for the emission of greenhouse gases. One estimate for the
entire EU food supply chain suggests that what we eat has more impact
on climate change than any other aspect of daily life, accounting for
31 per cent of the global warming potential of products consumed
within the then EU-15 (Tukker et al. 2006). This section concentrates
on agriculture’s contribution to climate change and how it will be
affected by it; chapter five looks in more detail at the stages beyond
the farm gate.
The contribution of agriculture alone varies significantly between
countries, reflecting the mix of different farming systems. For example,
in Ireland, with its large dairy industry and other livestock activities,
agriculture accounts for 28 per cent of the country’s total greenhouse
gas emissions, whereas for the EU as a whole agriculture’s contribution
is around 10 per cent. In the UK, total greenhouse gas emissions from
the food chain were estimated to be around 160 mt of CO2-equivalent
(see below for an explanation of this), representing around 22 per cent
of total UK economic activity (Defra 2009). Figure 4.1 provides a visual
breakdown of the UK food chain. It shows that primary food production
– comprising farming and fishing activities within the UK – account for
53 mt of CO2-equivalent emissions, or just over one-third of the total for
the entire food chain. Next by importance, amounting to 39 mt of
CO2-equivalent (around a quarter of total emissions), is the net balance
of food imports over exports, which relate exclusively to production
operations in countries outside the UK. The third largest source of food-
related greenhouse gas emissions (21 mt) arises from UK households in
their performance of shopping, storage and preparation of food. It is
worth noting that the relative share of this category has grown from
11 per cent in 2002 to 13 per cent of total emissions in 2006, with
almost half of this increase due to food shopping by car (Defra 2009).
By comparison, some of the other elements of the food chain appear to
be modest contributors of greenhouse gases. The environmental impacts
of these elements are discussed in chapter five.
 Agri-technologies 3%

Primary food
production 34%
Net trade
[emissions of imported
foods–exports]
24%

Households
13%
Food
manufacturing 8%
Food service
3%
Retail Transportation
6% 9%

Figure 4.1 Greenhouse gas emissions from the UK food chain, 2006
Source: adapted from Defra 2009: 46.

Effects of climate change

At the beginning of the section on energy in chapter three, it was noted
that the natural greenhouse effect is a positive feature of life on Earth;
however, the enhanced greenhouse effect is a consequence of the
increased atmospheric concentrations of greenhouse gases. These gases,
principally carbon dioxide (CO2), methane (CH4) and nitrous oxide
(N2O), absorb the reflected radiation from the Earth’s surface and are
warmed by it. As their atmospheric concentrations have grown – in the
case of CO2 from 280 parts per million by volume (ppmv) in the pre-
industrial era to 390 ppmv today – so the atmosphere has become
warmer: average global temperatures have risen by 0.76°C in the past
100 years. The phenomenon of human-induced climate change with
warming of the climate system is now regarded as unequivocal.
According to the evidence, the twelve warmest years on record within
the past 150 years have occurred during the past thirteen years. There
are particular anxieties in relation to the following.
l The rate of loss of the polar ice sheets – recent data from ongoing
monitoring indicate that the rate of loss is proceeding more quickly
than was predicted by the Intergovernmental Panel on Climate
Change Fourth Assessment Report (AR4) (IPCC 2007). The
consequences of ice sheet retreat include a reduction in the level of
reflectivity so that more heat is absorbed by the Earth’s surface,
particularly the oceans. Moreover, the volume of meltwater will
contribute to rising sea levels at a much greater rate than anticipated
by AR4, with studies since 2007 ranging from a 0.5 to 2 m rise by
2100 (Pew Centre 2009).
l Tipping points – rather than a continued linear response between
warming and ecological effects, scientists are concerned about
critical thresholds that could trigger accelerated rates of change. The
loss of Arctic sea ice and the retreat of the Greenland ice sheet are
particular causes of concern, as is the rate of thawing of permafrost
in the Arctic Circle that would release huge quantities of methane.
l An increase in the frequency and intensity of extreme events related
to meteorological phenomena such as hurricanes, tropical cyclones,
heatwaves, droughts and floods. A recent report draws on data from
the global reinsurance group Munich Re, and notes that “in 2007
there were 960 major natural disasters (the highest ever such figure)
with more than 90% being the result of extreme weather-related, or
climate-related events, together accounting for 95% of the 16,000
reported fatalities and 80% of the total $82 billion economic losses”
(Lancet 2009: 1706).
Of course, climate change will not be felt equally everywhere, and the
consequences for primary food production will also vary significantly.
For example, in mid- to high-latitude regions of the world (North
America, Russia, Central Asia, Northern Europe), it is thought that the
rise in ground-level temperatures will lengthen the growing season and
should increase yields, at least in the medium term. Increased levels of
atmospheric CO2 will also have a generally positive effect on crop
growth by enhancing photosynthesis. During the next two decades, it
may be possible for these regions to grow commercially crops that have
not been viable in the past, such as wine grapes and other temperature-
sensitive fruits, as well as to extend the northern limit of cereal
cultivation. Yet, as warming proceeds, there will be increased
 evapotranspiration and soil moisture losses that will be exacerbated in
many regions by a projected fall in precipitation levels. By mid-century,
excessively high temperatures and water shortages, as well as a
likelihood of increased plant disease, fungal infections and pest
outbreaks, will threaten crop yields.
For the developing regions of the low latitudes (within the tropics),
climate change brings not only warming, which is a concern where
crops may already be close to their limits for heat and water stress, but
greater variability and more erratic rainfall. Projections of changing
precipitation patterns will particularly affect rainfed agriculture – which
covers 96 per cent of all cultivated land in Sub-Saharan Africa, 87 per
cent in South America and 61 per cent in Asia. This will present an
enormous challenge in regions where there is already considerable
vulnerability, where the majority of households depend most on
agriculture and where there are fewer alternative sources of income, and
consequently more limited adaptive capacity. With climate change also
responsible for changes in the frequency and magnitude of more extreme
events – droughts, storms, flooding – it is apparent that in developing
regions there is increased risk of significant impacts on food production
systems. Potentially, the biggest losers from climate change are likely to
be people most exposed to the worst of its impacts and the least able to
cope while, cruelly, living in countries that have made the least
contribution to anthropogenic warming.

Agricultural greenhouse gases

But what is the contribution of primary food production to climate
change? As noted earlier, there are three principal greenhouse gases:
carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O), as well
as a number of industrial chemicals (some of the fluorocarbon family
used in refrigeration) that are long-lived with high global warming
potential. Although CO2 is the single most important greenhouse gas by
volume and is responsible for about 85 per cent of the UK’s contribution
to global warming, agricultural production is more closely associated
with emissions of methane and nitrous oxide. These gases are more
powerful than CO2, having a global warming potential twenty-three
times that of CO2 in the case of CH4, and 296 times for N2O. While CO2
is most closely associated with the burning of fossil fuels, CH4 is closely
linked to livestock, especially ruminants that possess four gastric
chambers through which they are able to break down cellulosic plant
material. This digestive system generates methane, which is released at
either end of the animal. Wetland rice cultivation is also a source of
methane. N2O emissions, on the other hand, derive from the
volatilisation of nitrogen from the soil, principally that derived from
chemical fertilisers. Much research has suggested that crops may not
take up even half the nitrogen made available through the application of
fertiliser; the remainder is either leached through the soil creating
pathways of nitrate contamination in ground and surface water, or is
released as N2O into the atmosphere.
Drawing on data in the AR4 of Working Group III of the IPCC (Smith
et al. 2007), it is possible to consider the contribution of agriculture to
the emission of greenhouse gases. It notes that in 2005, agriculture
accounted for between 5.1 and 6.1 Gt CO2-eq per year (i.e. 5.1 billion
tonnes of CO2-equivalent per year), which represents between 10 and
12 per cent of global anthropogenic emissions of greenhouse gases
(those derived from human activities, not natural ecosystem sources).
Of this, CH4 contributed around 3.3 Gt CO2-eq per year, and N2O
2.8 Gt CO2-eq per year, which represents 50 and 60 per cent,
respectively, of the total global anthropogenic emissions of these gases.
On the other hand, agriculture is responsible for only a relatively small
net contribution of CO2 (in the order of 40 million tonnes) despite the
large annual exchanges between the atmosphere and terrestrial
production systems. This excludes all post-farm-gate transportation and
processing activities. The individual sources of global agricultural
greenhouse gas emissions are shown in millions of tonnes of CO2-
equivalent in Figure 4.2. It must be remembered that emissions from
soils and from cattle stand out due to the higher global warming
potential of N2O and CH4.
However, as far as the two primarily agricultural greenhouse gases are
concerned, evidence from the recent past and projections into the short
term rather cloud optimistic prospects for the medium to long term.
Over the period 1990–2005, global agricultural emissions of CH4 and
N2O grew by 17 per cent, or by 60 mt CO2-eq per year. This growth was
most marked in those regions of the global South and economies in
transition, so that by 2005 they collectively accounted for three-quarters
of total global agricultural emissions. The regions of the developed
North, in contrast, displayed a decrease in the emissions of these gases,
led by Western Europe, which has been implementing a range of agri-
environmental measures. Forward projections of global agricultural
emissions to 2030 are not encouraging, however, with N2O emissions
 2500

2000
Million tonnes (CO2 -eq)

1500

1000

500

0
N2O CH4 Biomass Rice Manure Fertilser Irrigation Farm Pesticide
from soils from cattle burning production production machinery production
operations

Figure 4.2 Sources of agricultural greenhouse gas emissions

Source: drawn from data presented in Bellarby et al. 2008.

expected to increase by between 35 and 60 per cent due to more

widespread use of chemical fertilisers, and CH4 by 60 per cent in direct
proportion to an anticipated increase in livestock numbers.
In this regard, the challenge of mitigating climate change through
reduced emissions of these two greenhouse gases most closely linked to
food production becomes apparent. With rice such an important staple
crop in East and South-East Asia, as well as elsewhere, a projected
increase in irrigated area is likely to lead to a 16 per cent increase in
CH4 from paddy cultivation. Meanwhile, the Middle East and Africa
have the highest projected growth in agricultural emissions: a 95 per
cent increase from 1990 to 2020, much of which will be attributed to
N2O. Given the need to enhance food production capacity in Sub-
Saharan Africa, which has been marked by poor and declining levels of
soil fertility and low levels of fertiliser use, it is understandable that
efforts would be made to increase rates of fertiliser application in pursuit
of improved food security. Meanwhile, in Latin America and the
Caribbean, significant changes in land use and land cover, with forests
such those as in Amazonia converted to crop- and grassland and with a
resulting doubling of cattle numbers between 1961 and 2004, there have
been, and will continue to be, increases in emissions of CO2 and CH4.
As for climate change-mitigation options, the agricultural sector offers
some of the more technologically and commercially feasible
opportunities with the potential to change the position of primary food
production from the second largest source of greenhouse gas emissions
to a much lesser contributor, even a net sink. Although efforts should
continue to be directed toward ways of reducing CH4 and N2O, there are
opportunities to sequester carbon through improved land and soil
management practices. Some options are outlined in Box 4.1. There is
also some enthusiasm in certain quarters for the development of biochar.
Subjecting particular kinds of biomass to industrial processes such as
pyrolysis gives rise to gases that can be liquefied and used as fuels for
energy generation, with the residue comprising a highly porous form of
charcoal that, if added as an amendment to soils, is believed to enhance
fertility and to sequester significant quantities of carbon for hundreds of
thousands of years. It might indeed be considered as the magic bullet of
carbon capture and storage for the land-based sector’s contribution to
anthropogenic climate change (Biochar Research Centre 2009).

Box 4.1

Carbon sequestration options

1 Cropland management – mitigation potential up to 1.45 gigatonnes of
CO2-equivalent per year (Gt CO2-eq/year):
• avoiding bare soil through use of catch- and cover-crops between the
main commercial crops, thus preventing erosion and leaching of
nutrients, and locking up carbon;
• “smarter” applications of nitrogen fertiliser, for example in smaller
quantities and at the right time for optimum uptake by crops, and
making greater use of nitrogen-fixing legumes to compensate for less
fertiliser use;
• eliminating the burning of crop residues (such as corn stubble);
• reduced tillage – no-till farming avoids ploughing, and minimal soil
disturbance optimises carbon sequestration.
2 Grazing land management – mitigation potential up to 1.35 Gt CO2-
eq/year:
• reduced grazing intensity (lower livestock numbers) and use of fire (to
burn off noxious plant species) can enhance CO2 uptake in the soil and
in biomass.
 3 Restoration of degraded lands and drained wetlands – mitigation
potential up to 2 Gt CO2-eq/year:
• adding nutrients and organic amendments to degraded soils can restore
their productivity and carbon-holding capacity, to avoid the further
conversion of wetlands and to restore those no longer suitable for crop
production.
4 Improved water management – mitigation potential up to 0.3 Gt CO2-eq/
year, particularly in wetland rice farming.
5 Set-aside and improved manure management – mitigation potential
0.3 Gt CO2-eq/year, including converting cropland to agro-forestry.
6 Improved efficiency in manufacturing fertilisers – mitigation potential up
to 0.2 Gt CO2-eq/year: energy-saving and improved nitrous oxide-
reduction technology.

Source: adapted from Bellarby et al. (2008).

3. Global freshwater resources

In chapter three, within the evaluation of resources required for primary

production, the utter indispensability of water was made clear. Yet there
is a very uneven distribution of freshwater resources across the Earth
and, in some regions, the shortfall is becoming more acute not only for
food production, but for meeting basic human needs. It has been
calculated that agriculture accounted for 90 per cent of global freshwater
consumption during the past century. Yet the global population is
projected to increase to more than 9 billion by 2050, while the
international community struggles to make progress on the Millennium
Development Goals’ poverty targets (that prioritise meeting the water
and sanitation needs of the poorest), and all this at a time when the
global climate system is entering a warmer, more unstable phase. This
will inevitably place further stress on water resources (Khan and Hanjra
2008).
This section reviews the challenges posed by the utilisation of water for
food production, highlighting some of the problems arising from a
paradigm that appears to have treated freshwater as a limitless resource.
Recent years have, however, seen the emergence of new and important
conceptual innovations that reveal the short-sightedness of pre-existing
attitudes to water as an infinite gift of nature, and indicate possible
pathways toward a more rational and sustainable utilisation. While
outlining some of these new concepts, it will nevertheless become clear
 that without some significant changes to the underlying rationale of
international trade, the potential crisis facing global freshwater resources
will be realised.

Extracting water
Prior to the modern, post-1945 era, much of the world’s food needs, at
least beyond the leading industrialised countries, were met through
national or subnational provision, most commonly at the scale of the
river basin, with the moisture needs of agriculture supplied by rainfall or
through the diversion of surface water resources. In the latter case, the
vast majority of irrigation schemes largely comprised simple, gravity-fed
systems that drew from streams and rivers, and involved the channelling
of water along furrows, or the flooding of fields as in paddy rice
cultivation. Rice has been grown under irrigation in the Far East for
5000 years, with rice terraces in Banaue, the Philippines under
continuous irrigated cultivation for 3000 years, illustrating the
sustainable nature of these maintained structures. Although gravity-fed
systems are generally technologically simple and easily maintained, they
are inefficient as up to half the water supplied can be lost through
seepage, spillage and evaporation. In the absence of surface water
resources at a higher altitude than the fields, native ingenuity found
ways of lifting water, for example by using variations of Archimedes’
screw and the Persian wheel powered by people and animals, or wind-
powered pumps. However, it is with the development and widespread
availability of pumps powered by diesel oil and electricity that the
constraints of channelling surface water or shallow groundwater could
now be overcome, and the resources of deep aquifers exploited. Box 4.2
provides an example of how a particularly large subterranean water
resource can be seriously depleted by the widespread use of pumps
supplying high-pressure sprays and sprinkler systems.
The harnessing and management of large hydrological resources does
appear to offer enormous potential for agricultural and economic
development. Indeed, it has been argued that the first agricultural
civilisations – Ancient Egypt, Mesopotamia, China, India and pre-
Columbian Peru and Mexico – were characterised by a high degree of
social organisation based around irrigation and the management of
water. Although the notion of hydraulic civilisations of Karl Wittfogel
(1896–1988) has been subjected to profound critique from scholars,
there is nevertheless a plausible argument to be made that certain states
do indeed appear to be attracted by the social prestige and political
capital to be gained from engaging in large-scale hydraulic engineering
projects. Box 4.3 provides examples of giant dam construction and their
consequences.

Box 4.2

Mining the Ogallala aquifer

The Ogallala aquifer is an example of a non-renewable stock resource rather
than a renewable flow resource. It stretches under portions of eight states,
from South Dakota to Texas, covering the High Plains of the USA, a semi-
arid region marked by low (< 500 mm) and variable annual precipitation.
Without irrigation, the High Plains ecozone would support extensive grazing,
either of the native buffalo that once prevailed across this landscape, or the
cattle that came to replace them. Initial efforts to develop arable farming
resulted in the Dust Bowl years of the 1930s. Today, however, 94 per cent of
the water withdrawn from the aquifer supports 5.5 million ha of irrigated crop
production, largely of cereals, soybeans and cotton, and constitutes fully one-
fifth of the total annual US agricultural output, worth more than $20 billion.
The Ogallala was created several million years ago by streams flowing from
the Rocky Mountains, but geological activity has since cut this source of
mountain run-off so that there is no longer any effective recharge of the
aquifer. Systematic mining of the Ogallala began in the late 1940s and early
1950s, and the popular view was that the resource was inexhaustible. The
number of wells soared (in west Texas alone from fewer than 1200 in 1937
to 66,000 in 1971), as did the rate of withdrawal (a fivefold increase
between 1949 and 1974). In some areas, the rate of extraction was causing a
1.22–1.83-metre fall in the level of groundwater each year, while
precipitation (rainfall and snow melt) was able to restore just 1.27 cm. In
2000, total groundwater withdrawal from the aquifer for irrigation amounted
to 21 million acre-feet (or 26 km3). Area-weighted, average water-level
change from pre-development to 2005 shows a decline of 3.9 m, with some
areas recording declines in excess of 100 feet (30.5 m). In southern Kansas,
wells have dried up and farmers have had to revert to no-till dryland farming
methods in order to grow wheat. Elsewhere, smart robotic irrigation systems
are improving the efficiency of water use and reducing rates of extraction.
However, new developments threaten to put the aquifer under increasing
pressure: the expansion of the biofuels sector is a major economic incentive
for growing more maize, while there are plans to tap the aquifer in order to
supply drinking water to the cities of Dallas and even El Paso, over 1000 km
away.

Sources: adapted from McGuire (2007); Braxton Little (2009).

 Box 4.3

Damming the rivers

India lies third in the league of dam-building nations, and has been engaged
in the highly controversial Narmada River scheme, most attention to which
has focused on the Sardar Sarovar Dam. The height of the dam has been
raised on several occasions since its initial formulation, and is now expected
to reach 122 metres that will impound a reservoir 214 km in length and
involve the displacement of 240,000 people. Proponents of the scheme
argue, however, that the 66,000 km of canals will allow for the irrigation of
1,845,000 ha of low-productivity drylands.
China sits at the top of the international league of dam-building nations, with
around half of the world’s large dams. It has come under most criticism,
however, for its Three Gorges scheme, the world’s largest hydropower
project, which sets records for number of people displaced (more than
1.2 million); number of cities and towns flooded (thirteen cities, 140 towns,
1350 villages); and length of reservoir (more than 600 km). Critics argue that
the scheme is vulnerable to a number of environmental problems, ranging
from the hazards of industrial pollutants and siltation in the reservoir, to the
possibility of reservoir-induced seismicity. The latter notion suggests that
earthquakes may be triggered by the extra water pressure created by a large
reservoir, and that the construction of high dams in seismically active regions
should be avoided. More geophysical research is under way to improve
understanding of the propensity of regions to reservoir-induced seismicity.

Source: adapted from International Rivers (2009).

The construction of dams on rivers has been a focus of sustained

criticism over many years. Indeed, there is an established catalogue of
well documented social and environmental problems arising from large
dams. These include: problems associated with the impounding of water
and the creation of large, often quite stagnant lakes that occupy
significant areas of arable land and displace previously resident farm
families; the drowning of forests and the drainage into the reservoirs of
nutrients which significantly raises the potential for eutrophication; and
increased exposure of neighbouring populations to vector-borne diseases
(such as malaria and schistosomiasis). These constitute some of the
issues upstream of the dam, in the catchment zone. Below the structure,
other problems arise associated with irrigation schemes themselves,
particularly salinisation of the soil. This can occur in low-lying arid and
semi-arid regions, where groundwater is often quite heavily charged
with salts and irrigation causes waterlogging, the water table rises, and
intense evaporation leaves salts in the uppermost layers of the soil. It has
been estimated that around one-fifth of the world’s irrigated land suffers
from salinisation or waterlogging, with around 1.5 million ha per year
lost from production. The dangers of inappropriate irrigation cannot be
underestimated, as the final example in Box 4.4 illustrates.
Although large-scale dam building remains popular amongst technocrats,
the net rate of increase in irrigated area has fallen steadily for the past

Box 4.4

Draining the Aral Sea

The Aral Sea lies in Central Asia, occupying territory in Kazakhstan and
Uzbekistan, and in 1960 it was the fourth largest lake in the world. It has no
outlet, but equilibrium existed between inflow, provided by two principal
rivers, the Amu Darya and Syr Darya, and the evaporation rate of about 60
km3 per year. During the 1920s, the Soviet Government initiated a
programme of cotton cultivation in the region, and this was massively
expanded during the 1950s. Under the control of collective farms, millions
of acres of irrigated land were created, fed by large volumes of water
diverted from the two rivers before they reached the Aral Sea. Though it still
received an estimated 50 km3/year of water in 1965, by the 1980s the flow
of freshwater into the sea was zero. The Aral Sea consequently began to
shrink, the shoreline retreated, and more than 28,000 km2 of former lake bed
were exposed. Strong winds picked up an estimated 43 million tonnes of
sediment laced with salts and pesticides, and the resulting dust storms in the
region are attributed with causing a huge increase in incidences of
respiratory illness and cancers, especially of the throat.
As the Sea shrank – its surface area was reduced to half by the 1990s and its
volume by 75 per cent – the concentration of salts and minerals in the water
resulted in a dramatic alteration of the lake’s ecology. Fish stocks collapsed,
and with them the fishing industry, which had employed up to 60,000
people. By 2000, the Sea had split into two and the southern part had further
split into western and eastern lobes.
The disappearance of the Aral Sea has also had an effect on the regional
climate, its moderating influence giving way to shorter, hotter and drier
summers and longer, colder winters.
A time-series of satellite photographs showing the retreat of the extent of the
Aral Sea during the 2000s can be seen online at http://earthobservatory.
nasa.gov/Features/WorldOfChange/aral_sea.php

Sources: adapted from Millennium Ecosystem Assessment

(2005a); UNEP-GRID (2010).
four decades to less than 1 per cent per year as unexploited freshwater
resources become more limited. Many countries within the band
stretching from China through India and on through the Middle East to
North Africa either currently or will soon experience insufficient water
supply to maintain per capita food production from irrigated land.
Moreover, agriculture is facing increasing competition for freshwater
resources from other sectors: the needs of growing urban populations,
industry, fisheries, recreational use and demands for the protection of
natural ecosystems. It is estimated that water scarcity will affect 40 per
cent of the global population by 2050, and with climate change
scenarios suggesting reduced rainfall in some areas (and the likelihood
of increased flooding in others), water stress is likely to grow.
Consequently, it is being suggested that we need to switch our approach
from “humid zone thinking” (that encourages us to ask how much water
we need and where we get it from) to one better suited to the rapidly
emerging situation of water scarcity (how much water is there, and how
can we optimise its use?) (Crabb 1996). Water availability is a function
of nature and the hydrological cycle; water scarcity is a consequence of
inappropriate human activity ill-suited to the resources in a particular
location. A perspective that begins from an understanding of scarcity
and the urgent need for more sustainable management of water resources
will probably result in a variety of different responses.
Economists, for example, have promoted water charges as a means of
regulating use, believing that a market price for water will emerge that
will secure improved efficiencies in its use. But it is critical that
operationalising such measures do not become a means of bolstering
private profit over human need, as appeared to be the case in the
disastrous efforts to privatise water supply in Cochabamba, Buenos
Aires and Manila in the 1990s. Engineering agronomists accustomed to
achieving a “technological fix” for resource constraints are meanwhile
focused on devising new ways of improving the efficiency of water
delivery and uptake. These include “high-tech” sensors that calculate
crop needs, and robotic irrigation systems, as well as the genetic
engineering of more drought-tolerant crops.
On the other hand, there are a huge number of “lower-tech” but tried-
and-trusted measures that can be drawn upon. These range from drip
irrigation systems that deliver water to plant roots; through the use of
deficit irrigation, which delivers less water than the crop might
optimally use, but where the modest decline in yield is more than
 compensated by savings of water; to well established water conservation
methods such as low-tillage farming, mulching, rainwater harvesting or
the use of wastewater in urban and peri-urban areas.
While welcoming “more crop per drop” technical innovations, this
cannot distract us from more fundamental concerns about the underlying
assumptions of water use and its continued availability in some regions
of the world. It is clear there is a need for improved governance of
water at a variety of different scales: from the global through the
national level, to drainage basins and village committees, to individual
water users. Above all, a more integrated approach is required in which
a human right to safe drinking water and sanitation is established as the
first priority of any scheme engaged in harnessing hydrological
resources, whether for the generation of hydropower or the delivery of
irrigation. Naturally, it also raises questions about the logic of a global
agri-food system built upon the principle of comparative advantage,
where countries and regions have become almost “designated” producers
of certain commodities – reflecting their historical roles within the world
economy and its international division of labour, the relative value of
factors of production (especially land and labour), and domestic policies
– rather than reflecting their actual endowments of water and other
resources. This international trading system has consequently given rise
to the cultivation of unsuitable dryland environments with inappropriate
crops destined for distant markets, rather than prioritising the pursuit of
local food and water security.

Virtual water
To speak of our appetite for “eating water” is to draw attention to the
volumes of water locked up – or embedded – in primary foods, for
which the term “virtual water” is now widely used (Roth and Warner
2008). Different crops have different water requirements, which will
also vary according to the climates in which they are cultivated, and it is
possible to calculate the virtual water content of all foods. The global
average for maize, wheat and husked rice, for example, is 900, 1300 and
3000 m3 of water per tonne, respectively, the higher figure for rice
reflecting the levels of evaporation from flooded paddy fields as well as
the weight loss arising from processing the grain (Hoekstra and
Chapagain 2007). Livestock products have much higher virtual water
content than crop products because of their extended production time
and the nature of inputs required. For example, beef produced under
 industrial farming takes, on average, three years to reach the slaughter
weight of 200 kg of boneless beef. During this time, the animal
consumes nearly 1.3 t of grain, 7.2 t of roughage (pasture, silage), 24 m3
of water for drinking, and 7 m3 of water for servicing. By calculating the
amount of virtual water within the feeds, it is possible to estimate the
total volume of embodied water in the beef. Although this varies
between countries, Hoekstra and Chapagain (2007) suggest a global
average of 15,500 m3 per tonne, or, in terms that we can grasp more
easily, it takes around 15 m3 of water to produce 1 kg of beef. The
virtual water content of some selected primary food products is shown
in Table 4.1. As the virtual water content of any product will vary
depending upon the climate, the technology deployed and the yields
achieved, the table provides the limits of the range reported for a
number of selected countries by Hoekstra and Chapagain (2007) as well
as the global average.
In the same paper, the authors report that the total global volume of
water used for crop production at field level, i.e. before processing,
amounts to 6390 billion m3 per year. Rice constitutes the largest share,

Table 4.1 Global average virtual water content of selected primary foods (m3/tonne)

Product Range Global average

Rice (broken) 1822–4600 3,419

Wheat 619–2421 1,334

Maize 408–1937 909

Soybean 1076–4124 1,789

Coffee (roasted) 5790–33475 20,682

Tea 3002–11110 9,205

Beef 11019–37762 15,497

Pork 2211–6947 4,856

Sheep meat 3571–16878 6,143

Chicken meat 2198–7736 3,918

Eggs 1389–7531 3,340

Milk 641–2382 990

Source: adapted from Hoekstra and Chapagain (2007).

 Coffee
Cassava beans
2% 2%
Millet 2% Potatoes 1%
Coconuts 2%
Rice,
Sorghum 3%
paddy 21%
Barley 3%

Seed cotton
3%
Sugar cane
3%

Soybeans
4%

Maize
9%

Other 33%
Wheat
12%

Figure 4.3 Share of global field-level water use by different crops

Source: adapted from Hoekstra and Chapagain (2007).

consuming 1359 billion m3 per year, or 21 per cent of total volume used
for crop production, while wheat has a 12 per cent share. To put this in
perspective, wheat occupies nearly 40 per cent more cropped area than
rice and produces a little less in total volume of output (see Table 3.1).
The breakdown of global water use by crops is shown in Figure 4.3.
It is clear from these data that, depending upon the composition of our
diet, we eat a greater or lesser amount of virtual water. However, if
many of the food items that we consume are imported from other
countries, then we are also importing virtual water embodied in the
production of those items. For water-scarce countries, the concept of
virtual water has tremendous potential in guiding agricultural policy, for
it would suggest the most rational course of action would be to import
products requiring a lot of water in their production, rather than to
produce them domestically. This would result in water savings, relieving
the pressure on resources and allowing them to meet the needs of other
users. Jordan, for example, now imports 60–90 per cent of its food,
representing 5–7 billion m3 per year of water, which is in contrast to the
1 billion m3 per year it withdraws from domestic water sources (Roth
and Warner 2008). On the other hand, for countries rich in water
resources, the virtual water content of imported foods probably has little
bearing on their demand for products that cannot be grown domestically
as a consequence of climatic factors other than water availability. In this
respect, the notion of virtual water can only go so far as a guide to
domestic production and international trade.
Nevertheless, it is apparent that the international flow of virtual water
within global trade is significant. Chapagain and Hoekstra (2008)
undertook a comprehensive study of international virtual water flows for
the period 1997–2001 and estimated this amounted to 1625 billion m3
annually. Of this, 61 per cent related to international trade in crops and
crop products, and 17 per cent to trade in livestock products. Industrial
goods accounted for the remaining 22 per cent. Of particular interest
from this study is the identification of the major exporting and importing
countries of virtual water. Figure 4.4 provides summary data from a
more comprehensive table developed by Chapagain and Hoekstra, which
provides detailed accounts of virtual water imports and exports for
twenty-four countries. Figure 4.4 lists the eight most significant
exporters and importers of virtual water related to crop and livestock
products only.
The data are revealing: Australia, a continent-sized country that
comprises a great deal of desert and semi-arid environments, and which
in recent years has been experiencing serious drought, is the world’s
largest net exporter of virtual water. The USA is, however, the largest
gross exporter, that is, before subtracting its virtual water imports. India,
too, emerges as a net exporter of virtual water, with 21.5 billion m3 per
year of water exported in crops, mostly to other countries in Asia. This
is especially surprising given the state of India’s groundwater resources:
an article in the journal Nature highlights the unsustainable rates of
extraction largely for irrigation in the country’s north-western states of
Punjab, Haryana and Rajasthan, where excess withdrawal of 109 km3
occurred between 2002 and 2006 (Rodell et al. 2009). On a regional
basis, Africa is a net exporter of water to other continents, particularly
Europe. By contrast, the largest importer of virtual water – and by some
 Japan
Italy Balance of trade in
Russia crop products (Gm3/yr)
UK Balance of trade in
Germany livestock products
China (Gm3/yr)
Korea
Mexico

France
India
Thailand
Canada
Argentina
Brazil
USA
Australia
–100 –80 –60 –40 –20 0 20 40 60 80
Net importers Net exporters

Figure 4.4 Virtual water flows: leading exporters and importers

Source: drawn from data in Chapagain and Hoekstra (2008).

distance – is Japan, which imports crop and livestock products

containing over 77 billion m3 per year of water. On this basis, it could
be asserted that Japan’s resource and food security is achieved by
drawing down water (as well as land) resources in other countries.

Water footprints
The concept of virtual water is useful when assessing the implications of
international trade in primary foods: it helps reveal the hidden flows of
water between countries and regions. However, if we wish to get a
better understanding of an individual country’s level of consumption of
water, then the related concept of water footprint becomes especially
valuable. In recent years the term “ecological footprint” has become
widely used to represent the area of productive land and aquatic
ecosystems required to produce the resources used, and to assimilate the
wastes generated by, a population in a circumscribed area (country,
region, city) (Wackernagel and Rees 1996). The water footprint, in
contrast, indicates the volume of water required per capita to sustain a
population. In order to calculate the total water footprint of a nation, it is
necessary to quantify not only the water used to produce goods and
services for domestic consumption (endogenous water), but also that
virtual water which is embodied in imports of goods and services from
other countries (exogenous water).
Perhaps unsurprisingly, the USA emerges with the largest water
footprint per capita, 2480 m3 per year, with the Mediterranean countries
of Southern Europe not far behind. At the other end of the scale, the
lowest water footprint is represented by China, with an average of
700 m3 per year per capita. According to Hoekstra and Chapagain
(2007), the size of the global water footprint is determined largely by
the consumption of food and other agricultural products that comprise
86 per cent of the total, 73 per cent drawn from internal resources and
13 per cent from exogenous virtual supplies. The manufacture of
industrial products, in contrast, accounts for just 9 per cent of the global
water footprint, although this varies considerably between rich states
(e.g. the consumption of industrial goods accounts for 32 per cent of the
total water footprint of the USA) and developing countries (in India this
sector accounts for just 2 per cent). Domestic water consumption for
drinking and sanitation accounts for 5 per cent of the global water
footprint.
Hoekstra and Chapagain (2007) identify four factors that explain high
water footprints:
l The total volume of material consumption, which is related to gross
national income; examples here include the USA, Switzerland and
Italy.
l Climate – in areas of high evaporative demand, the water
requirement per unit of crop production is relatively large, a
situation facing Senegal, Mali, Sudan and Chad, amongst others.
l Water-inefficient agricultural practices are responsible for a low
output per unit of water used; Thailand, Cambodia and
Turkmenistan are amongst those guilty here.
l Finally, and significantly, a water-intensive pattern of consumption,
especially where meat is a major part of the diet. The USA averages
meat consumption of 120 kg per capita per year, a level that is three
times the global average, but the countries of Western Europe are
not so far behind. Given the rapid growth in meat consumption
worldwide, the implications of this dietary change for water use is
clearly of vital importance.
In drawing this section on water to a close, we can see that there is a
long-standing and deep-seated relationship between societal
development and the harnessing and management of freshwater
 resources. The capture and distribution of water for irrigation enabled
higher levels of agricultural productivity, supporting greater numbers of
people able to reach new cultural and technological achievements. As
countries sought to extend their influence over distant lands, new foods
were produced and transported around the world-shaping dietary habits.
As demand grew, first in line with population numbers, then with rising
material prosperity, food production required ever-greater quantities of
freshwater for growing crops and grazing animals. An innate tendency
toward the large scale meant bigger dams, larger pumps and greater
rates of extraction. All too soon, we began to come up against some
hydrological limits: the depletion of stocks (aquifers) and the
unsustainable utilisation of flows.
Water footprint analysis and the related concept of virtual water help to
sharpen our understanding of water-use efficiency in food production
and to highlight potential problems in regions with limited water
resources. Such analysis might encourage some countries to re-orient
cropping systems away from products requiring a high endogenous
water content to those requiring less, with the consequent shortfall in
domestic demand for those products met by imports from countries with
high positive water budgets and/or greater water productivity.
It also reveals significant inequalities between countries. For example,
some African countries have hardly any external water footprint simply
because they import so little; yet, despite experiencing water shortages
themselves, they export crops with a high virtual water content, such as
horticultural products, cocoa and cotton. Paradoxically, some European
countries (e.g. UK, Netherlands) with abundant water resources of their
own have a high external water footprint (accounting for up to 80 per
cent of their total) given their imports of agricultural and industrial
products. This situation reveals that an appreciation for the distribution
of water resources worldwide has played no part in shaping the pattern
of global trade, but ought to be factored into ongoing discussions that
are otherwise driven by entirely financial considerations.

4. Peak oil and its implications

The current level of primary food production – as well as much of the

post-farm-gate processing, manufacturing, distribution and retailing of
food, as chapter five shows – owes a very great deal to the use of fossil
fuels. Chemical fertilisers are manufactured from natural gas; oil drives
field machinery, irrigation pumps and post-harvest equipment. These
hydrocarbons derive from the ability of algae and plants to capture solar
energy many millions of years ago, to bloom and then to die. The decay
of this organic matter, buried deep under layers of sediment then
compacted and heated over a prolonged period, became natural gas or
crude oil depending upon the temperature and pressure to which it was
subjected.
Were fossil fuels to be in infinite supply, there would no reason to
suppose that the global agri-food system in its present state would not
continue to rest upon an abundance of cheap energy. Unfortunately,
there is a growing realisation that the evidence demonstrates otherwise:
that we are at, or close to reaching, a situation of “peak oil”, where
we have used up approximately half the geological endowment of
conventional oil and natural gas. This would suggest a near future of
much tighter energy markets as supply is unable to meet demand and
prices will rise. This will have direct and fairly immediate repercussions
for primary food production, as events during 2008 demonstrated
(discussed in chapter six).
Some of the most salient facts are as follows:
l The biggest oilfields in the world were discovered more than half a
century ago on the Arabian peninsula.
l The peak of oil discovery was 1965, and apart from the discoveries
of the Prudhoe Bay field in Alaska and the North Sea during the
1970s, there has been a relentless fall in discoveries since that time.
l An estimated 80 per cent of all oil used in the world comes from
fields discovered before 1973.
l Since 1981, we have been using more oil than has been found, and
the gap between discovery and consumption has been growing
wider with every passing year. In 2007, the world consumed six
barrels for every one that was discovered.
l The world is consuming 84 million barrels of oil each day, or
30 billion barrels per year. (A barrel comprises 42 US gallons,
35 UK gallons or 159 litres). Even a passing acquaintance with
geology will suggest that fossil fuels cannot last indefinitely
(Campbell 1997, 2003; Leggett 2005).
At one level, peak oil reflects the circumstances to which all individual
oil fields generally subscribe. That is, when initially tapped and under its
own pressure, oil flows freely on a rising curve of production until
reaching a maximum point, to be followed, as pressure falls, by a
 declining return until exhaustion. Plotting such output on a graph
provides a symmetrical, bell-shaped curve. Aggregating the individual
bell curves for each oilfield in the lower forty-eight states of the USA
was a task undertaken by geologist M. King Hubbert in 1956. Plotting
the production rates over the preceding 100 years, and using a generous
estimate for ultimate recoverable oil, Hubbert calculated that US
production would peak in 1971. Although widely ridiculed within the
industry, peak oil arrived in the lower forty-eight states a year earlier
than Hubbert’s prediction (Leggett 2005). In the United States, with its
largely unconstrained market driven by the principles of free enterprise,
the pattern of oil production was remarkably smooth and symmetrical.
Elsewhere, however, political interventions in the oil market – such as
that exercised by the Organisation of Petroleum Exporting Countries
(OPEC) in 1973 when they restricted supplies in order to effect a
substantial increase in crude oil prices – would result in more uneven
curves; wars, technological change and other unanticipated events also
have an effect on production patterns. In the UK, North Sea oil
production peaked in 1999 and output is now around 43 per cent lower;
Norway’s production, too, is now 25 per cent less than in 2001; while
the Prudhoe Bay field in Alaska produces just a third of the oil that it
did two decades ago (Mouawad 2008).

Future projections
The key issue, however, is: just how much oil is there? We will put
natural gas to one side for the moment and concentrate on what is called
“conventional” oil. This is the lightest crude that has dominated
production to date, and has been found in the largest reservoirs that were
easiest to locate and to exploit: close to the surface, often on dry land,
and flowing under their own pressure. Given that geologists, both on the
ground and remotely using the latest satellite technology, have mapped
the world extensively, one might expect a consensus on the total
ultimate recoverable resource. The world of oil, however, is one of
Machiavellian politics, where producer countries and oil companies
revise estimates of assets for purposes other than scientific transparency.
Fortunately, a network of independent analysts, led by a retired
petroleum geologist, Dr Colin Campbell, have worked assiduously over
the past fifteen or so years to strip away the obfuscation and political
spin associated with this data. According to the Association for the
Study of Peak Oil & Gas (ASPO), founded by Dr Campbell, the best
estimate for the total endowment of conventional oil prior to its
exploitation is a little less than 2 trillion barrels (2 1012) (Campbell
and Laherrère 1998). Of this, we have used to date around 950 billion
barrels, which leaves a little over 1 trillion barrels to exploit. While this
may seem sufficient to keep us powered for many years ahead, the
geological reality casts something of a shadow:
l Much of the remaining oil is heavier and of lower quality, with
higher sulfur content, and requires more effort to refine.
l It is generally found in smaller deposits below the category of giant
fields, defined as containing more than 500 million barrels or
capable of pumping 100,000 barrels per day. The 500 giant fields
comprise just 1 per cent of the total number of oil fields in the
world, yet contribute more than 60 per cent of global output (Höök
et al. 2009). Smaller fields are more costly to exploit given their
size.
l The new fields are likely to be found offshore, and increasingly in
deep water, where the difficulties and costs of extraction are very
much greater.
l The remaining 1 trillion barrels also includes the residual oil in the
original fields. This is the oil that no longer flows under natural
pressure, but must be pumped out using water or gas or by
fracturing the rock strata. Even with such advanced techniques as
directional drilling, the most oil that can be extracted from a
reservoir is 60 per cent of its total volume, and for many fields it is
a good deal less (35–40 per cent). The Ghawar field in Saudi
Arabia, the largest oil deposit ever discovered, with 87.5 billion
barrels, still accounts for 5 per cent of the world’s total daily
production, more than sixty years after its discovery. Yet to keep the
oil flowing, 7 million barrels of seawater are injected into the
reservoir every day (Campbell 2003; Leggett 2005).
Figures 4.5 and 4.6 set out the evidence and demonstrate why it is
possible to assert that we are at peak production. Figure 4.5 shows a bar
chart of past and possible future discovery and plots a production line to
the present: its future trajectory will probably be in a downward
direction, but the rate of its decline will depend upon a host of factors.
For example, greater efforts now to maximise recovery of oil in order
to maintain total output will see much sharper declines in the future.
Figure 4.5 demonstrates very clearly the points noted above – since the
early 1980s we have been using more oil than has been found, and with
each passing year this disparity between consumption and discovery
has been growing.
 60

50 Past discovery
Future discovery
40 Production
Gb/a

30

20

10

0
1930 1950 1970 1990 2010 2030 2050

Figure 4.5 Discovery and production of regular oil

Source: Colin Campbell, personal communication.

Figure 4.6 sets out the depletion curves as derived by Dr Colin

Campbell from his analysis of industry data, and include conventional
and non-conventional oil as well as natural gas. Regular (conventional)
oil has dominated output to date and will continue to do so into the
future, yet the evidence points to a likely peak in production around
2005. By comparison, the likely contribution of non-conventional
sources by volume appears to be relatively modest. This is especially
salient when set against the enormous financial and technological
demands required in their exploitation (e.g. drilling in deep ocean sea
beds or in the Arctic), as well as the potentially catastrophic
environmental impacts likely to result. That we have reached peak
production beyond which any further increase in output of regular oil is
now widely regarded as extremely unlikely; but with rising levels of
demand, especially from China and India, oil prices are set to move
generally upward. This is why there has been such enthusiasm for non-
conventional sources of oil, such as the Canadian tar sands (“heavy
oil”), as well as the development of biofuels. Box 4.5 sets out some of
the issues related to the exploitation of tar sands. The issue of biofuels
has even greater implications for primary food production, and is
explored in a little more detail below. A final observation with regard to
Figure 4.6 is drawn from Colin Campbell, who reminds us not to be
mesmerised by the date of peak. What matters, he says, is to recognise
Box 4.5

Canada’s tar sands: dirty oil

An estimated 2 trillion barrels of oil exists in the form of tar sands – about
the same quantity as conventional oil – although considerably more effort is
needed to retrieve it. Much of the current production is taking place in the
Province of Alberta in Canada, creating what has been called a modern-day
gold rush. But there are no individual prospectors here, only consortia of
major energy companies with a stake in the world’s largest energy project.
The constituent of tar sands is bitumen, a dirty, heavy oil. There are two
principal methods of extraction: open-cast mining, which involves the
removal of up to 75 metres of overburden (soil, clay and sand) to reveal the
seam of oil sands around 50 metres in depth; or in situ extraction, which
accounts for 80 per cent of production, involving high-pressure injection of
steam or solvents such as naptha. Fire may also be used to heat the bitumen
and coax it to the surface.
Processing involves washing the material in hot water in order to separate
the bitumen from the sand; the bitumen is then processed further to make
crude oil. The polluted water used in processing – three barrels for every
barrel of bitumen – is then pumped into enormous tailing ponds, which
cover over 100 km2.
The environmental balance sheet of this industry does appear to be truly
daunting. The volumes of freshwater being extracted for processing are
causing serious concerns for the integrity of the hydrological resources of the
region. The tailing ponds are leaving a terrible environmental legacy, with
huge volumes of highly toxic wastewater leaching into groundwater and the
river systems. Propane cannons are used to deter migratory flocks of birds
from settling on the ponds, mistaking them for freshwater feeding grounds,
leading to almost immediate death. Reports on human health impacts in the
wider region, particularly on First Nation peoples who continue to hunt and
fish for food, appear deeply worrying. This development has become
Canada’s fastest-growing source of greenhouse gas emissions, with oil from
tar sands generating 20–30 per cent more greenhouse gas emissions than
from conventional oil. Meanwhile, 15–20 per cent of Canada’s annual
production of natural gas is used to provide the process energy. This has
been described as reverse alchemy: turning clean-burning natural gas into
dirty, carbon-heavy oil.

Sources: adapted from Leggett (2005); Environmental Integrity

Project (2008); Government of Alberta (2010);
Polaris Institute (2010).
 50

45

40

35
Gb oil equivalent

30

25

20

15

10

5
0
1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050
Regular oil Heavy etc Deepwater Polar NGL Gas Non-con gas

Figure 4.6 The depletion curve for oil and gas

Source: Colin Campbell, personal communication.

the relentless decline that follows the peak, representing a future of ever-
falling output of regular oil (and gas), which unconventional sources
will only partly and temporarily (and at significant cost) help to buffer.

Biofuels
In contrast to the development of non-conventional oil as a way of
extending the supply of petroleum, biofuels have been heralded as a
means to offset dependence upon petroleum. Both in Europe and in the
United States, there has been considerable interest in developing energy
sources from biomass, that is, plant material derived from
photosynthesis and processed to produce a gas, liquid or solid that can
be burnt to release useful energy, for example in an internal combustion
engine. While biofuel technologies pre-date the Second World War, the
low price, ubiquity and high energy density of oil products made them
uncompetitive – at least until the oil-price rises and supply restrictions
of the 1970s, when some countries began seriously to explore ways of
reducing their dependence on imported petroleum. One such country
 was Brazil, which built much of its early economic development around
the automobile, and which quickly began to develop an ethanol industry
derived from sugar cane that, by 2006, produced over 16 billion litres,
around 20 per cent of the country’s total fuel consumption (Shurtleff and
Burnett 2008). Today, all automobiles in Brazil are required to burn an
E25 blend of gasoline (75 per cent) and ethanol (25 per cent), and there
is a growing fleet of “flex-cars” that run on any mix up to E100 pure
ethanol fuel. The successful transition in implementing such a measure
and the enormous foreign exchange savings made have encouraged other
states to follow suit, and world ethanol production has grown
accordingly (Figure 4.7).
The United States has embarked upon a rapid expansion of its own
ethanol programme, building new refineries to process maize (corn)
into fuel. By 2008, 134 ethanol plants were in production compared
with 68 in 2003. In 2007, some 81 million tonnes (or 20 per cent of
the US maize harvest) was diverted to ethanol production, yielding

20,000

18,000

16,000

14,000

12,000
Million gallons

10,000

8,000

6,000

4,000

2,000

0
1975 1980 1985 1990 1995 2000 2005 2010

Figure 4.7 World annual fuel ethanol production, 1975–2009

Source: Earth Policy Institute, www.earth-policy.org.
6.5 billion US gallons (around 24 billion litres). The industry is well
ahead of federal targets – for example, the Energy Independence and
Security Act of 2007 requires 8 billion gallons of renewable fuels to
be blended into the country’s fuel supply in 2008 – and was set to
achieve an installed capacity of over 13 billion gallons in 2008.
As oil prices rise, the economics of ethanol manufacture become
increasingly attractive, made more so by government subsidies paid
to distillers.
In Brazil, ethanol is competitive with gasoline/petrol when oil prices are
at $35–40 per barrel. Brazil also has a major comparative advantage
insofar as sugar cane produces twice as much ethanol per hectare than
corn (up to 8000 litres/ha compared with 4000 for corn). Critically, the
energy balance – that is, the ratio of the energy obtained from ethanol to
the energy expended in its production (energy returned on energy
invested, EROEI) – is far superior in the case of sugar cane: up to ten
times, compared with up to 1.6 times for corn. Besides many other
advantages, perhaps the most critical factor of concern here is that sugar
cane is not a staple food and, although occupying arable land, is not
otherwise distorting market prices. Corn, on the other hand, is a major
food ingredient within the United States, and corn exports provide
consumers in distant markets with subsistence. As ethanol distilleries
agree contracts for their feedstock with farmers, they are competing
directly with those who would otherwise eat that corn (an issue to
which we return in chapter six).
In the case of the European Union, where production of ethanol is
modest compared with the USA and Brazil, there has been more interest
in developing biodiesel fuels using vegetable oils as feedstock. A variety
of agricultural crops have been harnessed, including those widely grown
in Europe (rapeseed, sunflower) and those imported from the tropics (oil
palm, coconut, peanut). Domestically grown soybean accounts for 80
per cent of US biodiesel production. Because of strong environmental
criticisms that have drawn attention to tropical deforestation in countries
that have sought to expand production of commodities such as oil palm
for European biodiesel, the EU has reconsidered its initial target of
achieving a 5.75 per cent share of all transport fuels with biofuels by
2010 and 10 per cent by 2020. Instead, it has set new targets encouraging
the development of waste, non-food crops and what are referred to as
second- and third-generation biofuels. Second-generation biofuels focus
upon non-food biomass crops, such as quick-growing cellulosic grasses
and woody shrubs (e.g. miscanthus and switchgrass) to
 produce ethanol, and jatropha for biodiesel. Third-generation biofuels,
which are clearly at an even earlier stage of research and development,
involve the use of algae. Future editions of this book may need to
devote space to algaculture as a source of food and energy.
Initially, there was some enthusiasm for agro-biofuels, which were
regarded as environmentally benign, offering opportunities to reduce
dependence upon foreign oil and thereby enhance energy security, and
even to improve farm incomes. More recent analysis, however, has been
guarded, if not critical, particularly with regard to claims that agro-
biofuels are at least carbon neutral. As discussed earlier in this chapter,
the conversion of natural vegetation cover to biofuel production is likely
to liberate larger amounts of CO2 locked up in that vegetation than can
be captured by the new crop, and the process may significantly reduce
the carbon-sequestration capacity of soils. Furthermore, far from
creating new opportunities for farm incomes, large-scale biofuel
cultivation is likely to further the elimination of small farmers in the
interests of efficiencies of scale.
Consequently, the challenge of peak oil requires an urgent evaluation of
our predicament, laced by an understanding that whatever alternative is
developed is unlikely to possess the ubiquity, flexibility and price of oil.
Such alternatives need to meet more stringent criteria than those that
seem to exist with regard to the development of tar sands: new energy
sources should not generate such enormous environmental impacts, nor
should they compromise global food production capacity. The challenge
of peak oil ought to present an opportunity to look more closely at our
assumptions in continuing the extravagant patterns of energy
consumption that we have enjoyed hitherto. In the realm of food, it
might give us pause to consider the degree of technological and
institutional lock-in to oil-based production, distribution, retailing and
consumption, and offer an opportunity to think more closely about our
dietary options. The consumption of livestock products would
necessarily be high on this list.

5. Livestock and the rising demand for meat

Livestock’s impact on the environment is already huge, and it is growing and

rapidly changing. Global demand for meat, milk and eggs is fast increasing,
driven by rising incomes, growing populations and urbanization.
(Steinfeld et al. 2006: 3)
Meat consumption
We are naturally omnivores, physiologically evolved and culturally
adapted to eating a wide range of foods, including those of animal
origin. It has been argued that animal foods – whether scavenged,
collected, hunted or produced from domesticated species – have been a
universal feature of human behaviour, with valuable nutritional and
social implications. An issue that is of growing contemporary concern,
however, is whether we are consuming animal products, particularly
meat, at levels that are harmful to human health and to the environment.
Livestock products provide a range of nutrients, including protein,
minerals such as iron and calcium, vitamins including B12, and fat.
However, nutritional research over the past two decades has generated
a body of scientific evidence that has demonstrated the adverse effects
of animal fats (related to cardiovascular disease) and animal protein
(related to cancer). The environmental consequences of intensive
livestock-production systems are also significant and wide-ranging,
with meat production, for example, estimated to account for 18 per
cent of global warming, an effect greater than the transport sector
worldwide.
Pimentel and Pimentel (2003) have asserted that an estimated 2 billion
people worldwide live on a meat-based diet, while an estimated 4 billion
people live primarily on a plant-based diet. Yet the past decade has
witnessed a major expansion of meat eating such that, today, the
Pimentels’ ratio might well be in balance. This is a consequence of
rising levels of disposable income as well as dietary change that arises
from people moving to the cities. It has long been recognised that, at
low levels of income, indigenous staples such as cereals and tubers
dominate food intake. Beyond a certain income threshold, however,
there is a marked increase in meat consumption as rising disposable
income allows for greater dietary diversification and people opt for
higher sources of protein. With a marked upturn in economic fortunes in
China, India, some other Asian and Latin American countries, meat
consumption has grown rapidly. This “nutrition transition”, as it has
been termed, occurring within a single generation, has seen shifts in
some parts of the world from situations of widespread undernutrition
to aspects of overnutrition. This is an issue to which we return in
chapter six.
Table 4.2 provides a summary of meat consumption per capita by
region. It shows the sharp increase in meat consumption amongst the
 middle-income group of countries, which are those experiencing the
steepest improvements in household income. Here, meat intake has
grown from around 16 kg per person in 1970 to 46 kg per person in
2002. This is in sharp contrast to the low-income countries, which as a
group have registered a barely noticeable improvement, with a rise from
6.7 to 8.8 kg between 1970 and 2002. One country that has moved
between these economic categories during this period, and that can
demonstrate a level of meat consumption to match, is China. Although
Keyzer et al. (2005) note some discrepancies in official Chinese
statistical reporting of consumption levels, it is clear that the trend in
meat eating is sharply upward, rising from around 9 kg per person in
1970 to a reputed 52 kg in 2002, with pork comprising the most
significant element. If the FAO data are approximately correct, not only
did meat consumption in China double in little more than a decade
(from 26 kg per capita in 1990), but by 2005, China consumed more
meat than the entire population of the world in 1961 (Weiss 2007).
Amongst the high-income countries, on the other hand, meat
consumption has risen more slowly but steadily, although there is some
considerable variation between countries, even within Europe. While the
UK recorded a level of per capita meat intake of around 80 kg in 2002,
in Denmark the figure was 146 kg – exceeding even the USA’s
consumption of 125 kg.

Table 4.2 Meat consumption per capita by region (kg per person per year)

Region/class 2002 1990 1980

Asia 28 17 11

Europe 74 n/a n/a

North America 123 111 107

South America 70 50 47

Sub-Saharan Africa 13 14 15

High-income countries 94 85 79

Low-income countries 9 8 7

Middle-income countries 46 29 22

Source: World Resources Institute, http://earthtrends.wri.org/searchable_db (figures rounded by the author;

n/a, not available).
Livestock systems
Large-scale and intensive livestock production really began to take off
from the 1950s. Prior to this, the raising of animals was much more
closely tied to the availability of local resources and a degree of
multifunctionality at farm level, such as the grazing of crop residues in
fields as part of rotations in order to avail of the manure, the use of pigs to
root and break up the sod after a long fallow, or the conversion of organic
wastes to meat and eggs. The herding of ruminants (cows, sheep, goats)
also involved local daily or more long-distance seasonal movements in
order to take advantage of different grazing resources; for example,
patterns of transhumance in mountainous regions with cattle and sheep
taken to the high summer pastures (where their milk was turned into
butter or cheese), or long-distance pastoralism, a particular feature of
nomadic cultures, which proved an effective evolutionary adaptation to
the sparse grazing of arid environments. In all these cases, livestock
products were central to human survival, with milk and its fermented
derivatives, blood and meat contributing variously to the human diet.
Although grazing land takes up over one-quarter of the ice-free terrestrial
surface of the planet, livestock also account for around one- third of
arable land that is given over to the production of animal feeds. The
expansion of pasture in regions where there was no previous history of
livestock grazing has seen considerable environmental destruction,
perhaps the most controversial being the level of deforestation in Amazonia
and Central America. During the 1970s, the World Bank substantially
increased its loans to Central American countries for agricultural and
rural development, ostensibly to improve basic human needs and poverty
alleviation. But the majority of loans went to support the production of
beef for export (largely for US hamburgers), creating precious few jobs
and limited livelihood opportunities for the rural poor. In Brazil, the
push into Amazonia – though now joined by the expansion of soybean –
was initially driven by ranchers using cattle to displace small-scale farmers
and to control, as cheaply as possible, large areas of land. Consequently,
although it is easy to represent cattle as primary instruments of
environmental destruction, they are really ciphers of more powerful
economic and political drivers of change. Box 4.6 reminds us that cattle
culture has deep historical roots, if a poor modern-day reputation.
According to FAO figures, there are an estimated 1.3 billion cattle in
the world, with the greatest numbers in India, Brazil and across
Sub-Saharan Africa. A full breakdown of livestock numbers by world
region is provided in Table 4.3. If the South accounts for three-quarters
of the global cattle herd, it has a lesser share in the total number of pigs
and poultry, which are increasingly produced in intensive operations,
and a slightly higher share in small ruminants (sheep, goats, camelids),
which are most commonly grazed more extensively. The bald figures
can disguise enormous differences in systems of production and the
economies they support.
For the cattle-herding peoples of East, Southern and Sahelian Africa, as
well as for pastoralists elsewhere, grazing ruminant animals across dry
and challenging environments is an effective means of survival,
provided the herds and flocks move regularly between pasturelands and
waterholes. The use of established migratory routes ensures that grazing

Box 4.6

Bovine culture
In an impassioned book, Jeremy Rifkin traces the deep-seated significance of
the bull and the cow in human culture: from its first domestication in
Mesopotamia and its yoking to a plough, the bull became a hugely important
symbol of divinity throughout the Mediterranean (consider, for example, the
Minoan civilisation). Bovine culture then spread, via the Nile Valley, from
Egypt deep into Africa, creating pastoralist societies from the east to the
very south of the continent. Meanwhile, the eastern migration of Eurasian
nomads brought the cow to India, where it became, over 2000 years, a
sacred animal, a goddess of abundance. This is in sharp contrast to Spain,
where it is the machismo of the bull that is celebrated, at least by part of the
population, through the sacrificial rituals of the bullring. The conquest of the
Americas also brought its “cattlising”, with both conquistadors and catholic
priests introducing the animals to new lands from the Pampas of Argentina
and Uruguay through large swathes of Venezuela and Mexico to Texas and
Florida. The extermination of the buffalo across the Great Plains of the USA
by 1870, as well as the First Peoples who lived alongside them, opened up
the range to cattle ranching. With the railway providing their transport to the
stockyards of Chicago, and with the development of refrigerated shipping by
the 1880s, American beef found its way onto European tables.
Cattle have consequently played a critical role in changing the ecology of
many habitats around the world, especially in the Americas. It is little
wonder that, for Rifkin, the creation of the modern cattle complex is a truly
malevolent force, responsible for a major desecration of the world’s
ecosystems.

Source: adapted from Rifkin (1992).

Table 4.3 Population of the principal commercial livestock categories by world region

Region Population (million head)

Cattle Pigs Small Poultry

ruminants

North America 111 73 9 2,059

Latin America and Caribbean 358 77 116 2,256
Brazil 177 32 24 878
Europe 101 165 142 1,329
Commonwealth of Independent
States 58 31 60 558
West Asia and North Africa 32 0.7 227 1,263
Sub-Saharan Africa 213 20 370 862
South Asia 246 15 299 701
India 191 14 181 377
East and South-East Asia 153 529 346 5,995
China 104 452 289 3,830
Oceania incl. Australia 65 8 266 199

Total: North 327 285 400 4,519

Total: South 984 632 1,322 10,628

Total: world 1,311 918 1,722 15,147

Source: assembled from data in Steinfeld et al. (2006).

resources are used in a sustainable way, allowing for their full recovery
between visits. With the end of colonial rule came the creation of states
demarcated by borders drawn across traditional migration routes. The
imposition of head taxes, the introduction of the monetary economy that
led to the breakdown of traditional bartering arrangements, and efforts to
sedentarise nomadic pastoralists all contributed to the creation of an
emerging ecological problem, which was made worse by the drilling of
wells to create permanent waterholes. As herds grew in these settled
areas, overgrazing has led to significant problems of desertification. In
the meantime, pastoralists, for whom their animals represent their sole
assets as well as their cultural identity, encroach onto the lands of
farmers in their search for grazing, contributing to escalating tensions
and conflict in a context of resource constraints.

Factory farming
If the arid regions of Sub-Saharan Africa, where thin cattle browse on
thorny shrubs, illustrate the timeless relationship between animals and
the availability of local resources, the past half-century has seen the
severing of this link in the development of modern livestock production.
With the emergence of grain feeding of animals, beginning in the USA
in the 1950s and spreading quickly to Europe, the Soviet Union and
Japan in the 1960s, suddenly there were no local resource constraints to
the number of animals that could be raised. Although this has had an
utterly transformative effect on the production of monogastrics (pigs and
poultry), it has also had a large impact on the cattle sector in the USA.
In contrast to the scarcity of grazing in Sub-Saharan Africa, the
overcrowded feeding lots of Kansas and Colorado, where cattle are
fattened on a diet of corn and supplements, makes us appreciate the
immensely adaptable nature of the bovine. Box 4.7 draws from Michael
Pollan’s account of the feedlot system, and describes the way in which
up to half of all beef is produced in the United States today.
But the rise of feedlot production for beef has been outstripped by
developments that have transformed the “farming” of pigs and poultry.
High stocking densities, confinement practices and measures aimed at
“speeding up” the growth cycle of animals have become standard
procedures for those production units officially known in the United
States as CAFOs (confined, or concentrated, animal feeding operations),
elsewhere colloquially as “factory farms”. In recent years, organisations
concerned with animal welfare (e.g. Compassion in World Farming)
have worked hard to bring to wider public attention the conditions under
which poultry (layers, broilers) and animals (pigs, veal calves) are kept.
But the intensification of livestock production, driven by the imperative
of reducing costs and thereby making meat and dairy ever cheaper, has
brought little sustained public opprobrium. Most of the criticism at such
production practices occurs only episodically at times of outbreak of a
zoonotic disease (one that is capable of being passed from animals to
humans).
Modern animal production did not set out to be cruel or wasteful, but to
be efficient, to get more output for less input. Above all, modern
methods were designed to reduce labour, with other inputs either freely
available (air, water) or relatively cheap (energy, feeds) (Gussow 1994).
As part of the process of intensification and specialisation described in
chapter two, farms began to lose the small-scale and multifunctional
aspects of animal husbandry in which a few pigs and small flock of hens
were an almost ubiquitous feature of the average European or North
American farmyard. Gradually, a combination of market incentives and
public policy regulation (on the grounds, ironically, of animal health and
food safety) encouraged specialised and large-scale production in
Box 4.7

Feedlot beef
Born on a ranch on the Great Plains, a calf spends its first six months eating
grass, which is what it is physiologically evolved to do. The animal’s rumen,
the first of four compartments in a bovine stomach, contains powerful
bacteria capable of breaking down and digesting the cellulosic material that
is utterly indigestible to humans. Growing meat on grass makes ecological
sense, particularly in places like South Dakota, where the land is incapable
of growing row crops without large amounts of irrigation and chemicals.
Cattle that feed on pasture, or on crop residues, or by-products of food
processing are able to metabolise such material and therefore do not compete
with humans directly for food. However, calves born on the grasslands are
introduced to maize (corn) at six or seven months, and by nine or ten months
are part of the huge bovine population that comprises the American feedlot
beef production system.
Consequently, feedlot cattle – or at least the specific steer which Pollan
bought on a ranch and followed through the system – are given a mixed diet
of around 15 kg/day, consisting of corn, liquefied fat (possibly rendered from
animal carcasses), protein supplement (comprising molasses and urea), silage
and antibiotics. The latter are critical in keeping the animal healthy – or, at
least, healthy enough until its slaughter after five months – as the diet puts
enormous strain on the animal’s capacity to digest a diet it was never
designed to consume. Yet, while it converts this daily intake into 1.5 kg of
body weight, the greater part of this intake is converted into a waste stream,
some of which is drained off into lagoons, some of which simply finds its
way down into the groundwater.
Pollan makes clear the connectedness of the feedlot: back to the fields that
provide the bulk of the cattle feed, to the vast monoculture of corn that is fed
by “a steady rain of pesticide and fertiliser”; and, following this further
upstream, to the oil fields of the Persian gulf. Downstream of the cattle
feedlot is, metaphorically and physically, the Mississippi Delta and the Gulf
of Mexico, where 5800 square miles (according to NASA) of coastal waters
have been rendered lifeless because of the excessive nutrient loading that
derives from fields and feedlots.
Pollan’s calculation was that his purchased steer consumed nearly a barrel of
oil during its life of 15 months, enabling it to grow from a 36 kg calf to a
beast of 545 kg. Though inefficient, bovine metabolism has been transformed
into another dimension of food industrialisation driven by fossil fuels.

Sources: adapted from Pollan (2006); NASA (2009).

 dedicated units. Such production systems lent themselves to new
opportunities for agro-industrial appropriationism, supplying dedicated
feeds, medicines, housing, waste-management equipment, and scientific
expertise in producing higher-yielding and faster-maturing breeds. Little
wonder that some of the world’s largest grain companies (Cargill,
ConAgra) began to develop strong commercial interests in livestock
production. Recalling the quotation from Page (1997) in chapter two
regarding hog production in the USA, the diffusion of capital-intensive
techniques resulted in a dramatic reduction in the number of farms
raising pigs, and a sharp increase in the average number of pigs per
farm. Almost everywhere that dedicated pig units are created, the
numbers of animals are no longer counted in their tens or dozens, but in
their hundreds or thousands. Consequently, factory farms have come to
account for around 40 per cent of global meat production by volume,
with around three-quarters of the world’s poultry, 68 per cent of egg
production, and about half of the world’s pig meat produced in confined
feeding operations (Weiss 2007).
As the concentration and stocking density of factory-farmed animals has
increased, one can reflect that these creatures seem to have lost their
sentient character to become much like machines housed in rural
factories. Animal welfare organisations have drawn attention to
confinement practices that include the use of cages for battery hens and
pens for nursing sows that do not even allow sufficient room to turn; and
have highlighted common surgical procedures including debeaking,
docking tails, clipping piglets’ teeth, and castration without anaesthetic.
Managing reproduction in pigs can involve surgically relocating a boar’s
penis so that it can identify sows in a fertile state but allows the semen to
be collected by a handler, which can then be carefully allocated in order
to artificially inseminate twenty sows. Above all, the extraordinary effort
to drive down costs of production in order that meat can become an ever-
cheaper food staple has led to a high-throughput, high-speed system. All
of this in order that pork (and chicken, for parallel practices occur in the
poultry industry) is available at ever-cheaper prices. Box 4.8 provides
some further details of intensive pig production in the USA.

Environmental impacts
Given the global spread of intensive livestock production systems
capable of producing the high volumes of meat demanded by societies
experiencing improved economic prosperity, it is clear that animal
 Box 4.8

A pig’s life
More than 90 per cent of pigs raised for meat today are raised indoors in
crowded pens of concrete and steel. They never get to go outside or root
around in pasture, and don’t even have straw to bed down in. The most
tightly confined of all are the breeding sows. Under the factory’s rigid
production schedule, they are made to produce litter after litter as quickly as
possible, which means that they are pregnant for most of their lives. During
their pregnancies, which last about 16 weeks, most American sows are
confined in “gestation crates” – steel-barred crates or stalls just a foot longer
than their bodies, and so narrow that the sows cannot even turn around. Of
the 1.8 million sows used for breeding by America’s ten biggest pig
producers, about 90 per cent are kept in this manner.
When the time comes to give birth, they are also confined in a farrowing
crate. [This] keeps the sow in position, with her teats always exposed to her
piglets. She is unable to roll over – and this, the defenders of the crate say,
ensures that she will not roll on top of, and perhaps smother, her piglets.
In Europe, widespread public concern about the close confinement of sows
led to the European Union asking its scientific veterinary advisory
committee to investigate the impact of gestation crates on the welfare of
sows. Pigs like to forage and explore their environment. In a stall, they
cannot, and after a time they become inactive and unresponsive, or carry out
meaningless, repetitive motions that are signs of stress. The scientific
committee concluded, “sows should preferably be kept in groups”. The EU
is now phasing out sow crates by the end of 2012, and requiring that sows
be given straw that they can play around with, to reduce the stress of
boredom. Even before the new law comes into effect, Britain and Sweden
acted to ban sow stalls. All of the 600,000 breeding sows in Britain now
have, at least, room to turn around and can interact with other pigs.

Source: adapted from Singer and Mason (2006), 45–7.

numbers have long outstripped the capacity to feed them on locally

available crop residues, by-products and food waste. Instead, increasing
volumes of cereals and oil seeds are produced for dedicated use in the
animal feeds sector, and there are growing concerns that this represents
a source of considerable uncertainty for long-term human food security.
In 2002, the livestock industry accounted for 670 mt of grains, over
200 mt of processed oil seeds and pulses, and 150 mt of roots and tubers
(Steinfeld et al. 2006). The particular combination of feed concentrates
depends naturally upon the livestock, but also upon the comparative cost
advantages within different regions. For example, maize is the dominant
cereal used in animal feeds in the USA, Brazil and China, while wheat
and barley prevail in Canada and Europe. According to Pimentel and
Pimentel (2003), US livestock consume more than seven times as much
grain as is consumed directly by the entire American population;
the amount of grains fed to US livestock is sufficient to feed about
840 million people following a plant-based diet. Grain fed to livestock
as a proportion of the total grain consumed now accounts for some
60 per cent of the total grain consumption of the USA, 69 per cent in
Australia, and 73 per cent in Canada.
Soybean meal is a ubiquitous element of pig and poultry rations
globally, and demand for it over the past four decades has increased at a
faster rate than total meat production, suggesting a net increase in its use
per unit of meat. It was noted above how soybean was strongly
associated with deforestation in Amazonia and how quickly output has
increased, with Brazil, Argentina and the USA today accounting for
about 80 per cent of global production. This places importing countries
at something of a disadvantage when it comes to specifying standards.
For example, the UK imports about 90 per cent of its 3 mt of soymeal
from Brazil and Argentina. However, these two countries are well
advanced in converting their soybean production to genetically modified
(GM) varieties, while the UK, as part of EU policy, seeks to import only
non-GM soy. The processing of soybeans, it should be said, gives rise to
about 19 per cent oil, which can be used elsewhere in the food industry,
and 74 per cent meal, which is used for animal feed.
While it has been argued that feeding grains to livestock represents a
buffer through which grain surpluses can be utilised, thus supporting
farm-gate prices, the more convincing argument points to the spread and
intensification of cereal and oil seed farming to produce low-cost inputs
for the feedstock industry. The bottom line is that the conversion of
plant to animal protein is inefficient and involves a significant net loss.
According to Steinfeld et al. (2006), livestock consume 77 million
tonnes (mt) of protein in feedstuffs that could potentially be used for
human nutrition, whereas the products that livestock supply contain only
58 mt of protein. In dietary energy (calorie) terms, the net loss is even
higher, particularly when calculations include all the fossil energy inputs
into the cultivation of feed crops. According to Pimentel and Pimentel
(2008), chicken broiler production is the most efficient, with an input of
4 kcal of fossil energy per 1 kcal of protein produced. Thereafter,
input–output ratios for other livestock products produced from
concentrates deteriorate quickly, with pork at 14:1, eggs at 39:1 and beef
at 40:1. As they argue, if animals were fed on good-quality forage rather
than grains, the energy inputs could be reduced by half.
In a context of rising global demand for meat, there are consequently
serious anxieties about how the increased demand for feedstuffs –
variously estimated at between 1 and 1.9 billion t over 1999 levels by
2030 – will be met while also providing for an extra 2 billion more
people. There is also concern about the various waste streams derived
from intensive livestock operations. With livestock estimated to
contribute around 18 per cent of the global warming effect – principally
derived from a 37 per cent share of methane emissions and 65 per cent
of nitrous oxide – the link to climate change is well established.
Regionally and locally, livestock wastes are a major source of ammonia
and create considerable damage to air quality, seriously affecting the
health and wellbeing of residents in proximity to feeding operations and
slaughter facilities. Water pollution, too, is strongly associated with the
inadequate management of animal manure, of which an estimated
500 mt was produced by feeding operations in 2003. As Michael
Pollan reminds us, one of the striking things that any intensive animal-
feeding operation does is to take the elegant ecological solution of
livestock manure applied to fields for fertilisation and neatly divide it
into two new problems: a fertility problem on the farm (which is
resolved with chemical fertilisers), and a pollution problem arising from
the livestock.
It is the scale of this pollution problem that has become a serious public
health concern. The reasons why this waste is no longer valuable
nutrient material that can be returned to the soil to perform a fertilisation
function are as follows:
l Because of the high level of antibiotics routinely used as
prophylactics to keep animals from falling ill, there are high levels
of pharmaceutical residues in the waste stream.
l This has also led to the development of antibiotic-resistant
pathogens that can prove very dangerous to human health, as normal
broad-spectrum antibiotics (e.g. tetracyclines) are no longer
effective.
l The widespread use of growth hormones to speed weight gain in
animals also creates a residue problem in the wider environment
and, together with certain forms of pesticides, has created hormonal
changes and mutations in aquatic life such as frogs and fish.
 l The waste stream can contain zoonotic pathogens (parasites,
bacteria, viruses) that can be harmful to human health.
l The waste stream comprises effluents high in ammonia and other
airborne pollutants that are a nuisance (odours), can contribute to
acidification of terrestrial ecosystems, and are greenhouse gases.
Liquid wastes high in nitrates can contaminate groundwater and
reduce oxygen in surface waters, resulting in eutrophication; they
also contain higher levels of heavy metals and phosphorus, which
bio-accumulate from fertilisers and feeding supplements.
It is little wonder, then, that there should be growing environmental
concerns about the meatification of global diets (Weiss 2007). The
spread of factory farms across large parts of Asia and Latin America
(having already become the norm across North America and Europe),
designed to satisfy the rising urban demand of newly affluent
households for meat, brings increasing pressure on land to produce
feeds, and on the wider environment to cope with waste streams.
However, drawing attention to such predicaments is not to make a moral
claim for vegetarianism. We remain omnivores physiologically adapted
to meat consumption; the question is whether the carnivorous appetites
promoted by the food industry have distorted the balance in our food
intake. Seeking an answer to this question is something to which we
return in chapter seven.

6. Summary

This chapter sets out four major challenges for primary food production
as we look toward the future. Climate change, freshwater depletion,
“peak oil” and rising demand for meat represent the key issues that need
to be addressed – and with some urgency – if we are to ensure stability
and security of the global food system in the medium term. Dealing at
some length with these four separate challenges is not to suggest that
other issues are not also important. Maintaining the world’s stock of
plant and animal genetic diversity, comprising those varieties that have
been carefully bred to produce food, as well as their wild relatives and
the wider stock of biological resources, might for many represent the
most pressing single issue that we face. Others would argue that the
scale of land-use/land-cover change, particularly the conversion of
tropical moist forests to arable farming or livestock grazing, represents
the single most immediate problem.
 The four issues identified here do not represent the only challenges that
we face – and they cannot be resolved in isolation. In this regard, they
help to demonstrate the need to embark upon a new course for primary
food production that is less energy-intensive, more efficient in its use of
water and other resources, and that results in lower greenhouse gas
emissions and other forms of pollution. In short, primary food
production in all its manifest forms around the world must endeavour to
work within sustainable parameters. Chapter seven briefly sketches out
what these might constitute. In the meantime, we need to examine the
environmental consequences of the food system beyond primary
production – the focus of chapter five.

Glossary
Agri-commodities Agricultural products produced for the global market and
whose value is established by the parameters of supply, demand and
speculation.
Agri-food system Encompasses all those activities related to the production
(cultivation, rearing, capture); processing (refining, manufacturing);
distribution (transportation); sale/retail, preparation; and consumption of food.
The food system embraces a more holistic and dynamic understanding,
recognising complex relationships between different components, including
feedback loops from consumption that can reshape aspects of production. In
this it possesses a superior explanatory capability to the term food chain.
Agro-ecosystem Represents human manipulation and alteration of a natural
ecosystem for the purpose of establishing agricultural production. Ecosystem
properties, such as energy flows and nutrient cycling, are harnessed in such a
way as to optimise productivity of the desired crop(s).
Alternative food networks Widely used term that embraces a variety of new
arrangements linking together producers, consumers and other
actors in more fluid and loosely bound sets of relations that together
represent an alternative to more standardised food supply chains.
Appropriationism A process whereby activities and resources previously
under the control of farmers and local artisans (manufacture of farm tools and
mechanical aids, soil fertilisers and seeds) become the basis of an autonomous
sector of innovation and control, leading to the accumulation of value by
industrial interests at the expense of farmers.
Consumerism The fostering of desire to acquire material goods.
Distanciation A term possessing several different philosophical meanings, it
is used here in a more prosaic sense to convey the way in which a
perceptual separation is established between the realm of production and the
site of consumption, such that it conceals the circumstances in which food
may be produced.
Entitlements Can be regarded as rights of ownership or use to the
products of land or labour, which determine people’s access to food.
For example, a farmer growing her own food is entitled to what she has
grown (less any obligations to others). The entitlement of a labourer
working for wages is constituted by what he can buy with those wages;
in the absence of employment such entitlements collapse, leaving
labouring families vulnerable to hunger.

Food chain A term commonly used by agri-food policy makers and

business analysts to refer to the succession of stages from primary
production to consumer purchases of food. Unlike a food systems
approach, food chains tend toward a linear representation of discrete
sequential stages that are frequently labelled as “farm to fork” or “field
to plate”.

Food security Conventionally, “food security exists when all people, at

all times, have physical and economic access to sufficient, safe and
nutritious food to meet their dietary needs and food preferences for an
active and healthy life” (FAO 1997). Alternatively, community food
security exists when “all residents obtain a safe, culturally acceptable,
nutritionally adequate diet through a sustainable food system that
maximizes community self-reliance, social justice, and democratic
decision-making” (Winne n.d.).

Food sovereignty The right of peoples and sovereign states to

democratically determine their own food and agricultural policies.

Global warming potential GWP is a measure of the ability of a

greenhouse gas to trap heat in the atmosphere. This measure combines
its efficiency to absorb and radiate heat, and its lifespan in the
atmosphere. The GWP of products is calculated on the basis of the total
greenhouse gas emissions arising from their production, use and
disposal.

Horizontal consolidation A process whereby a company merges with

or acquires another within the same sector, thus increasing their share of
that particular market segment.

Productivist/productivism An approach to agricultural production that

emphasises output or yield per unit area and effectively disregards all
other considerations, such as the long-term consequences of resource
depletion or social justice. Productivism represents the polar opposite to
sustainability.
Relocalisation A vital aspect of alternative and sustainable food
networks that seek to reconnect with the place of production, in contrast
to the perceived “placelessness” of globalised food “from nowhere”.

Substitutionism The capability developed by food manufacturers to

interchange different raw materials as ingredients for their final food
products, including the displacement of primary foods by industrial
(synthetic) substitutes.
Sustainable consumption The use of goods and services for the
satisfaction of basic needs and to enhance the quality of life, while
minimising the use of natural resources and toxic materials as well as
the emissions of waste and pollutants over the life cycle.

Sustainable food system Combining a concern with the present and

future generations, an SFS is one that provides sufficient healthy food to
meet current needs while ensuring the integrity of agro-ecosystems and
the wider environment to meet the food needs of generations to come. It
fosters greater local production and distribution in order to ensure that
nutritious food is available, accessible and affordable to all. An SFS is
built on humane practices, building equitable and just trading
partnerships with distant producers, protecting farmers, workers and
consumers.

Vertical integration A strategy of agri-food companies to develop

commercial interests upstream and/or downstream of their existing core
competence, usually by acquisition (take-over) of businesses in that area.
http://www.oda-alc.org/biblioteca/