80 Pedogenesis Review

See
discussions, stats, and author profiles for this publication at:

https://www.researchgate.net/publication/231582005
Quantifying Processes of
Pedogenesis
Article in Advances in Agronomy December 2011
DOI: 10.1016/B978-0-12-386473-4.00006-3
CITATIONS
READS
11
885
3 authors, including:
Budiman Minasny
University of Sydney
342 PUBLICATIONS 6,929
CITATIONS
SEE PROFILE
Alex B Mcbratney
University of Sydney
516 PUBLICATIONS 16,404
CITATIONS
SEE PROFILE
Available from: Budiman Minasny

Retrieved on: 23 September 2016
C H A P T E R
O N E
Quantifying Processes of
Pedogenesis
Uta Stockmann, Budiman Minasny, and Alex. B. McBratney
Contents
1. Introduction
2. Conceptual Models of Soil FormationFactors, Processes,
Pathways, Energy
2.1. Factors
2.2. Processes
2.3. Pathways
2.4. Energy
2.5. Summary
3. Soil Weathering and Production
3.1. Production of soil from parent materials
3.2. Chemical weathering of bedrock to soil
3.3. Summary
4. Soil Mixing: Vertical and Lateral Movements
4.1. Bioturbation
4.2. Soil creep
4.3. Rainsplash
4.4. Modeling pedoturbation
4.5. Summary
5. Models of Soil Formation Based on the Concept of Mass Balance
5.1. Landscape evolution models
5.2. Modeling soil formation in the landscape
5.3. Summary
6. Conclusions
Acknowledgments
References
Faculty of Agriculture, Food

New South Wales, Australia
and
Natural
Resources,
Advances in Agronomy, Volume 113

ISSN 0065-2113, DOI: 10.1016/B978-0-12-386473-4.00006-3
The
University
of
Sydney,
#
2
5
5
8
9
10
12
13
13
18
22
23
25
30
31
31
32
34
34
36
39
42
44
65
Sydney,
2011 Elsevier Inc.

All rights reserved.
Uta Stockmann et al.
Abstract
Our knowledge of plant and animal growth and development is far superior to
that of the evolution of soil, yet soil plays a fundamental role in natural
ecosystems. To understand the complexity of soil systems we need to explore
processes that lead to its formation. Research in pedogenesis has been focused
on formalizing soil-forming factors and processes to ultimately model soil
formation in the landscape. Early models described soil formation qualitatively
and were mostly limited to a description of soil evolution in the landscape. They
led to the development of qualitative models of pedogenesis based on empirical observations and later to quantitative models of pedogenesis based on
empirical equations or detailed differential equations derived from fundamental
physics. This review highlights the main models of pedogenesis and focuses on
models and rates of pedogenic processes such as the production of soil from
weathering of parent materials, and vertical and lateral movements in the soil
profile. It will become clear that field and laboratory work is needed to improve
and validate quantitative models of pedogenesis. In order to estimate and verify
model parameters, it is therefore of importance to collect real-world data.
1. Introduction
The following review will present and discuss various models of
pedogenesis. Since comprehensive reviews on models of soil formation
have been presented by Hoosbeek and Bryant (1992), Amundson (2004),
Schaetzel and Anderson (2005), and Minasny et al. (2008), here, however,
we only summarize the main models and focus on models of soil-formation
processes such as the weathering of parent materials and soil mixing.
Soil is a very complex system composed of a variety of interconnected
physical, biological, and chemical processes. It exists at the interface of the
atmosphere, biosphere, hydrosphere, and lithosphere. The interface or zone
where soil-formation processes take place has become known recently as
the critical zone (Brantley et al., 2007), where rocks meet life (see Fig. 1 and
Box 1). Here, soil weathering, soil mixing, and soil erosion processes occur
over several time scales, from the colloid (mm), grain (mm), soil horizon
(cm) and soil profile (m) scales to the landscape (km), and global (Mm) scale.
The importance of soil is also reflected in the recent US National
Research Council publication Landscapes on the Edge: New Horizons
for Research on Earths Surface (NAS, 2010) that addresses the challenges
and opportunities in Earth surface processes. The Earths surface is defined
as a dynamic interface where physical, chemical, biological, and human
processes cause and are affected by forcings in the Earth system. The book
clearly states the importance of pedogenesis: Soil formation is not, however, only of academic interest. Our food comes from plants grown in soil.
Quantifying Processes of Pedogenesis
La
ter
al
so
il t
ran
Soil
sp
Bedrock
Particle trajectory
Critical zone
Regolith
Water flow paths
or
Bioturbation
Erosion
Soil
production
Lowering/breakdown
of parent material
Material
Process
Soil
(material with horizons)
Physical, chemical,
biological, and transport
Soil creep
Eluviation
Illuviation
Processes
Soil production
Regolith
(parent material for soil
Weathering of underlying
production/form of
parent material
weathered friable rock with
structural characteristics and
Biological
fresh primary minerals of
parent rock)
Weathering
front advance Bedrock
Chemical
Physical
Uplift
Figure 1
The critical zone (based on Anderson et al., 2007; Graham et al., 2010).
The rapid rate of soil erosion due to land use relative to the slow rate of
transformation of rock into soil endangers soil resources worldwide. The
fate of soils, the base of agriculture, is of great concern.
To understand the complexity of the soil system it is important to
investigate soil-formation processes quantitatively. Ultimately, quantifying
pedogenesis should give us answers to questions such as
(1) How does soil form?
(2) At what rate does soil evolve over time? and
(3) How fast are rates of soil turnover occurring in the soil profile and
what influence do they have on pedogenesis?
Over the years, soil scientists have formalized concepts and models of
soil formation to improve our knowledge of pedogenesis. Based on the
degree of computation, these models describe soil formation qualitatively
and/or quantitatively. Further, based on the complexity of the structure
used in the models, they describe soil formation empirically (functionally) or
mechanistically. Generally, functional models are limited to a description of
pedogenic factors, but can also be based on empirical equations, whereas
mechanistic models are based on the mechanisms that can be formulated as
mathematical equations (Hoosbeek and Bryant, 1992).
Early models of soil formation were limited to a description of soil
evolution in the landscape or were based on simple empirical equations.
BOX 1 The critical zone
The term critical zone is used quite extensively in recent literature of Earth
Sciences. The critical zone is defined as the external terrestrial layer extending from the outer limits of vegetation down to and including the zone of
groundwater, which sustains most terrestrial life on the planet (Brantley
et al., 2006; Fig. 1).
The critical zone is described as the zone where chemical, biological,
physical, and geological processes are combined to control the development
of soils and ecosystems. It is known as a complex mixture of air, water, biota,
organic matter, and earth materials (Brantley et al., 2007). Within the critical
zone, a weathering engine transforms bedrock and biomass into soil, the
living skin of the Earth (Anderson et al., 2007). The weathering engine is
driven by physical and chemical weathering processes that fracture, grind, and
dissolve the bedrock; and biological weathering and turbation processes
(Anderson et al., 2007). Within the critical zone, soil acts as an open system
that is subject to elemental gain and losses. Studying this central component of
the critical zone is imperative, since knowledge of soils is still limited despite
their fundamental importance (Brantley et al., 2007). Rates of soil production
and loss, bedrock or outcrop weathering, and erosion have been estimated in
the literature, but a comparison is often challenging (Brantley et al., 2007). For
instance, for undisturbed forested landscapes rates of soil production and soil
loss are assumed to be balanced within the critical zone, varying between 7
and 80 mm per 100 years (0.07 and 0.8 mm yr 1). In contrast, weathering
rates estimated from field data for the transformation from bedrock to regolith
range between 0.05 and 10 mm per 100 years (0.00050.1 mm yr 1).
At present, interdisciplinary research is focusing on exploring how chemical, physical, and biological processes work together within the weathering
engine. In 2006, a working group was formed, the so-called Critical Zone
Exploration Network (www.czen.org), to emphasize the demand in integrating new tools to estimate the processes within the critical zone from field data
and therefore to answer process-orientated research questions (Brantley et al.,
2006). Questions related to the formation of soil as a major component of the
critical zone that need to be explored and answered are (cited from Brantley
et al., 2006): (1) What controls the thickness of the critical zone? Research is
focused on how fast and deep weathering of fresh bedrock occurs and what kind
of agents are involved, and therefore (2) What controls the rate of chemical and
physical weathering? (3) What controls the vertical structure and heterogeneity
of the critical zone? Here, research is focused on the mechanisms that ultimately
lead to a certain soil type and produce individual soil horizons.
In 2007, volume 3 (Number 5) of the magazine Elements explored and
reviewed the interdisciplinary knowledge and future research on The Critical
Zone focusing on its physical and chemical controls and biogeochemical
agents and expressing the need to study the human imprint on the critical
zone over the past 250 years (Amundson et al., 2007; Anderson et al., 2007).
However, there has been a shift of interest toward mechanistic modeling.

Mechanistic models of soil formation implement soil-forming processes to
describe soil formation quantitatively. They require a detailed understanding
of pedogenic processes. For instance, Bockheim and Gennadiyev (2000)
identified a total of 17 soil-forming processes that lead to the formation of a
soil profile. These processes include argilluviation, biological enrichment of
base cations, andisolization, paludization, gleization, melanization, ferrallitization, podzolization, base cation leaching, vertization, cryoturbation, salinization, calcification, solonization, solodization, silification, and anthrosolization.
While there are models that simulate individual processes, no mechanistic
model can be found in the literature that is able to simulate the listed processes
of soil formation simultaneously, resulting in a soil profile.
This review will focus on exploring and quantifying the pedogenic processes of the physical and chemical weathering of bedrock, the formation of soil
horizons, and the rate of soil-mixing processes, short-term and long-term.
2. Conceptual Models of Soil Formation

Factors, Processes, Pathways, Energy
The extent of soil formation is believed to be dependent on local site
characteristics. To model the evolution of soil in the landscape, we need to
know which factors and processes are important for describing pedogenesis
quantitatively. In the following, conceptual models of soil formation are
reviewed briefly; they form the basis of mechanistic soil-formation models.
2.1. Factors
The origin of the soil-forming factors equations presented in the following
is discussed in more detail in Box 2.
Dokuchaev is known as one of the first soil scientists who formulated an
equation of soil-forming factors in 1886 (Volobuyev, 1974). The Russian
soil scientist linked the formation of soil to environmental factors using a
descriptive equation:
P f K; O; GB;
where P is the soil, K is the climate, O are the organisms, G is the ground or
parent rock, and B is the time.
In the Western world, Shaw (1930) can be seen as the first soil
scientist, who published an equation that described potent soil-forming
factors. In his equation he stated that soil (S) is formed from parent materials
(M) by a combination of climatic factors (C) and vegetation (V) as a function
BOX 2 Background: The equation of soil-forming factors
Jennys state factor model is seen as the most well-known model of pedogenesis based on soil-forming factors. This often referred to and cited model
states that soil (s) is created as a function of climate (cl), organisms (o), relief (r),
parent material (p), time (t), and additional, unspecified factors (...) (refer to
equation 3).
Jenny published his renowned state factor model in 1941 in his book
titled Factors of soil formation. A system of quantitative pedology. However, previous to Jennys state factor model, the soil scientist Shaw, also
formulated an equation of soil-forming factors.
In 1930, Shaw published a formula of potent factors in soil formation
in the science journal Ecology, describing soil formation as being influenced
by parent materials (M), climatic factors (C), vegetation (V), and time (T), as
well as the processes of erosion and deposition (D) (refer to equation 2).
It was brought to our attention that Shaw presented his conceptual model
of soil formation at the Second International Congress of Soil Science in
Leningrad in 1930. Following Shaws presentation published in the proceedings of the congress, there is a lengthy discussion of Shaws model equation.
For instance, a comment was made by Prof. Romell: He . . .suggested in
order not to hurt the feelings of the mathematicians to simply put S equal to a
general function of the other symbols with:
S f M; C; V ; T ; D:
However, Shaw did not develop his model any further before he died
suddenly in 1939. Further, we learned that Jenny was also a presenter at the
Second World Congress of Soil Science and that he interacted with Shaw in
Berkely (University of California). Subsequently, we hypothesize that Jennys factors of soil-formation equation was developed following the discussions of Shaws paper in 1930.
In the Russian soil science community, however, Dokuchaev has been
credited with first formulating factors of soil formation in his publication on
Key points in the history of land evolution in the European Russia in
1886, published in Russian (Florinsky, 2011a). Dokuchaevs work on soil
formation was introduced to the Western world through the English translation of Afanasievs paper The classification problem in Russian soil science in 1927. Afanasiev cited and discussed Dokuchaevs work including
his hypothesis on soil formation. The English translation of Dokuchaevs
hypothesis states that Every dry land vegetative soil is in all instances a mere
function of the following factors of soil formation:
1)
2)
3)
4)
5)
the nature of the parent rock

the climate of the given locality
the mass and character of vegetation
the age of the country, and finally
the relief of the locality
Here, Dokuchaevs hypothesis is only published as a sentence. However,

in a later publication from 1899 (Dokuchaev, 1899) this hypothesis on soil
formation is written as the mathematical formula shown in equation 1 (e.g. as
associated with Dokuchaevs work in Volobuyev, 1974 and Schaetzel and
Anderson, 2005 where a precise citation of the equation is not given). The
factor of relief or topography does not appear in the mathematical equation
although Dokuchaev discussed its importance on soil formation (Florinsky,
2011b).
Later on Zakharov (1927) published a precursor of Jennys equation in
one of the first fundamental Russian textbooks on soil science (Florinsky,
2011b) where soil (p) formation is a function of parent rock material
(.G..), organisms (P..OpG.), climate (.), age of the terrain (Bop.
cTp.) and topography (P-):
p f M:G:P:; P::OpG:; :; Bop: cTp:; P-:

We also know that both Shaw and Jenny were aware of Dokuchaevs
earlier work on soil-formation factors. Shaw told Jenny he ( Jenny) was
going to be the new Dokuchaev (see Encyclopedia of Soils in the Environment: Jenny, Hans (Amundson, 2005)). Jenny acknowledged that his state
factor equation included factors first proposed by Dokuchaev, however,
emphasized the conceptual differences behind the two formulas, which
he expressed in the first sentence of his renowned book from 1941: As a
science grows, its underlying concepts change, although the words remain
the same (Amundson, 2005; Jenny, 1941).
of time (T). In addition to these soil-forming factors, he also included the

processes of erosion and deposition (D) to describe soil formation in the
landscape:
S M C V T D:
Although presented as a mathematical equation, Shaws soil-formation

model is only a factorial model, listing the major soil-forming factors. Shaw
(1930) emphasized that the influence of these factors in developing soil is
not uniform, but rather changes with local conditions.
The most well-known conceptual model of soil formation is Jennys

(1941) state-factor model, which is also called the clorpt model. It comprises independent variables or state factors that define the state of a soil
system. Hence, these state factors are not considered as formers or creators
of the soil (S).
S f cl; o; r; p; t; . . .;
where cl is the climate, o are the organisms, r is the topography, p is the

parent material, and t is the time, and . . . stands for additional, unspecified
factors. The state factors are independent from the soil system and vary in
space and time (Amundson and Jenny, 1997).
In its original form the state-factors model is unsolvable. To be solved,
the indeterminate function f needs to be replaced by certain quantitative
relationships. Hence, the clorpt equation has been formalized in quantitative ways based on empirical field observations, where a single factor is
defined by keeping the other factors constant (Minasny et al., 2008).
Empirical models were developed to describe soil formation in the form
of quantitative climofunctions, biofunctions, topofunctions, lithofunctions,
and chronofunctions, mostly based on numerically intensive statistical
methods (McBratney et al., 2003; Yaalon, 1975). Based on the clorpt
model of soil formation McBratney et al. (2003) formulated the scorpan
model, which indeed applies empirical quantitative relationships to predict
soil properties from landscape attributes at specific locations in the landscape. The scorpan model is written as:
Sc=Sa f s; c; o; r; p; a; n;
where Sc are the soil classes and Sa are the soil attributes, s is the soil, c is the
climate, o are the organisms, r is the topography, p is the parent material, a is
age, and n is space or the spatial position. The model is used quite extensively in the field of digital soil mapping to predict the recent state of the soil
(soil properties), but is not intended for, and can not be applied, for longterm soil-formation predictions.
2.2. Processes
One of the first soil scientists who described soil formation as processes
instead of factors was Simonson (1959). He considered two processes
important for the evolution of soil, the accumulation of parent materials
and the differentiation of soil horizons in the soil profile. Further, he
described the evolution of soil types as a function of additions (i.e., organic
matter), removals (i.e., soluble salts), transfers (i.e., humus and sesquioxides),
Additions
Additions
Translocations
Transformations
Removals
Removals
Figure 2 Soil profile evolution as a function of additions, removals, translocations, and

transformations.
and transformations (i.e., primary minerals into secondary minerals) as

demonstrated in Fig. 2:
s f addition; removal; translocation; transformation:
The soil-forming processes approach by Simonson (1959) can be seen as

one of the conceptual frameworks for mechanistic models of soil formation
implementing physical laws (Minasny et al., 2008). However, the original
work is still a qualitative description.
2.3. Pathways
Johnson and Watson-Stegner (1987) introduced the concept of pathway
models. They viewed soil evolution as a result of genetic pathways. Their
model considers soil as a complex open system with changes in soil thickness
and increasing genetic complexities with time. They stated that soil (S)
forms progressively (P) and regressively (R) along interacting pathways:
S f P; R;
10
where P stands for progressive pedogenic conditions, including processes

and factors that promote horizonation, developmental (assimilative)
upbuilding, and/or subsurface deepening; and R stands for regressive pedogenic conditions, including processes and factors that promote haploidization, retardant (nonassimilative) upbuilding, and/or surface removal. Soil
evolves along these progressive and regressive pathways, where some might
be dominant over others. This is yet another qualitative description.
2.4. Energy
Models of soil formation based on the concepts of energy describe pedogenic factors and processes implementing the principles of energy or concepts of thermodynamics. The most well known and cited soil scientist
addressing this possibility has been Runge (1973). However, in their review
Minasny et al. (2008) emphasized the work of Volubuyev from Azerbaijan.
Volobuyev published various papers on linking pedogenic processes with
laws of energy. The most relevant models for estimating soil formation from
energy laws are included in the following paragraphs.
Runge (1973) presented a different type of factorial model, formulating
soil evolution based on energy:
S f o; w; t ;
where S is the soil, o is the organic matter production (renewing factor), w is

the amount of water available for leaching (developing vector), and t is time.
Climate and relief are expressed within the vector w. This energy model
relies on gravity as the main source of energy, driving the infiltration of
water in the soil, which is responsible for horizonation. The model considers solar energy indirectly in the production process of organic matter.
The energy model of Runge (1973) is only useful in a qualitative way,
because actual quantitative thermodynamical calculations are not implemented in the model (Hoosbeek and Bryant, 1992).
In an unpublished thesis, Regan (1977) studied soil formation through
energy processes and created an energy model of soil formation. He based his
studies on soils derived from limestone and marine washed sands in Florida,
USA. Regan (1977) calculated the total amount of energy needed to form
soil by implementing the energy from sunlight, the energy flux from carbon
dioxide production by organisms, wind, and temperature; the chemical
energy of rain; the kinetic energy developed from sloped surfaces; the
chemical free energy of phosphorus and the gravitational energy from uplift
processes. Running the energy model steady-state conditions of soil formation are reached after only 525 years for soils with sandy parent materials, and
after 375 years for soils with calcareous parent materials with a rate of
approximately 5.65 103 kJ m 2 yr 1. Rates of soil formation with time
11
are obviously underestimated, but nevertheless this model can be seen as a

good example for linking the amount of energy needed for soil formation
with vegetation and urban growth (Minasny et al., 2008).
The quantification of processes of energy transformation during soil
formation was addressed intensively by Volobuyev. For instance, he
described the expense of energy in the process of soil formation applicable
for all climatic zones as follows (Volobuyev, 1974):
Q Ra Re1=mK ;
where Q is the expenditure of energy on soil formation, R is the energy of

solar radiation, a are the available energy sources, K is the relative wetness,
and m is a factor expressing the participation of biota in energy exchange.
Following on, Volobuyev and Ponomarev (1977) investigated various
thermodynamic aspects of soil-forming processes. They calculated Gibbs
free energy (DG) and entropy (S) for different soil types from individual
Gibbs free energy and entropy values of soil minerals (see Volobuyev and
Ponomarev, 1977, Table 1, p. 6) showing that the thermodynamic characteristics of soil minerals vary significantly for the soil types studied. In
addition, they identified two soil groups based on their energy expenditure
during mineral formation: (1) one that is characterized by a decrease in
Gibbs free energy and an increase in entropy (2) and one that is characterized by an increase of Gibbs free energy and a decrease in entropy.
Further, Volobuyev et al. (1980) used these Gibbs free energy potentials
for soils to predict their infiltration or leaching capacity. They showed that
the lower the Gibbs free energy levels of soils, the higher their infiltration
capacities. Based on calculations from Volobuyev and Ponomarev (1977)
and Volobuyev et al. (1980), Minasny et al. (2008) presented Gibbs free
energy and entropy for different soils, rocks, and minerals, as shown in
Fig. 3.
Soils enriched with SiO2, Al2O3, Fe2O3, and CaCO3 and large quantities of residual minerals have low Gibbs free energy (which is lost
during weathering) and high entropy. In the order of higher energy and
lower entropy, this is followed by phyllosilicate minerals, carbonates, and
soluble salts. A decrease in Gibbs free energy and an increase in entropy are
associated with minerals that have higher intensity of leaching and are more
resistant to weathering.
Volobuyev (1984) also formulated an energy model to apply Dokuchaevs equation quantitatively (Eq. (1)):

Pc wR 0:67
Q Rr exp
;
9
mP p
12
1000
900
Gibbsite
800
Kaolinite
Oxisols
Olivine
Mollisols
600
500
Vertisols
400
S (kJ kg-1)
700
Quartz
Gypsum
300
200
100
16,000 14,000 12,000 10,000 8000
6000
4000
2000
0
0
G (kJ kg-1)
Figure 3 Gibbs free energy (DG) and entropy (S) for different soils, rocks, and
minerals (graph is based on Minasny et al., 2008).
where Q is the (annual) expense of energy on soil-forming processes, R is

the radiant solar energy, P is the relative wetness, m is the biological
activity, r is the radiation balance, p is the atmospheric precipitation,
w (chemically bound water of mineral soil components) is the rate of
mineral transformations in soils, and Pc is water, such as water that is
fixed in the mineral, faunal, and floral component of soils. Dokuchaevs
soil-forming factors, as demonstrated in Eq. (1), are represented by R and
P (climate, K), by m (organisms, O), by Pc and w (parent rock, G), and (m)
by p and r.
2.5. Summary
Conceptual models of pedogenesis have been, and are still being developed, for the past 100 years. These include the factorial, processes,
pathways, and energy models. However, these models are mostly interpreted qualitatively, although some have been applied in a quantitative
way, that is, the factorial model of Jenny that could be solved by applying
empirical quantitative relationships to predict soil properties from landscape attributes. These conceptual models can form the basis of mechanistic models.
13
3. Soil Weathering and Production

As described in the previous sections, several processes are responsible
for transformations, translocations, additions, and removals in the soil system. Ultimately, these processes and their associated transformations of
energy result in the formation of a particular soil profile. Further, the
dynamics of the interacting chemical, physical, and biological processes
are believed to be induced by different parent materials and climates.
The horizonation and differentiation of the soil profile is the result of
transformation processes like soil weathering and soil mineralization,
decomposition and humification, and aggregate formation as well as translocation processes like eluviation and illuviation.
A common perception in pedology is that pedogenesis is a product of
mostly downward moving processes such as leaching that lead to the formation
of interrelated layers, the A and B horizons (Huggett, 1998). Pedoturbations,
the so-called soil-mixing processes are seen as processes that are working
against horizonation rather than promoting it, because of possible mixing of
surface and subsoil materials induced by mixing agents such as soil biota,
soil moisture changes, and periodic freezing and that are resulting in the
subsequent homogenization of the layers in the soil profile (Huggett, 1998).
The following sections will explore how processes of soil formation are
modeled or estimated with field data in the literature.
3.1. Production of soil from parent materials

According to NAS (2010): The breakdown of bedrocka major factor in
Earth surface processesis among the least understood of the important
geological processes.
The evolution of soil has been explained vastly with the help of chronosequences over time scales of up to millions of years. Traditional theories
of soil evolution along chronosequences explain soil development progressively under the influence of environmental factors until soil development is
in equilibrium (Huggett, 1998). Accordingly, it is believed that the development of a certain soil type is preset in a certain landscape, that is, in the
German soil classification scheme, on limestone parent materials Rendzinas
are formed that eventually evolve into Brown Earths. For instance, chronosequences were created based on conceptual ideas and observations in the
field with the help of successional stages of vegetation by placing them in a
chronological order, and by exploring soil profiles that developed on
surfaces of known age (Schaetzel and Anderson, 2005). One discrepancy
in formulating chronosequences is the assumption of constant soil-forming
factors except time. This is especially unlikely for the soil-forming factors
14
climate and vegetation cover. New views in evolutionary pedogenesis tried

to explain the nonlinear behavior of soil development by assuming that soils
evolve through continual formation and destruction, and consequently
might progress, regress, or stay constant depending on environmental conditions (Huggett, 1998).
Chronosequences can be transformed into chronofunctions by plotting
soil and landscape properties against time (or age) using time as the independent variable, based on Jennys state-factor equation (Schaetzel and
Anderson, 2005):
SSoil ft timecl; o; r; p; . . . :
10
Further, statistical models can be applied to express chronosequence data

(soil and landscape properties) mathematically by fitting curves of soil
evolution with time. Schaetzel et al. (1994) reviewed types of mathematical
functions commonly used in chronofunctions (Fig. 4).
Sigmoid
Soil property
Power
Logarithmic
Exponential
Linear
Time
Figure 4 Types of mathematical functions commonly used in chronofunctions

(adapted from Schaetzel et al., 1994). S-shaped or sigmodial curve, general form of
equation: Y 1/(a b exp(t)); power functions, general form of equation: Y atb;
logarithmic functions, general form of equation: Y a b(log t); exponential functions, general form of equation: Y a exp(bt); simple linear functions, general form of
equation: Y a bt.
15
Demonstrated in Fig. 4 chronofunctions might be modeled using:

(1)
(2)
(3)
(4)
(5)
simple linear behavior Y a bt

single logarithmic behavior Y a b(log t)
exponential behavior Y a exp(bt)
power functions Y atb
or nonlinear-sigmodial functions Y 1/(a b exp(t)).
Linear functions suggest that the soil system evolves at a constant rate
through time whereas logarithmic models imply that the soil system is in
steady state or will reach a steady state eventually some time in the future.
Nonlinear-sigmodial chronofunctions propose that the soil system evolved
along periods of rapid pedogenesis followed by decreasing rates (Schaetzel
et al., 1994).
Following the concepts of chronofunctions, production rates of soil
from bedrock parent materials were assumed to decrease with time and
were discussed to follow a linear or nonlinear function of time (Colman,
1981). However, there are two concepts of soil production that have been
discussed much more extensively in the geomorphology literature than in
pedology (Fig. 5).
Relative production rate
0.8
k2 = 0
0.6
k2/k1 = 3
0.4
0.2
k2/k1 = 1.5
0
0
0.2
0.4
0.6
0.8
Soil thickness (m)
1.2
Figure 5 The rate of soil production versus soil thickness. Here, both the exponential
and the humped soil-production model are presented graphically. Both axes are dimensionless. Soil production is presented graphically depending on different values of the
parameters k1 and k2 (see Eqs. (12) and (13)). If k2 0 the soil production equals a
depth-dependent exponentially decreasing soil-production function. If k2/k1 0 soil
production shows a humped function.
16
(1) The exponential soil-production model

This model states that the formation of soil declines exponentially with
increasing soil thickness. This theory was discussed conceptually by Ahnert
(1977). He assumed that the rate of soil production from hillslopes decreases
exponentially with increasing thickness of the overlying soil mantle. The
exponential decrease of the rate of soil production with increasing soil
thickness is believed to be dependent on the soil temperature and water
penetration through the soil profile. The exponential decrease of the temperature range with increasing soil depth is seen as a factor for reduced
weathering with increasing depth below the soil surface (Minasny and
McBratney, 1999). Further, the occurrence of moisture related processes
like freeze-thaw also decreases exponentially under increasing soil depth.
Heimsath et al. (1997, 1999) verified the exponential decline of soil
formation with increasing soil depth with field data from the Tennessee
Valley in California, USA, and found field evidence for the theory of
exponential decline of soil production with increasing soil depth and parameterized the potential weathering rate of bedrock. According to this
research, soil production is the greatest when bedrock is just exposed and
decreases with increasing soil depth. The derivation of soil-production rates
(SPRs) from field data became possible with the measurement of in situ
terrestrial cosmogenic nuclide (TCN) concentrations from soil parent materials such as bedrock or saprolite, sampled underneath different soil depths.
The exponential decline of soil production (SPR) with increasing soil
thickness is described as (Dietrich et al., 1995; Heimsath et al., 1997):
SPR P0 expbh;
11
where P0 ([L T 1], mm kyr 1) is the rate of weathering of bedrock at

h 0, h ([L], cm) is the soil thickness and b ([L 1], cm 1) is a rate constant,
a length scale that characterizes the decline in soil production with increasing soil thickness.
(2) The humped soil-production model
The second concept explains the conversion of bedrock or saprolite to
soil using a humped function (Humphreys and Wilkinson, 2007). This
theory was first introduced by Gilbert (1877) and later discussed by
Carson and Kirkby (1972). Accordingly, soil production is highest underneath a nonzero soil depth. Hence, the weathering of bedrock is greatest
underneath an incipient soil depth and slower underneath exposed bedrock
or an already thick soil mantle.
The main explanation for the occurrence of a humped model of soil
formation is the maximization of chemical and physical weathering processes under an initial soil depth. The presence of water is discussed as a
17
major factor in this phenomenon by Carson and Kirkby (1972). Water is

considered as an important agent for chemical weathering of bedrock or
saprolite into soil. Carson and Kirkby (1972) give an example for their
reasoning: On an exposed site, like bedrock, water tends to run off, which
lowers the rate of chemical weathering. This characteristic also applies for
very thin soils where water runs off fast, because of insufficient pore space in
the soil to hold the water. In soils with a thick soil cover the circulation of
water also tends to be slow, which reduces the rate of weathering. Furthermore, because water tends to run off bare rock, processes of freeze-thaw are
also limited underneath shallow or deep soil covers. Granger et al. (2001)
proposed that the presence of moisture in an already thin soil cover seems to
be very important for the chemical weathering of granite rock to gruss,
which is relatively resistant to weathering.
Wilkinson and Humphreys (2005) discussed the role of fauna and flora
to explain the occurrence of a humped model of soil production. They
argue that animals and plants require a moderate soil mantle to promote soilproduction processes. Yoo et al. (2005) applied this theory when modeling
the population of pocket gophers in relation to soil thicknesses. In addition
to mesofauna, plant roots make an important contribution to the
weathering of saprolite by disturbing the soil-saprolite interface. Most likely
those disturbances create channels and therefore access for weathering
agents (i.e., water) into the soil profile. Biota are also able to penetrate
through the soil and saprolite into the zone of unweathered rock. However,
Wilkinson and Humphreys (2005) also argue that biota will reach the soil
bedrock interface less frequently under a thick soil cover.
Minasny and McBratney (2006) presented the humped model as a
continuous double exponential function (see Fig. 5):
@e
P0 expk1 h expk2 h Pa
@t
12
where P0 ([L T 1], mm kyr 1) represents the rate of weathering of

bedrock, h ([L], cm) is the soil thickness, k1 represents the rate of mechanical
breakdown of the rock materials, k2 is the rate of chemical weathering, Pa is
the weathering rate at steady-state condition ([L T 1], mm kyr 1) with
condition k1 < k2. As k2 0, the humped function is reduced to the
depth-dependent exponential soil-production function (Eq. (11)). The
critical thickness, hc, where weathering is at maximum is given by:
hc
lnk2 =k1
:
k2 k1
13
An empirical parameterization of the humped model from field data is

still to be achieved, although Heimsath et al. (2009) assumed a humped form
18
of soil production for a study site at Arnhem Land in Northern Australia for
a landscape dominated by outcrops and soil depths no less than 35 cm.
Dietrich et al. (1995) applied the exponential and also the humped model
to simulate soil formation in a catchment of the Tennessee Valley in
California, USA. With a chosen maximum soil production at 25 cm, the
predicted landscape formed is characterized by sharply curved ridges and
outcrops. In such an environment with humped soil production, soil depths
below the peak of soil production are assumed to be unstable and together
with soil erosion this will lead to a stripping of the soil to bedrock.
Consequently, no soil depths that are in equilibrium should be observed
in the field for less than the peak in soil production. However, the modeling
results of Dietrich et al. (1995) were more consistent with field observations
when using an exponential model. Later work by Heimsath et al. (1997)
showed that soil formation indeed followed an exponential function at the
study site of 0.077 mm yr 1 with no soil mantle. Following Heimsath et al.
(1997), a small range of different authors applied the concepts of deriving
SPR from TCN data. SPRs derived from TCN range between 0.004 and
0.4 mm yr 1 (Heimsath et al., 2000, 2001a,b, 2002, 2005, 2009; Wilkinson
et al., 2005).
Current research is also focused on finding the conversion rate of
bedrock to regolith (Brantley, 2010; Graham et al., 2010). In pedology,
bedrock as well as regolith is parent material for soil production. Different to
hard fresh rock, regolith is a form of weathered friable rock that still has
the structural characteristics and fresh primary minerals of the parent rock
and is transformed into soil through physical disruptions. Regolith is already
seen as a hospitable substrate, in particular for the soil flora. Published rates
of regolith formation for hard granitic rocks are relatively low and range
between 0.004 and 0.02 m yr 1 (Dosseto et al., 2008; Graham et al., 2010).
3.2. Chemical weathering of bedrock to soil

The chemical weathering of bedrock to soil is an important process in
forming the soil mantle. Rates of chemical weathering have been estimated
from both laboratory and field-based studies applying a variety of investigative methods. However, a time-dependent difference between rates in the
laboratory and field rates was found (Lasaga et al., 1994). Calculated laboratory rates are much faster when compared to observed field rates by up to 5
orders of magnitude (Brantley et al., 2007), yet both methods show a
decrease of chemical weathering with time.
In the laboratory, chemical weathering rates mostly depend on the
experimental setup like pH, the percolation rate, and the weathering state
of the silicate minerals studied, that is, fresh or partly weathered minerals.
For instance, White and Brantley (2003) investigated the long-term dissolution of plagioclase (weathering rate R in mol m 2 s 1) extracted from
19
fresh and weathered granite. In one of the longest experimental setups, of

over 6 years, they determined parabolically decreasing weathering rates of
7.0 10 14 mol m 2 s 1 for fresh Panola Granite, and found significantly
less chemical weathering rates of 2.1 10 15 mol m-2 s-1 for partially
weathered Panola Granite, which reached steady-state in only 2 months.
After extrapolating decreasing weathering rates of fresh plagioclase with
time, they concluded it would take several thousand years of reaction to
replicate the rate of the naturally weathered plagioclase under identical
experimental conditions.
Chemical fluxes in a watershed can be determined using the solute discharge flux Qi,dis for a chemical species i based on a mass-balance approach:
Qi;dis Ci;dis
V
;
At
14
where Ci,dis is the chemical concentration of a chemical species i, V is the

fluid mass, A is the geographic area of the watershed, and t is time (White
and Blum, 1995).
In watershed studies, chemical weathering kinetics of silicate (temperature dependence of dissolution rate) are often modeled empirically with the
help of the Arrhenius equation, which describes the dependence of chemical weathering rates on the temperature (WX) in natural systems (White and
Blum, 1995):
WX A expDEX =RT ;
15
where A is an empirical constant that incorporates the effects of surface area

and surface reactivity, DEX is the activation energy for the weathering
reaction that releases element X (kJ mol 1), R is the universal gas constant
( J K 1 mol 1), and T is the absolute temperature (K). Based on silicondioxide (SiO2) and sodium (Na) fluxes, activation energies for chemical
weathering were calculated to be 59.4 and 62.5 kJ mol 1, respectively.
A range of chemical weathering rates estimated from field data (usually
expressed by M L 2 T 1) can be found in the literature, a selection is
presented in the Appendix, Table A1 (parts 1 and 2). Part 1 summarizes
chemical weathering rates based on elemental fluxes (loss and gain) in watersheds, and the chemical composition of the parent materials and weathering
products studied. For instance, Colman and Dethier (1986) summarized rates
of chemical weathering with time based on catchment mass loss from solutes,
varying between 0.003 and 0.04 mm yr-1 for different climates and parent
materials (Dethier, 1986; Paces, 1986; Pavich, 1986; Velbel, 1986).
In another study, Alexander (1988) calculated rates of chemical weathering
based on mass-balance equations for 18 watersheds with noncarbonate lithologies. The predicted rates of about 0.00170.16 mm yr 1 were mostly based
20
on the volume of runoff water and the soil to rock ratio, that is, the mass of
soil/mass of bedrock weathered to produce that amount of soil.
Summarizing Table A1, part 1 (Appendix), we can see that rates of
chemical weathering vary between 0.00017 mm yr 1 calculated for a
tropical environment by Owens and Watson (1979) and 0.473 mm yr 1
calculated for a humid-tropical environment by Wakatsuki and Rasyidin
(1992). However, the majority of compiled chemical weathering rates from
these watershed studies ranged between approximately 0.01 and
0.1 mm yr 1. A pattern of chemical weathering rates being grouped by
study areas with similar climate regimes and parent materials was not readily
discernible in the data.
In the current literature, various studies are now focusing on calculating
chemical weathering rates of bedrock in situ to investigate soil-formation
processes. Calculated rates are presented in Table A1, part 2 (Appendix),
with Table A2 (Appendix) listing the main characteristics of the environments
studied. The so-called weathering indices are used to estimate the extent of
chemical weathering based on mass-balance calculations or the ratio between
the chemistry of fresh parent rock to that of weathered rock or soil (Taylor and
Eggleton, 2001). Generally, the rock to soil ratio of an immobile, slowly
weatherable mineral is used to estimate rates of chemical weathering based
on its loss and gain. In most of the studies, zircon (Zr) is used.
Before discussing the studies that applied this approach, the term total
denudation rate used in these publications needs to be clarified: The term
total denudation rate refers to the combined rate of chemical and physical
weathering of rock to soil.
In publications of Riebe et al. (2003, 2004a,b), Green et al. (2006), Yoo
et al. (2007), and Burke et al. (2009, 2007), total rates of denudation or
weathering were substituted with rates of soil production derived with
TCN. To estimate the extent of chemical weathering, determined chemical
weathering rates (using weathering indices) were subtracted from rates of
soil production (using TCN). In the publication of Dixon et al. (2009),
however, total denudation rates were calculated by combining determined
chemical weathering rates (using weathering indices) and rates of soil
production (using TCN). TCN-derived SPRs were therefore substituted
with rates of physical weathering.
Riebe et al. (2003, 2004a,b) calculated in situ rates of chemical
weathering using weathering indices in conjunction with SPRs derived
from in situ TCNs. In these studies, SPRs are substituted with total denudation rates to estimate the degree of chemical weathering.
The conservation of mass equation for the chemical weathering rate as a
fraction of the total denudation or weathering rate is written as:

W D 1 Zrrock =Zrsoil ;
16
21
where W is the chemical weathering flux in [M L 2 T 1] or [L T 1], D is

the total denudation rate, and [Zr]rock and [Zr]soil are the concentrations in
rock and soil of Zr.
Riebe et al. (2003, 2004a,b) estimated chemical weathering to vary
between as low as 0.0041 mm yr 1 and as high as 0.14 mm yr 1. At the
study site in Rio Icacos, Puerto Rico (Riebe et al., 2003), results showed
that chemical weathering accounted for 58% and 68% of the total loss by
chemical and physical weathering. On average, 0.021 mm yr 1 was
accounted for by chemical weathering of saprolite and 0.054 mm yr 1
was attributed to combined chemical weathering and physical soil
weathering. Results from the study site at the Santa Rosa Mountains in
Nevada (Riebe et al., 2004b) showed that chemical weathering rates
decreased with increasing altitude from 0.02 to 0 mm yr 1, suggesting a
trend of dominance of physical erosion with increasing altitude. Sparsely
vegetated, high-altitude crystalline terrain in particular seemed to be characterized by very slow silicate weathering rates. For 42 different study sites
with granitic parent materials located in diverse climate regimes, the data
analysis implied that chemical weathering rates increase proportionally with
supply rates of fresh material, that is, supply-limited weathering (Riebe
et al., 2004a).
In the majority of the following studies, the conceptual ideas of Riebe
et al. (2003, 2004a,b) have been employed. Green et al. (2006) conducted a
study in the Bega Valley in south-eastern Australia to quantify chemical
weathering along a hillslope, implementing SPRs of the area as analyzed by
Heimsath et al. (2000). They calculated that mass loss by chemical
weathering accounted for 3555% of the total mass loss from the hillslope.
Results from Burke et al. (2007) indicated a decrease of chemical
weathering of saprolite with increasing overlying soil thickness. At the
study site Point Reyes in California, USA, chemical weathering accounted
for 1351% of total denudation rates. Assumptions were made that spatial
variation in chemical weathering is controlled by the topography of the
study sites. Accordingly, weathering rates decrease with slope across the
divergent ridge and increase with upslope contributing area in the convergent swale. Results also showed that measurements of saprolite abrasion pH
are closely related to weathering indices, assuming that the intensity of
chemical weathering decreases linearly with an increase in saprolite pH
from 4.7 to 7.
Yoo et al. (2007) combined geochemical mass balance with sediment
transport to predict rates of soil chemical weathering and transport on
hillslopes. This research was conducted at a site in the south-eastern highlands (Frogs Hollow) in south-eastern Australia. Results showed losses of
soil chemical weathering rates of 0.029 mm yr 1 on the ridge and consequently gains of 0.029 mm yr 1 at the lowest slope positions. Additionally,
Yoo et al. (2007) investigated soil residence times by linking soil transport
22
and topography. Results indicated residence times for soil of 4 kyr on the
ridge to 0.9 kyr at the base of the hillslope.
Burke et al. (2009) compared the extent of chemical weathering for field
sites in the lowlands and highlands around the area of the Bega Valley in
south-eastern Australia and found indications for lower rates of chemical
weathering for the highland (47%) compared to the lowland sites (57%).
Dixon et al. (2009) applied a similar approach for the Sierra Nevada
Mountains in California, USA. Results implied that chemical weathering
of saprolite to soil peaks at mid elevations compared to high and low
elevation sites and that physical erosion rates increase with both saprolite
weathering rates and intensity. In contrast to previous publications, total
denudation rates were calculated as the sum of total chemical weathering
and SPRs.
3.3. Summary
Soil mantled landscapes are believed to be the result of two main
concepts of soil formation: (1) the exponential soil-production model
and (2) the humped soil-production model.
The exponential decline of soil production with increasing soil thickness was verified with field data in situ. Calculated rates of weathering of
parent material are as low as 0.004 and as high as 0.4 mm yr 1.
The humped model was used to explain soil formation conceptually for
some landscapes, but an empirical parameterization is still to be achieved
from field data.
Laboratory derived chemical weathering rates of rocks or minerals are
generally more rapid than rates estimated from field data.
Minimum rates of chemical weathering derived from field data are as
low as 0.01 mm yr-1 with maximum rates of about 0.5 mm yr 1.
The majority of chemical weathering rates estimated from catchment
based mass loss of elements range between 0.01 and 0.1 mm yr 1 with
mean values of about 0.060 mm yr 1.
Rates of chemical weathering derived with so-called weathering indices
vary from 0 to 0.144 mm yr 1 with mean values of about
0.022 mm yr 1 and are relatively similar to chemical weathering rates
estimated from catchment based methods (Figs. 6 and 7), although rates
derived with weathering indices were generally lower.
Rates of total denudation were estimated for all chemical weathering
studies based on weathering indices using TCNs. Generally, calculated
rates of chemical weathering were lower than rates of physical
weathering.
Both datasets on chemical weathering rates (Appendix, Table A1, part 1
and 2) show no distinct pattern of weathering rates being dependent on
different climates and parent materials of the sites studied.
23
Riebe et al. (2004b)
Riebe et al. (2004a)
Riebe et al. (2003)
Green et al. (2006)
Burke et al. (2009)
Burke et al. (2007)

0
0.05
0.1
Rate of chemical weathering (mm/yr)
0.15
Figure 6 Box-plot distributions of chemical weathering rates derived from weathering

indices (Appendix, Table A1, part 2). Graph shows the range in the datasets derived
from in situ field data using weathering indices; the line in the middle of the box refers to
the median of the distribution.
4. Soil Mixing: Vertical and Lateral Movements

Soil-mixing processes ultimately result in the detachment, transport,
sorting, and deposition of material within the soil mantle and on its surface
(Paton et al., 1995). Consequently, these pedoturbations initialize vertical
and lateral movements in the soil mantle.
Various forms and agents of pedoturbation can be seen in Table 1. Here,
soil creep is characterized as mainly induced by gravity-related mass movements (graviturbation). However, creep is also caused by bioturbation via
processes like burrowing and the growth and decay of roots which is
followed by the refilling of generated holes from the upslope side. In
Table 1, pedoturbation caused by water is restricted to rain splash and
movement of water down the profile, excluding overland lateral transport
processes and deposition by water, that is, rill or gully soil erosion caused by
water. The most relevant pedoturbations for pedogenic studies are soil creep
and bioturbation, as well as eluviation and illuviation processes that transport soil particles due to the movement of water down the soil profile.
Table A3 (Appendix) summarizes various rates of pedoturbation that are
found in the literature by the mixing agents involved. Some rates of
pedoturbation were excluded, because of their specific nature and these
24
Weathering
indices
Other
0.1
0.2
0.3
0.4
0.5
Rate of chemical weathering (mm/yr)
Figure 7 Box-plot distributions of chemical weathering rates derived from field data
(Appendix,Table A1, parts 1 and 2). Graph demonstrates range in weathering indices
datasets compared to different methods (generally catchment based mass balance from
solutes); the line in the middle of the boxes refers to the median of the distribution.
Table 1
2005)
Forms and agents of pedoturbation (based on Schaetzel and Anderson,
Form of pedoturbation
Soil-mixing agents
Aeroturbation
Anthroturbation
Aquaturbation
Gas, air, wind

Humans
Water (rain splash, eluviation/illuviation
processes)
Shrinking and swelling of clays
Freeze-thaw activity, ice crystals
Crystals, such as ice and various salts
Animals, including insects
Plants
Mass movements, such as creep
Extraterrestrial impacts such as comets and
meteorites, and human-generated
impacts, that is, artillery shells and bombs
Earthquakes
Argilliturbation
Cryoturbation
Crystalturbation
Faunalturbation (Bioturbation)
Floralturbation (Bioturbation)
Graviturbation
Impactturbation
Seismiturbation
25
are discussed in the text only. Listed rates of pedoturbation were mainly
based on observed pedoturbation processes in the field such as ants moving
soil particles or litter, or by examining and interpreting the products of such
processes (Schaetzel and Anderson, 2005). Most estimates were therefore
based on more or less short-term observations rather than estimating longterm mixing rates from dating soil particles in situ. We can see that rates of
pedoturbation vary extensively in their rapidity and that ants, termites, and
earthworms are one of the major agents causing pedoturbations.
4.1. Bioturbation
The influence of plants and animals on soil formation was not recognized
profoundly in early publications of soil science. For instance, Jenny (1941)
did not examine the role of animals as part of the o factor, because of lack of
sufficient observational data. . .. Further, Carson and Kirkby (1972) nominated faunal and floral mixing agents as being of secondary or minor
importance in causing soil creep. Carson and Kirkby (1972) defined processes of soil creep (only) as diffusion processes or mass movements in the
soil profile, mostly caused by re-working of the soil surface, because of soil
moisture and temperature changes such as frost cycling and resulting from
the steady application of downhill shear stress. Nevertheless, they acknowledged possible rates of soil creep produced by soil fauna and flora in addition
to those resulting from mass movements. On the other hand Paton et al.
(1995) argued that bioturbation processes play a major role in forming a soil
profile (see Boxes 3 and 4).
In the current literature, there is still debate about how much influence
bioturbation actually has on soil transport or soil-mixing processes and
therefore pedogenesis, and if bioturbation results in net downslope movement or not (Amundson, 2004; Wilkinson et al., 2009). Biological activity
in soils mostly depends on the soil structure including porosity, texture and
stone content; water content; the roots that hold the soil together; and
available nutrients. Often the A-horizon or topsoil is the part of the soil
profile where most bioturbation processes take place and therefore would
be equivalent to the so-called biomantle (Amundson, 2004).
It is believed that bioturbation processes are able to either promote the
development of horizons (horizonation) in texture contrast soils or impede soil
horizonation altogether (haplodization) through the homogenization of the
topsoil and the subsoil as well as material transport between the subsoil and
the topsoil. Furthermore, it is also believed that bioturbation processes can be
responsible for the formation of stone layers, generally occurring at the interface between the A and B horizons (Lobry de Bruyn and Conacher, 1990;
Muller-Lemans and van Dorp, 1996; Paton et al., 1995; Wilkinson et al.,
2009). Particularly, small mammals like gophers are thought to cause the
creation of a stone layer in some areas of the soil profile (Gabet et al., 2003).
26
BOX 3 The Macquarie school of bioturbation
Research from Macquarie University in Sydney, Australia, identifies bioturbation processes as one of the key components of pedogenesis. Based
predominantly on observations in Australia, Paton, Humphreys and Mitchell
(1995) published A New Global View on Soils, describing pedogenesis as
depending predominantly on bioturbation, slope processes and the landscape
studied, and introducing an alternative view on the genesis of soils in
pedology. Geoff Humphreys was credited with first using the term bioturbation in pedology by Don Johnson (the term and concept of bioturbation
is originally from ichnology and oceanography describing the displacement
and mixing of sediment particles and solutes by benthic fauna or flora) (also
refer to Box 4).
Paton et al. (1995) proposed that the formation of texture contrast soil
resulted from interacting slope and pedogenic processes with bioturbation
processes being the main drivers in forming the clay rich subsoil horizon.
They argued that the downward movement of clay via lessivage has been
overemphasized and in fact plays a minor role in forming duplex soil profiles
in landscapes. Therefore, rather than transporting fine materials down the
profile via leaching processes, clay and silt-sized materials are moved down
the soil profile or downslope by biological agents resulting in an accumulation of coarser materials on the top of the profile. Consequently, the authors
put more emphasis on lateral rather than vertical movements in the soil
profile. It was noted that especially the deeply weathered so-called duplex
soils in tropical Australia would have formed under the proposed circumstances. Further, they used these processes to explain the formation of basal
stone layers occurring below the so-called biomantle; the zone that is
produced largely by bioturbation in the soil profile ( Johnson, 1990). Stones
that are too large for the bioturbating agents to be transported to the soil
surface will accumulate below the burrowing depths of the soil fauna.
Following on, Wilkinson and Humphreys (2005) concluded that rates of
bioturbation are significantly faster than rates of soil production. They hypothesized that soil-mixing rates exceed SPRs by up to 3 orders of magnitude for
study sites in south-eastern Australia, and based their argument on soil-mixing
rates by faunal agents of about 0.42910.6 mm yr 1 and SPRs of about
0.0130.061 mm yr 1. In 2009, Wilkinson et al. (2009) reviewed estimates
of bioturbation that were extrapolated from short-term observations, mostly
recorded by researchers from their school. In this context, the importance of
biogenic agents in pedogenic studies was again emphasized. Bioturbation rates
for earthworms were reported to be as high as 1050 t ha 1 yr 1
(0.8334.167 mm yr 1) with some reaching 100 t ha 1 yr 1
(8.33 mm yr 1). Ants, termites, and vertebrates recorded comparable rates of
approximately 15 t ha 1 yr 1 (0.0830.417 mm yr 1); for ants rates of about
510 t ha 1 yr 1 (0.4170.833 mm yr 1) were also noted.
27
Wilkinson et al. (2009) acknowledged the role of Charles Darwin in

emphasizing the effect of biota on soil. Darwin can be seen as one of the first
scientists who observed and recorded mixing of plant and mineral matter in
soil by earthworm activities. He proposed a mixing rate of 0.244 cm2 yr 1
for temperate maritime environments. Furthermore, Darwin made an inference that around 26,886 earthworms live in an acre (4046 m2) of pasture soil
and that each earthworm would ingest 20 ounces of soil each year
(560 g yr 1, which translates to a total amount of 15 t of soil annually).
(Source: Pedometron, The Newsletter of the Pedometrics Commission of
the IUSS, No. 20, p. 14.)
Following research initiated by Humphreys, who died unexpectedly in
2007, Wilkinson is continuing his legacy at The Macquarie School,
investigating the importance of biological mixing of soils and its implications
for pedogenesis.
BOX 4 The term bioturbation in pedology
At the 19th World Congress of Soil Science in Brisbane, Australia (16

August 2010), we learned that the term bioturbation was introduced to
pedology by Winfried E. Blum and Robert Ganssen in 1972 in their
publication about Soil-forming processes of the Earth for the journal
Die Erde, at least to the German speaking soil scientist (Blum and
Ganssen, 1972). Blum and Ganssen (1972) used the term bioturbation
for explaining horizon blurring processes by burrowing animals like earthworms, insects, or small mammals who habituate the soil environment. They
noted that bioturbation processes by termites in tropical and subtropical
environments resulted in a complete mixing of soil horizons of up to 3 m
of depth, destroying pedogenic structures most likely formed by lessivage.
However, it is also suggested that termites could be partly responsible for their
creation, because of selective transfer of certain fine soil particle sizes, clay and
silt, to the surface (Lobry de Bruyn and Conacher, 1990).
4.1.1. Faunaturbation
Agents that predominantly cause faunaturbation are earthworms, ants, termites, gophers, wombats, ground squirrels, and mice (Muller-Lemans and
van Dorp, 1996). Hole (1981) classified those agents into two groups,
exopedonic (outside the soil) and endopedonic (inside the soil) agents. He
also listed 12 soil fauna related activities that take place within a soil body:
28
mounding, mixing, forming voids, back-filling voids, forming, and destroying peds; regulating soil erosion, movement of water and air, plant litter,
nutrient cycling, biota, and producing special constituents. In the following
rates of soil turnover by faunal agents are going to be discussed. Most of the
discussed rates are presented in Table A3.
Hole (1981) summarized that Arctic ground squirrels and pocket
gophers are able to excavate up to 1.5 mm of soil each year, that birds
were able to turnover the complete litter layer of a subtropical rainforest
floor in Australia in only 3 weeks, and that earthworms required only up to
300 years for forming an ochric epipedon by mixing of the topsoil (A2) and
the litter layer (O).
Samedov and Nadirov (1990) concluded that biota, especially earthworms and woodlice can increase soil productivity significantly. Earthworms and woodlice promoted the neutralization of acid (pH 3.45.5)
decay products and their alkalinization (to pH 6.88.6) due to digestion of
the soil and plant residues. They also influenced the total chemical composition of the soil, changing Gibbs free energy of soil (DG) from 967 to
992 or to 1063 kJ 100 g1. Further, Samedov and Nadirov (1990) also
hypothesized that earthworms and woodlice might participate in the formation of secondary soil minerals with low lattice energies.
Different to Hole (1981), Muller-Lemans and van Dorp (1996) reported
that it took less than 20 years for earthworms to turnover the topsoil of
European grasslands, which resulted in intensive and more or less homogeneous mixing of the topsoil. For earthworm activities, transport rates of
around 2 kg m 2 yr 1 (equivalent to 1.6 mm yr 1) of dry matter from deep
soil to top soil are listed.
In Amundson (2004) turnover rates of soil by earthworms were cited to
be of around 3 years for the upper 10 cm of a Canadian prairie soil and
around 700 years for the upper 50 cm of a soil in England, resulting in a
homogenization of the upper soil profile. Further, mixing rates of 360 years
for the upper 75 cm of soil were reported for ground squirrels in California.
Buchan (2010) noted that earthworms are able to digest approximately
90 t of soil per ha each year (equivalent to 7.5 mm yr 1). Comparing the
listed studies, we can conclude that earthworms seem to play a very active
role in intermixing the layers in the soil profile.
Gabet et al. (2003) reviewed rates of bioturbation caused by earthworms
(between 0.54 and 10 mm yr 1), ants (between 0.005 and 1.8 mm yr 1),
and termites (between 0.013 and 0.41 mm yr 1). They summarized that
earthworms are able to excavate tunnels up to 5 mm thick in the soil that
either collapse or promote the formation of macropores. The latter would
increase the soil porosity by 310 times. Different to earthworms they
suggested that ants are able to generate relatively deep burrows into the
soil. However, ants were classified as selective burrowers who base their
burrowing activity on the grain-size of the soil, accordingly being absent in
29
most very fine-grained soils. Gabet et al. (2003) also noted that vertebrates
usually dig burrows of up to 25 cm in diameter down to 30 cm below the
soil surface with dens reaching depths of up to 1 m. However, the role of
vertebrates in reaching and breaking up the parent material and therefore
promoting soil formation is assumed to be minor.
Yoo et al. (2005, 2007) proposed that an increasing number of fauna
participate in soil mixing as soil thickens, especially larger organisms like
wombats which were observed to preferentially habituate the base of
hillslopes. Naturally, thicker soils are a more preferable habitat for vertebrates, because of increased plant productivity, soil moisture, and space for
burrowing and nesting. At a study site in the Tennessee Valley in California
with known SPRs (Heimsath et al., 1999; McKean et al., 1993), Yoo et al.
(2005) investigated and modeled pocket gopher activities in relation to
sediment transport and soil thickness. Modeling results suggested that
pocket gophers spend only about 9 kJ of energy annually on producing
downslope transport of soil, which is equivalent to 1% of the total energy
expenditure on burrowing. However, it was presumed that most of the
energy is used for shearing, mixing, and elevating. Yoo et al. (2005)
estimated gopher induced gross sediment flux to be of 50140 cm2 yr 1,
which relates to soil turnover rates of only 40100 years for the first 50 cm
of soil of an area of 1 m2.
4.1.2. Floraturbation
Processes that initiate floraturbation are plant growth and decay and also tree
throw.
Gabet et al. (2003) reviewed the impact of plants on soil-mixing rates.
They reported that the axial (up to 1.45 MPa) and radial (up to 0.91 MPa)
pressure caused by root growth is sufficient enough to break up bedrock and
therefore contribute to physical weathering, and would also be substantial
enough to push up a column of approximately 100 m thick soil. It was also
presented that plant roots are able to enter soft rock matrices through small
cracks, promoting the breakdown of the rock. After decay plant roots leave
behind macropores that become important preferential pathways for subsurface water flows. Further, they stated that tree throw results in uprooting
and consequently in the excavation of soil, leaving a pit behind which is
prone to be refilled. Thus, tree throw is another agent that can promote the
homogenization of the surface or subsurface soil. Gabet et al. (2003) also
discussed a publication of Gill and Jackson (2000) who reported increased
turnover rates for roots with increasing temperature and therefore implied
that sediment flux by root growth and decay could increase with increasing
global temperatures.
Tonneijck and Jongmans (2008) investigated the vertical distribution of
organic matter in volcanic ash soils in Ecuador. They implemented a semiquantitative micromorphological analysis of soil faunal pedofeatures and
30
high-resolution radiocarbon dating. Results implied that bioturbation is

highly responsible for the vertical distribution of soil organic matter, with
illuviation and roots playing a minor role in transporting soil organic matter
in soils. However, the authors assumed that vertical transport of soil organic
matter via leaching plays an insignificant role in volcanic ash soils, because of
large metal to soil organic matter ratios which limited the mobility of soil
organic matter. Here, bioturbation did not result in the homogenization of
the topsoil and subsoil, but in a gradual increase of organic carbon in the soil
profile. This was explained by the presence of endogeic species, who tend to
move soil horizontally (Anderson, 1988), and transport material over short
vertical distances only. Movement over short vertical distances was also
explained by the occurrence of upward directed shifting of bioturbation in
response to soil thickening due to soil organic matter accumulation.
4.2. Soil creep

Soil creep defines the slow mass wasting process of soil on a slope, under the
influence of gravity. (Source: Glossary of Soil Science terms, Soil Science
Society of America.)
Saunders and Young (1983) listed various rates of soil creep in mm yr 1,
referring to the downslope movement of soil particles near the soil surface.
The highest rates were recorded for temperate climates predominantly
affecting soil movement in the surface soils down to the first 25 cm with
rates of about 0.52 mm yr 1 for the temperate maritime zone and generally higher rates for the temperate continental zone of about 210 mm yr 1.
For tropical environments, rates of approximately 45 mm yr 1 were listed.
Recorded rates of solifluction, a process often difficult to separate from soil
creep, were recorded to be as low as 1 mm yr 1 and as high as 300 mm yr 1
for polar and montane climate zones with most rates clustered around
10100 mm yr 1. It was also stated that the majority of rates of solifluction
affected soil movement down to 50 cm of depth.
Heimsath et al. (2002) used optically stimulated luminescence (OSL) to
investigate process of soil creep at a study site with known soil production in
the Bega Valley in south-eastern Australia. They defined soil creep caused
by burrowing agents (e.g., worms, ants, moles) and tree throw as the main
factor for creeping soil. With the help of OSL, they determined when soil
grains had last visited the soil surface and measured vertical soil-mixing rates
of 0.10.4 mm yr 1. Results indicated that the amount of grains that have
visited the surface decreased with increasing soil thickness. Heimsath et al.
(2002) used a Monte Carlo simulation to demonstrate particle transport.
Results of the simulation assumed that soil creep involved independent
movement of mineral grains throughout the soil body and that the grains
were reburied or eroded by overland flow upon reaching the soil surface.
31
Kaste et al. (2007) calculated diffusion-like mixing rates of 0.10.2 mm yr 1

for the Bega Valley in south-eastern Australia using fallout radionuclides
(7Be and 210Pb). These mixing rates are comparable to mixing rates of quartz
grains of 0.10.4 mm yr 1 calculated by Heimsath et al. (2002).
4.3. Rainsplash
Lobry de Bruyn and Conacher (1990) cited annual erosion rates caused by
rain splash of 0.25 mm yr 1.
4.4. Modeling pedoturbation

In their review, Gabet et al. (2003) provided quantitative models of bioturbation and sediment transport. They formulated a general slope-dependent
model to determine the horizontal volumetric flux of sediment (qsx) caused
by root growth and decay.
qsx
xrt
;
rr
17
where x (m) is the net horizontal displacement of soil, r (kg m 2) is the root
mass per unit area, t (per year) is the root turnover rate, and rr (kg m 3) is
the density of root material. They then calculated values for sediment flux
by root growth and decay for three different types of vegetation with
2.1 10 4 m2 yr 1 for temperate grasslands, 6.8 10 4 m2 yr 1 for
sclerophyllous shrubs, and 8.8 10 4 m2 yr 1 for temperate forests.
Gabet et al. (2003) also quantified the horizontal alteration of soil along a
hillslope caused by tree throw applying the following equations:
vol
distance events events

;
event
event
area
time
2
xn xd xu W Dsin y;
p
qs
18
19
where qs is the horizontal sediment flux, xn is the long-term net horizontal

transport distance, xd the horizontal distance of displacement of the root
plate centroid caused by trees that were falling directly downslope, xu the
horizontal distance of displacement of the root plate centroid caused by trees
that were falling directly uphill, W the width of the root plate, and D the
depth of the excavated pit.
Gabet et al. (2003) calculated horizontal sediment flux (qs) of
8 10 4 m2 yr 1 on a 10 slope by substituting Eq. (19) with Eq. (18)
for an excavated pit width of 4 m, a pit depth of 0.7 m, a mound volume of
32
4 m3, and an uprooting rate of 4 trees ha 1 yr 1 (with qsx equal to
4.8 10 3 m 2 yr 1 sin y).
Salvador-Blanes et al. (2007) considered bioturbation processes and
the redistribution of soil particles in their model of soil profile evolution.
In considering the addition and removal of soil particles to or from
the soil profile and also their translocation within the soil profile, the
process of horizonation was implemented in the model. Soil mixing
was modeled by incorporating parameterized bioturbation processes resulting from earthworm, ant, and termite activity (assumed surface casting of
15 kg m 2 yr 1, cited from Paton et al., 1995). Due to a lack of parameterization, illuviation, and eluviation processes were not taken into account.
However, running the model resulted in stone layers in the soil profile,
which was attributed to high mixing velocities because of bioturbation.
4.5. Summary
Soil-mixing rates by faunal agents reported in the literature are relatively fast
compared to the earlier discussed rates of chemical and physical weathering
(Fig. 8):
Lobry de Bruyn and Conacher (1990) published soil-mixing rates from
ants and termites ranging between approximately 0.002 and 0.950 mm
(2) Soil mixing

rates
(1) Chemical
weathering rates
6
8
Rates (mm/yr)
10
12
Figure 8 A comparison of soil-mixing rates and chemical weathering; the line in the
middle of the boxes refers to the median of the distribution.
33
of soil turnover each year (Appendix, Table A3). Soil-mixing rates from
ants and termites taken from the selection of other published data range
from 0.002 to 1.8 mm yr 1. However, soil turnover rates by ants and
termites seem to be of similar orders of magnitude for the compared
studies (Fig. 9).
Similar to ants and termites, most soil-mixing rates caused by earthworms fell into the range of about 0.54 mm of soil turnover each year,
although much higher rates of 7.5 and 10.6 mm yr 1 were also
recorded. Higher rates of soil alterations by earthworms may occur
because unlike ants and termites earthworms also ingest the soil material, in addition to moving soil particles by mounding and burrowing.
Further, turnover rates of the whole topsoil by earthworms were
reported to be as fast as 300700 years and resulted in a homogenization
of the soil material.
Soil turnover rates recorded for vertebrates range from 0.08 to
1.5 mm yr 1 and are very similar to mixing rates recorded for ants,
termites, and earthworms.
In situ measured rates of soil creep predominantly caused by burrowing
agents and tree throw are much lower than reported assumed turnover
rates by ants, termites, and earthworms. These rates, calculated by
(7) Soil creep,
solifluction
(6) Rainsplash
(5) Vertebrates
(4) Termites
(3) Earthworms
(2) Cicades
(1) Ants
0
4
6
8
Soil mixing rates (mm/yr)
10
12
Figure 9 Box-plot distributions of soil-mixing rates from field data (presented in

Table A3, Appendix). Values of soil-mixing rates in mm yr 1 are plotted separately
for discussed soil-mixing agents; the line in the middle of the boxes refers to the median
of the distribution.
34
applying sophisticated laboratory methods to investigate pedogenic

processes in situ, range between 0.1 and 0.4 mm of soil turnover each
year. However, rates of soil creep determined using other methods
are either very similar to mixing rates from ants, termites, and earthworms with 0.55 mm yr 1, or exceed these rates by far, that is,
10300 mm yr 1 for polar and montane environments.
Comparing published rates of soil turnover with chemical weathering
rates (Fig. 8) demonstrated that the range in turnover rates of soil by
mixing agents exceeds the range of chemical weathering rates
significantly.
5. Models of Soil Formation Based on the

Concept of Mass Balance
Models of soil formation in the landscape couple lateral transport laws
from the field of geomorphology with pedogenic processes. For natural
environments, downslope movement is generally induced by biological and
physical processes, aided by gravity (Amundson, 2004). Soil transport is then
modeled as random diffusive-like transport or oriented down a slope gradient. In the following, the basis for landscape evolution models is presented,
followed by the implementation of soil in landscape evolution models.
5.1. Landscape evolution models

Landscape evolution usually results from processes of erosion and deposition. In particular, landscape evolution models replicate long-term hillslope
evolution via the principles of the conservation of mass. Additionally they
use transport algorithms to describe the movement of material on slopes
(Ahnert, 1977; McKean et al., 1993; Willgoose, 2005). Recent landscape
evolution models have successfully coupled hillslope evolution with channel and tectonic processes to model the evolution of natural, undisturbed
landscapes spatially as a whole rather than restricted solely to hillslope
development (Willgoose, 2005).
As described in Minasny et al. (2008), landscape evolution models
simulate the downslope transport of sediment with the help of the following
simple continuity equation, which describes diffusive-like transport:
qs Krz;
20
where qs is the sediment flux, which is proportional to the slope rz, and K
is an equivalent to a diffusion coefficient with dimensions (length)2 (time) 1
and z is the elevation. The simplest model simulating soil formation in the
35
landscape implements the change in elevation (soil thickness) as a function

of material transport (sediment flux):
@z
rqs ;
@t
21
where z is the elevation (L), t is the time (T), qs is the material flux (L3 T 1),
and r is a partial derivative vector.
Carson and Kirkby (1972) are seen as one of the first researchers to
introduce the change of soil thickness into the continuity equation of
hillslope evolution models (Eq. (21)). For the transport of soil on a hillslope,
the equation has the written form: (soil transport in) (soil transport out
over a unit length of slope profile) (increase of soil thickness due to
expansion during weathering of bedrock) (decrease of elevation of land
surface). The differential equation of the described formula is:
@S
@y
m 1W ;
@x
@t
22
where S is the mean soil transport, x is the horizontal distance from the
divide, m is the volume of mineral soil produced from the weathering unit
volume of bedrock, W is the rate of lowering of bedrock surface through
weathering, y is the elevation, and t is the time. For the change in soil
thickness, the continuity equation has the written form: (increase of soil
depth) (increase of elevation of land surface) (reduction of elevation
of bedrock) or in differential equation form:
@z @y
@S
W mW
;
@t
@t
@x
23
where @z/@t is the rate of soil increase with z as the soil thickness.
Carson and Kirkby (1972) also presented two limiting processes for the
rate of weathering and material transport along a hillslope. The first process
is weathering limited; hence, the transport process is faster than the
weathering rate. The second process is transport limited; hence, the rate
of weathering is faster than the transport process. If the weathering rate (soil
production) and the transport rate (erosion) equal each other, then the soil
depth is at a local steady state (Burke et al., 2007). Both equations, described
above, can be applied to transport-limited or weathering-limited removal.
Dietrich et al. (1995) and Heimsath et al. (1997, 1999) introduced soil
into the continuity equation of mass transport along a hillslope (Eq. (21)):
rs
@h
@e
rr rqs ;
@t
@t
24
36
where h is the soil thickness, rs and rr are the bulk densities of soil and rock,
e is the elevation of the bedrock-soil interface, t is time, and qs is the
sediment flux in the horizontal direction. Therefore, the change in soil
thickness at any point on the hillslope is a function of the divergence of the
sediment flux, and the rate of conversion from parent material to soil.
If the steady-state soil thickness is reached (@h/@t 0), then soil production (@e/@t) can be described as:
@e
r
s Kr2 z;
rr
@t
25
where z is the ground surface elevation. The key variables that drive soil
thickness on hillslopes are assumed to be the slope curvature, the transport
coefficient, the SPR, and time (i.e., Amundson, 2004).
5.2. Modeling soil formation in the landscape

In 1975, Yaalon discussed conceptual models of pedogenesis and raised the
question, can soil-forming functions like Jennys (1941) be solved? Yaalon
(1975) expressed the need for collecting field data to model soil formation
mechanistically. Therefore, he stated that the future of modeling soil formation relies on replacing statistically based models with physically more
significant mathematical models.
Yaalon (1975) emphasized the importance of implementing spatial differentiation (catenation), the relief factor (catenary slope), material fluxes on
a given pedomorphic surface (vertical and lateral), and different parent
materials and time steps in modeling soil formation. Within the proposed
model, he considered vegetation as the dependent variable of the ecosystem,
varying with soil properties.
Huggett (1975) proposed the first mechanistic model of soil formation in
the landscape. His publication can be seen as the first application of landscape evolution modeling in soil science. Huggett (1975) introduced the
concept of a homomorphic model to simulate the soil system on the catena
scale over millennial time scales. A homomorphic modeling approach
groups . . .several components. . . of the soil system . . .to form one single
element in the model. . ..
More importantly, he suggested simulating soil formation in a 3D soil
profile. He also proposed to simulate soil in form of layers more or less
organized parallel to the parent material. For instance, within the soil profile
the rate of weathering should decrease with depth, whereas the bulk density
should increase. Huggett (1975) stated that in terms of landscape evolution
contours that concave downslope should lead to convergent flowlines and
convex contours should lead to divergent flowlines and that flow should tend
37
to converge in hollows and diverge over spurs. All thalweg and valley basins
should join in one complex network, based on 1st and 2nd order stream lines.
Kirkby (1977, 1985) presented the first comprehensive mechanistic
model of soil profile development in interaction with a hillslope model
based on mass balance, which can be seen as a valuable basis for quantitative
soil profile modeling or even soil catena modeling. To predict the development of soils over time quantitatively, the mass-balance model estimates the
flow of water through the soil and the uptake of nutrients from the soil. The
soil profile development model consists of three parts, the weathering
profile, the inorganic profile, and the organic profile.
Kirkbys soil profile model is based on the accumulated amount of soil
deficit, w, which represents the amount of parent rock converted into a soil
mantle:
1
1 P dz;
26
z0
where P is the proportion of bedrock remaining in the soil profile

(unweathered rock) and z is the depth below the soil surface. P unweathered parent material, takes values between zero and one.
The mass-balance model for the accumulated soil layer of the soil profile is

@w
@
1 PS
;
27
J S
@t
@x
PS
where PS are the properties remaining in the part of the soil profile where
mechanical erosion occurs (mostly at the surface), x is the distance from the
divide, t is the time, S is the mechanical sediment transport measured in the
x direction, and J is the chemical sediment transport.
Within the model inorganic minerals are treated as mixtures of elementary
oxides, which dissolve independently to create ions (Gibbs free energy to
calculate equilibrium reactions). In the model, the soil organic matter material
experiences production, accumulation, vertical mixing, and decomposition.
Mixing rates of organic matter act as a diffusion process; transporting organic
material from areas of high to those of low concentration. At the surface
accumulated organic matter is transferred downward through vertical mixing
in balance with an upward mixing of inorganic material. The most important
product of decomposition of organic matter is CO2. Soil CO2 distribution is
simulated with gaseous diffusion of the produced CO2. The soil pH is
calculated as a result of this CO2 distribution.
Kirkby (1985) enhanced his soil profile model by also simulating the
processes of percolation, equilibrium solution, leaching of solutes, ionic
diffusion, organic mixing, leaf fall, organic decomposition, and mechanical
denudation. Each process leads to second-order linear partial differential
equations.
38
Minasny and McBratney (1999, 2001) introduced a basic mechanistic

quantitative soil-formation model based on the conceptual ideas of Huggett
(1975) and Heimsath et al. (1997). Their soil-formation model considers soil
formation spatially at the catena scale and is written for predicting regolith
depth and soil formation at the landscape scale over millennial time scales,
simulating soil formation over tens of thousands of years. The continuity
equation of the model is based on the assumption that the change of soil
thickness over time depends (1) on the production of soil from the
weathering of parent material and (2) on the transport of soil through
natural surface erosion. The rate of weathering and the erosive diffusivity
are the main parameters used in the soil-formation model. The weathering
of bedrock is characterized as an exponential decline with increasing soil
cover thickness based on Eq. (11), applying concepts from Dietrich et al.
(1995) and Heimsath et al. (1997). It is assumed that the weathering rate of
bedrock to soil, P0, is mainly controlled by the climate whereas the empirical constant, b, is believed to depend on the thermal properties of the rock
or soil. The transport rate of materials is defined similar to Darcys law for
water transport in soils whereas the movement of materials within the
landscape is expressed in terms of diffusive transport. The erosive diffusivity
of the soil material is assumed to be controlled by factors of soil erosion, soil
physical properties, and climate. Running the soil-formation model over
millennial time scales showed that the initial SPR is high and that it slows
down exponentially until it reaches steady state at around 40,000 years,
applying a value for P0 of 0.19 mm yr 1.
Based on the soil-formation model of Minasny and McBratney (2001),
Salvador-Blanes et al. (2007) improved the modeling of in situ long-term soil
profile evolution by implementing particle size transformation, horizonation (layers of regolith formed from bedrock with each time step) and
bioturbation processes. Horizonation is introduced in the model by taking
into account the translocation of particles within the soil profile. However,
due to the influence of bioturbation processes, running this model resulted
in the formation of stone layers in the soil profile. Running the model also
showed that after 10,000 years a soil profile of about 1.25 m is formed. This
profile, however, lacked any horizonation. After a simulation time of
20,00080,000 years the thickness of the soil profile increased to a depth
of 1.80 m and by that time period a total of three horizons were developed
in the soil profile.
The soil-formation models discussed predict soil profile development in
situ or along a hillslope starting from parent materials and ending with the
formation of soil being in equilibrium, which was simulated over tens of
thousands of years and modeling steps of 1001000 years.
In the following soil-formation models, the evolution of soil is not
simulated as starting from in situ bedrock. The presence of soil is assumed
already and processes that lead to the formation of the soil profiles studied
39
are simulated. Generally, these models predict soil in yearly time steps up to
1000 years. They are not designed to simulate soil profile development from
parent rock over millennial time scales, but to model processes in soil.
Sommer et al. (2008) presented a soil-landscape genesis model that
simulates the evolution of a hummocky agricultural landscape with a timesplit modeling approach. Fallout radionuclide data indicated regressive processes of pedogenesis during the past 50 years for the studied landscape. For
the progressive period, soil formation was therefore modeled by simulating
pedogenic processes such as decarbonization and carbonization, silicate
weathering, and clay translocation that were quantified by a spatially
distributed, coupled water flow, and solute transport model. For the modeling of the regressive period, a modified version of the dynamic, physically
based EROSION-3D model was used, which allows the spatially distributed
modeling of soil erosion and soil deposition. In their modeling approach,
Sommer et al. (2008) coupled geologicalhistorical information with spatially distributed pedogenic processes that experienced recent changes to
better simulate the current development of the landscape studied.
Finke and Hutson (2008) introduced a model to simulate formation of
calcareous soils formed on loess (15,000 BP to present). Model simulations
were based on Jennys factors of soil formation ( Jenny, 1941) using the
LEACHC model. Additional soil-forming processes were implemented to
describe the effect of the soil fauna and flora on various soil properties,
assuming an annual rate of bioturbation of 30 t ha 1 (2.5 mm yr 1). Running the model showed that it takes about 1297 years to decarbonize 1 m of
loess parent material. Results also showed a clear effect of bioturbation on
soil formation. Considering bioturbation processes in the model reduced the
decarbonization time of the topsoil and affected clay migration processes.
5.3. Summary
The models of soil formation reviewed couple lateral transport laws from
modeling hillslope formation in the landscape with pedologic processes.
These models simulate soil formation in a quantitative-mechanistic sense
based on principles of mass balance.
However, some model soil formation over millennial time scales starting
from parent materials and ending with the formation of soil being in
equilibrium whilst others simulate soil processes that lead to the formation
of a certain soil type over centennial time scales, already assuming the
presence of the soil body.
An overview of all models presented and discussed in this literature
review is given in Table 2. In Table 2, models of pedogenesis are summarized according to the concepts applied in the models listed.
Table 2 Summary of models of soil formation presented in this literature review (factors, processes, pathways, energy, and mass-balance
soil-formation models)
Model concept
Model description
Model type
Reference
Factorial Energy
Soil-forming factors
S f(cl,o,r,p,t,. . .)
Qualitative (empirical)
Soil-forming factors
S f(s,c,o,r,p,a,n); S f(time)cl,o,r,p,. . .
Quantitative-empirical
Dokuchaev (1886), Shaw

(1930), Jenny (1941)
McBratney et al. (2003);
chronofunctions
Simonson (1959)
Soil-forming processes S f(addition, removal, translocation,

transformation)
Soil evolution along
S f(progressively, regressively)
pathways
Energy
Soil energy model
S f(o,w,t)
Soil energy model
Soil formation is modeled through
quantifying processes of energy, that is,
radiation, kinetic energy, etc.
Soil energy model
Soil formation is modeled via energy
expenditure
Soil energy model
Thermodynamic aspects in soil-forming
processes, that is, Gibbs free energy and
entropy of soil minerals
Mass balance
SPR P0exp( bh)
Soil weathering:
Soil-production
verified with field data
function
Qualitative
Qualitative
Johnson and Watson-Stegner

(1987)
Qualitative
Mechanistic
Runge (1973)
Regan (1977)
mechanistic
mechanistic
Volobuyev (1974)
Volobuyev and Ponomarev
(1977)
Quantitative-mechanistic Heimsath et al. (1997)
@e
@t
P0 expk1 h expk2 h Pa Quantitative-mechanistic Dietrich et al. (1995), Furbish

and Fagherazzi (2001),
Minasny and McBratney
(2006)
W D(1 [Zr]rock/[Zr]soil)
Soil weathering:
Quantitative-mechanistic Riebe et al. (2003, 2004a,b),
Coupling chemical
Burke et al. (2007, 2009),
physical weathering
Green et al. (2006), Yoo et al.
(2007)
Pedoturbations:
Monte Carlo simulation of independent
Quantitative-mechanistic Heimsath et al. (2002)
Soil creep
movement of soil grains in soil profile
Soil weathering:
The hump function
Soil landscape model

Soil landscape model
Soil profile model
Soil profile model
Soil process model
Catena scale, 3D, homomorphic; lateral

and vertical fluxes of material
Soil formation over millennial time scales,
coupling weathering, diffusive transport
and erosive diffusivity, fluxes simulated
within the system
Soil profile formation in interaction with
hillslope processes, fluxes simulated
within the soil system
Soil profile evolution over millennial time
scales, implementing horizonation and
bioturbation, fluxes simulated within the
system
Soil landscape evolution via simulating soilforming processes over centennial time
scales, that is, silicate weathering, soil
erosion
Quantitative-mechanistic Huggett (1975)

Quantitative-mechanistic Minasny and McBratney (1999,
2001), Dietrich et al. (1995),
Heimsath et al. (1999)
Quantitative-mechanistic Kirkby (1977, 1985)
Quantitative-mechanistic Salvador-Blanes et al. (2007)
Quantitative-mechanistic Sommer et al. (2008), Finke and

Hutson (2008)
Models are grouped based on the concepts used and the type of the model (based on Minasny et al., 2008).
42
6. Conclusions
In pedology, most studies have focused on the prediction of soil
properties from landscape attributes at specific sites based on empirical
quantitative relationships (McBratney et al., 2003). However, to understand
and improve knowledge of pedogenesis, a mechanistic model that can
simulate the development of soil under various conditions or scenarios
needs to be formalized. This growing movement in pedology has directed
the focus towards mechanistic modeling of soil formation. From this review
we have learned that in order to verify mechanistic models of soil formation,
it is essential to quantify pedogenic processes with field data.
By reviewing research related to this topic, we can conclude that
processes of pedogenesis have been estimated from field data, particularly
rates of soil weathering and rates of soil mixing. However, we also learned
that the majority of these estimates were based on watershed studies or
observational data and that it is necessary to further investigate soil formation processes in situ to develop models of soil formation further and to be
able to model better soil evolution.
Colman and Dethier (1986) discussed that the largest constraint in
estimating weathering rates (chemical and/or physical) from field studies is
the lack of independent age estimates of the weathered materials studied. In
recent years, sophisticated laboratory techniques have become available that
can be used to quantify such processes of soil formation in situ.
In situ terrestrial cosmogenic nuclides (TCNs) and optically stimulated
luminescence (OSL) are such tools to investigate pedogenesis quantitatively
via numerical dating. Both techniques have been used widely in the fields of
geology and geomorphology until they became accessible for pedology
related studies. TCN and OSL can now be applied to investigate longterm processes of pedogenesis on millennial time scales; and OSL can also be
used to investigate short-term processes of pedogenesis on a time scale of a
few hundred years. The concentration of TCN in soil parent materials can
be used to investigate processes of soil weathering in situ by deriving rates of
soil production. Dating individual sand-sized quartz grains with OSL can be
applied to explore processes of soil mixing by estimating rates of potential soil
turnover, and also to estimate the age of soil horizons.
In the following paragraphs, we will answer and reflect on our questions
about pedogenesis formulated in Section 1 to this literature review:
(1) How does soil form?
We know that soil forms via a combination of interrelated physical,
chemical, and biological processes.
Traditional concepts in pedology explain soil formation progressively
depending on environmental factors and processes along a sequence of
preset soil types with soil formation eventually reaching equilibrium.
43
Further, soil profile evolution is described as a product of mostly downward moving processes that lead to the formation of interrelated layers or
horizons. Alternately, soil profile evolution is described as depending predominantly on a combination of bioturbation and slope processes.
Chronofunctions have been applied to explain soil formation along linear
and nonlinear functions, that is, exponential, polynomial, or sigmodial behavior. However, we learned that the formation of soil from parent materials was
predominantly explained by two concepts: (1) the exponential soil-production model and (2) the humped soil-production model. More recently,
sophisticated laboratory methods have made it possible to verify and parameterize the concept of exponential decline of soil formation with increasing soil
thickness, using field data over millennial time scales. Soil production following a humped model with maximized soil production at an intermediate soil
depth was assumed for some field sites, but a parameterization of this function
is still lacking. Evidence for the humped concept comes from observing high
rates of chemical and physical weathering under initial soil cover in the field.
(2) At what rate does soil evolve?
We learned that rates of soil formation are initially fast and decrease with
time. We also learned that field data have confirmed the concept of
exponential decline of weathering of parent materials to soil with increasing
soil thickness in situ (TCN-derived). Estimated rates of soil production from
field data were as low as 0.004 mm yr 1 and as high as 0.4 mm yr 1.
Modeling soil formation mechanistically over millennial time scales by
implementing an exponential soil-production model with relatively high
rates of weathering of bedrock to soil of 0.19 mm yr 1 resulted in steady
state after around 40,000 years.
Results from studies comparing the rate of chemical and physical
weathering in situ showed that chemical weathering rates were generally
lower, accounting for less than 50% of the total weathering rate. Mean rates
of chemical weathering were recorded to be about 0.022 mm yr 1 with a
maximum rate of 0.14 mm yr 1.
(3) How fast are rates of soil turnover occurring in the soil profile, and
what influence do they have on pedogenesis?
We learned that mixing rates of soil are much faster than rates of soil
production from parent materials. Mixing rates as high as 10.6 mm yr 1
were recorded and we learned that it is assumed that rates of bioturbation
exceed SPRs by up to 3 orders of magnitude. Particularly, earthworms seem
to play a very active role in turbating the soil profile.
We can also conclude that in most of the reviewed pedoturbation
studies, especially bioturbations impeded the horizonation of a soil profile
by homogenizing the soil material. However, bioturbation processes promoting the formation of horizons in the soil profile were also proposed.
44
BOX 5 What defines soil science?
Churchman (2010) from the University of Adelaide identified three aspects

that are unique to the discipline of soil science, these aspects distinguish soil
science among scientific disciplines generally. These are
(1) the formation and properties of soil horizons,
(2) the occurrence and properties of aggregates in soil, and
(3) the occurrence and behavior of soil colloids.
He added that the possibility that these aspects could be reduced to other
sciences is dismissed because they can be explained more usefully at a larger
size scale or by a more complex context than those belonging to more basic
sciences such as physics or chemistry. The three unique aspects are considered to comprise a research tradition for soil science, which, due to their
ultimate irreducibility, is constituted as a special science. These unique
aspects cross sub-discipline boundaries so that both soil science and soils
should be considered holistically rather than via the separate sub-disciplines
through which they have often been studied in the past.
Determining rates of soil-mixing in situ by dating individual soil particles

(OSL, fallout radionuclides) resulted in rates of soil turnover of about
0.2 mm yr 1. These rates are lower than those estimated from observational
field studies, where soil turnover was generally recorded to be higher than
0.5 mm yr 1, for a variety of mixing agents.
Implementing relatively high rates of soil turnover of up to about
4 mm yr 1 in mechanistic models of soil formation, however, resulted in
the formation of stone layers in the soil profile. Does this result implicate
that high rates of bioturbation are questionable?
However, within this review it became apparent that work on quantifying soil-forming processes in situ to verify and improve mechanistic models
of soil formation is still minimal and therefore vital (i.e., Bockheim and
Gennadiyev, 2009; Pelletier and Rasmussen, 2009).
Ultimately, a soil model should be able to simulate the formation and
generation of the three unique aspects in soil science of soil horizons, soil
aggregates, and processes of soil colloids which is presented further in Box 5.
ACKNOWLEDGMENTS
We would like to thank Dr. Tom Vanwalleghem and Dr. Stephen Cattle for their useful
comments on the chapter; Dr David Hammer for suggesting a possible link between Shaws
work, the 1930 World Conference and Jennys formulation, and Dr. Igor V. Florinsky for
his clarification on Dokuchaevs soil-formation formula.
Appendix
Table A1, part 1 Chemical weathering rates (CWR) derived from elemental fluxes (loss and gain) in watersheds and chemical products of
the parent materials and weathering products studied
Reference
ID
CWR
(mm
kyr 1)
Alexander
(1985)
Maryland
5.81
0.00581
California
California
Maryland
Wales
Wales
British Columbia
Luxembourg
Hong Kong
Hong Kong
Hong Kong
England
England
Virginia
Virginia
Java
England
Idaho
10.79
4.98
19.92
40.67
13.28
54.78
19.92
6.64
5.81
9.96
11.62
14.94
18.26
83
116.2
38.18
64.74
0.01079
0.00498
0.01992
0.04067
0.01328
0.05478
0.01992
0.00664
0.00581
0.00996
0.01162
0.01494
0.01826
0.083
0.1162
0.03818
0.06474
CWR
(mm yr 1)
Elevation
(m)
Climate
Geology
(Continued)
Table A1, part 1
Reference
Alexander
(1988)
(Continued)
ID
CWR
(mm
kyr 1)
CWR
(mm yr 1)
Elevation
(m)
Wyoming
Wyoming
New Mexico
New Mexico
New Mexico
Hawaii
Mexico
Mexico
Papua
Wyoming
Colorado
CA
91.3
166
24.9
49.8
240.7
83
91.3
298.8
282.2
132.8
58.1
11
0.0913
0.166
0.0249
0.0498
0.2407
0.083
0.0913
0.2988
0.2822
0.1328
0.0581
0.011
220
Adamellite
CH, BC, Can.

CS, BC, Can.
England
France
Idaho
Luxemburg
Maryland
Minnesota
Maryland
Maryland
60
78
55
106
65
46
37
20
5.8
20
0.06
0.078
0.055
0.106
0.065
0.046
0.037
0.02
0.0058
0.02
185
945
232
152
478
90
280
98
67
70
Till/quartz diorite
Till/quartz diorite
Granite
Gneiss
Adamellite
Metashale
Greenstone (metabas.)
Till/gabbro
Schist
Serpentinite
Climate
Geology
Cleaves,
1993
Dethier,
1986
Edmond
et al.,
1995
Owens and
Watson,
1979
Paces, 1986
N. Hamp.
Scotland
V1, Victoria
V2, Victoria
Wash.
WF, Wales
WP, Wales
ZJ, Zimbabwe
ZR, Zimbabwe
Baltimore,
Maryland
Pacific Northwest
29
160
33
57
85
26
19
18
1.7
9.1
0.029
0.16
0.033
0.057
0.085
0.026
0.019
0.018
0.0017
0.0091
300
542
144
203
1150
105
90
160
60
0.003
Guayana Shield,
South America
10
0.01
Rhodesia-Juliasdale
4.4
0.0044
1900
Tropical
Granite
Rhodesia-Rusape
Bohemian Maffif,
X-0
0.17
8.9
0.00017
0.0089
1600
724
Tropical
Warm temperate
X-8
14
0.014
744
Warm temperate
X-7
32
0.032
635
Warm temperate
Granite
Biotitic gneiss with
muscovite, sillimanite,
quarzites
quarzites
quarzites
Warm temperate
Cold temperate
Till/gneiss
Till/granite
Dacite
Dacite
Till/quartz diorite
Till/wacke
Till/wacke
Adamellite
Adamellite
Plagioclase-muscovitequartz
Sedimentary and
metamorphic
Granite
(Continued)
Table A1, part 1
(Continued)
CWR
(mm
kyr 1)
CWR
(mm yr 1)
Elevation
(m)
Climate
Geology
250
Cool temperate
Occoquan granite
Metamorphic rocks
Reference
ID
Pavich, 1986
Pavich, 1989
Virginia Piedmont
Appalachian
Piedmont
Appalachian
Piedmont
Papua New Guinea
4
20
0.004
0.02
0.004
58
0.058
Humid tropical
Volcanic material
Watershed 27,
Coweeta
Southern Blue
Ridge
Hubbard Brook
37
0.037
Warm temperate
Garnet, plagioclase
37
0.037
Cold temperate
Gneiss
47
0.047
Cold temperate
Iu, granitic
Iu, basic pyroclastic
224
473
0.224
0.473
Humid subtropical
Humid subtropical
Ruxton,
1968
Velbel, 1986
Wakatsuki
and
Rasyidin,
1992
Metamorphic rocks
Table A1, part 2 Chemical weathering rates (CWR) derived in situ in applying weathering indices and TCN-derived total denudation rates for
study sites in and outside Australia (rates of t km2 yr1 were converted to mm yr1 in assuming a bulk density of soil of 1200 kg m3)
Reference
ID
Soil
thickness
(cm)
Burke et al.
(2007)a
Pit 6
15
35 m Myr 1
0.035
59 m Myr 1
0.059
Pit 20
Pit 9
Pit 7
Pit 17
Pit 19
Pit 5
Pit 16
Pit 8
Pit 15
Pit 25
Pit 26
Pit 14
Pit 10
Pit 1
Pit 24
Pit 12
Pit 21
FH1
15
29
40
40
40
48
55
60
70
70
70
75
80
90
100
120
120
52
44 m Myr 1
13 m Myr 1
31 m Myr 1
6 m Myr 1
17 m Myr 1
27 m Myr 1
5 m Myr 1
4 m Myr 1
2 m Myr 1
4 m Myr 1
4 m Myr 1
1 m Myr 1
7 m Myr 1
1 m Myr 1
1 m Myr 1
1 m Myr 1
1 m Myr 1
3.9 1.2 m Myr 1
0.044
0.013
0.031
0.006
0.017
0.027
0.005
0.004
0.002
0.004
0.004
0.001
0.007
0.001
0.001
0.001
0.001
0.004
59 m Myr 1
58 m Myr 1
50 m Myr 1
50 m Myr 1
50 m Myr 1
50 m Myr 1
36 m Myr 1
36 m Myr 1
18 m Myr 1
18 m Myr 1
18 m Myr 1
18 m Myr 1
28 m Myr 1
14 m Myr 1
11 m Myr 1
11 m Myr 1
11 m Myr 1
18.7 1.0 m Myr 1
0.059
0.058
0.050
0.050
0.050
0.050
0.036
0.036
0.018
0.018
0.018
0.018
0.028
0.014
0.011
0.011
0.011
0.019
FH2
FH3
50
55
11.7 0.6 m Myr 1

8.5 0.7 m Myr 1
0.012
0.009
19.5 1.0 m Myr 1

17.6 1.0 m Myr 1
0.020
0.018
Burke et al.
(2009)b
CWR (published rate)
CWR
(mm yr 1)
Total denudation
rate (published rate)
Total
denudation rate
(mm yr 1)
(Continued)
Table A1, part 2
Reference
(Continued)
ID
FH4
FH5
FH6
FH7
FH8
FH9
FH10
FH-B1
FH-B2
FH-B3
FH-B4
FH-B5
FH-2MP
Highland
mean (from
above)
NR0
NR1
NR2
NR3
NR4
NR5
NR6
NR7
NR-A1
CWR
(mm yr 1)
Total denudation
Total
denudation rate
(mm yr 1)
65
40
104
58
65
48
55
50 (96)
30 (100)
70 (120)
70 (120)
25 (60)
35
55
7.3 0.5 m Myr 1

15.2 0.7 m Myr 1
3.5 0.2 m Myr 1
8.6 0.6 m Myr 1
6.9 0.6 m Myr 1
9.5 0.8 m Myr 1
0.0 m Myr 1
5.0 1.1 m Myr 1
11.0 1.41 m Myr 1
5.0 0.5 m Myr 1
7.0 0.4 m Myr 1
11.0 0.7 m Myr 1
7.8 1.5 m Myr 1
8.1 0.8 m Myr 1
0.007
0.015
0.004
0.009
0.007
0.010
0.0
0.005
0.011
0.005
0.007
0.011
0.008
0.008
14.4 1.0 m Myr 1

23.8 2.2 m Myr 1
6.6 1.0 m Myr 1
16.6 1.0 m Myr 1
14.4 1.0 m Myr 1
20.3 1.2 m Myr 1
17.6 1.0 m Myr 1
17.6 1.0 m Myr 1
27.4 4.5 m Myr 1
11.4 1.0 m Myr 1
11.4 1.0 m Myr 1
30.6 4.5 m Myr 1
24.5 5.0 m Myr 1
19.7 1.7 m Myr 1
0.014
0.024
0.007
0.017
0.014
0.020
0.018
0.018
0.027
0.011
0.011
0.031
0.025
0.020
65
75
72
72
49
30
58
75
60
7.6 0.5 m Myr 1

6.9 0.4 m Myr 1
7.2 0.2 m Myr 1
9.5 0.4 m Myr 1
12.0 0.6 m Myr 1
15.6 1.0 m Myr 1
9.7 0.5 m Myr 1
6.7 0.4 m Myr 1
7.2 0.6 m Myr 1
0.008
0.007
0.007
0.010
0.012
0.016
0.010
0.007
0.007
14.4
11.8
12.6
12.6
19.9
29.1
16.6
11.8
14.2
3.0 m Myr 1
2.0 m Myr 1
2.0 m Myr 1
2.0 m Myr 1
4.0 m Myr 1
5.0 m Myr 1
2.0 m Myr 1
1.0 m Myr 1
1.0 m Myr 1
0.014
0.012
0.013
0.013
0.020
0.029
0.017
0.012
0.014
Soil
thickness
(cm)

Green et al.
(2006)
NR2-MP
SN-B1
SN-B1B
SN-B3
SN-B4
SN-B5
SN-B6
H2B1
H2B2
Lowland mean
(from
above)
Transport
distance 0 m
Transport
distance
315 m
Transport
distance 15
30 m
Transport
distance 30
45 m
Transport
distance
> 45 m
50
50
65
40
40
55
60
70
60
58
11.2 0.5 m Myr 1

11.1 0.6 m Myr 1
6.6 0.6 m Myr 1
12.2 0.9 m Myr 1
8.8 0.0 m Myr 1
8.8 0.7 m Myr 1
11.8 0.3 m Myr 1
3.5 0.4 m Myr 1
6.5 0.6 m Myr 1
9.1 0.5 m Myr 1
0.011
0.011
0.007
0.012
0.009
0.009
0.012
0.004
0.007
0.009
N.A.
20.0 1.2 t km 2 yr 1
0.017
N.A.
28.6 1.0 t km 2 yr 1
0.024
N.A.
29.4 1.6 t km 2 yr 1
0.025
N.A.
23.0 1.3 t km 2 yr 1
0.019
N.A.
19.3 2.1 t km 2 yr 1
0.016
17.6 2.0 m Myr 1

19.5 2.0 m Myr 1
14.4 1.0 m Myr 1
23.8 2.0 m Myr 1
23.8 2.0 m Myr 1
17.6 1.0 m Myr 1
16.0 1.0 m Myr 1
7.2 1.0 m Myr 1
16.0 1.0 m Myr 1
16.6 2.0 m Myr 1
0.018
0.020
0.014
0.024
0.024
0.018
0.016
0.007
0.016
0.017
(Continued)
Table A1, part 2
(Continued)
CWR
(mm yr 1)
Total denudation
Total
denudation rate
(mm yr 1)
Reference
ID
Soil
thickness
(cm)
Riebe et al.
(2003)c
RIS1
50150
47 11 t km 2 yr 1
0.039
79 18 t km 2 yr 1
0.066
RIS2
RIS3
Average soil
samples
(above)
Rio Icacos,
Puerto Rico
RI-1
RI-4
McNaab
Track, New
Zealand
MT-3
MT-4
MT-5
Chiapas
Highlands,
Mexico
50150
50150
59 14 t km 2 yr 1
59 13 t km 2 yr 1
56 13 t km 2 yr 1
0.049
0.049
0.047
101 23 t km 2 yr 1

97 2 t km 2 yr 1
90 21 t km 2 yr 1
0.084
0.081
0.075
51 10 t km 2 yr 1
59 9 t km 2 yr 1
0.043
0.049
87 15 t km 2 yr 1
97 14 t km 2 yr 1
0.073
0.081
88 17 t km 2 yr 1
115 25 t km 2 yr 1
58 10 t km 2 yr 1
0.073
0.096
0.048
195 31 t km 2 yr 1

235 45 t km 2 yr 1
131 20 t km 2 yr 1
0.163
0.196
0.109
Riebe et al.
(2004a)
SS
Jalisco
Highlands,
Mexico
ST-1
ST-3
ST-4
ST-5
Panola Mtn.,
GA, USA
PM
Jalisco
Lowlands,
Mexico
RT-1
RT-2
Santa Rosa
Mtn., NY,
USA
SR-1
SR-3
SR-4
SR-6
SR-7
SR-10
34 6 t km 2 yr 1
0.028
122 14 t km 2 yr 1
0.102
166 42 t km 2 yr 1

84 14 t km 2 yr 1
158 69 t km 2 yr 1
173 44 t km 2 yr 1
0.138
0.070
0.132
0.144
556
212
622
549
71 t km 2 yr 1
22 t km 2 yr 1
72 t km 2 yr 1
59 t km 2 yr 1
0.463
0.177
0.518
0.458
10 2 t km 2 yr 1
0.008
23 3 t km 2 yr 1
0.019
112 25 t km 2 yr 1

104 22 t km 2 yr 1
0.093
0.087
462 50 t km 2 yr 1

399 46 t km 2 yr 1
0.385
0.333
7 3 t km 2 yr 1
0 3 t km 2 yr 1
15 4 t km 2 yr 1
16 4 t km 2 yr 1
24 7 t km 2 yr 1
2 6 t km 2 yr 1
0.006
0.000
0.013
0.013
0.020
0.002
106
132
144
104
117
117
11 t km 2 yr 1
14 t km 2 yr 1
15 t km 2 yr 1
11 t km 2 yr 1
12 t km 2 yr 1
12 t km 2 yr 1
0.088
0.110
0.120
0.087
0.098
0.098

(Continued)
Table A1, part 2 (Continued)
Reference
ID
Sonora Desert,
Mexico
CE-3
JC-1
Fall River,
Sierra
Nevada,
USA
FR-2
FR-5
FR-6
FR-8
Antelope Lake,
Sierra
Nevada,
USA
AL-4
AL-5
AL-9
AL-10
Adams Peak,
Sierra
Nevada,
USA
AP-3
AP-4
AP-5
Soil
thickness
(cm)
CWR
(mm yr 1)
Total denudation
Total
denudation rate
(mm yr 1)
31 11 t km 2 yr 1
34 13 t km 2 yr 1
0.026
0.028
194 32 t km 2 yr 1

191 20 t km 2 yr 1
0.162
0.159
99 30 t km 2 yr 1
75 16 t km 2 yr 1
19 7 t km 2 yr 1
7 1 t km 2 yr 1
0.083
0.063
0.016
0.006
485 92 t km 2 yr 1

384 54 t km 2 yr 1
104 25 t km 2 yr 1
40 4 t km 2 yr 1
0.404
0.320
0.087
0.033
4 t km 2 yr 1
20 t km 2 yr 1
13 t km 2 yr 1
3 t km 2 yr 1
0.013
0.016
0.022
0.010
72 8 t km 2 yr 1
86 11 t km 2 yr 1
119 12 t km 2 yr 1
91 8 t km 2 yr 1
0.060
0.072
0.099
0.076
0.020
0.005
0.017
140 13 t km 2 yr 1

100 9 t km 2 yr 1
162 15 t km 2 yr 1
0.117
0.083
0.135
15
19
26
12

24 3 t km 2 yr 1
6 4 t km 2 yr 1
20 9 t km 2 yr 1
Riebe et al.
(2004b)d
AP-11
AP-13
Fort Sage,
Sierra
Nevada,
USA
A1
A2(s)
A3(s)
A4(s)
Sunday Peak,
Sierra
Nevada,
USA
SP-1
SP-3
SP-8
Nichols Peak,
Sierra
Nevada,
USA
NP-1
SR-10 (top of
transect)
SR-3
SR-1
SR-4
SR-6
SR-7 (base of
transect)
14 6 t km 2 yr 1
18 6 t km 2 yr 1
0.012
0.015
93 12 t km 2 yr 1
124 12 t km 2 yr 1
0.078
0.103
5 4 t km 2 yr 1
13 4 t km 2 yr 1
32 12 t km 2 yr 1
111 43 t km 2 yr 1
0.004
0.011
0.027
0.093
83 7 t km 2 yr 1
63 17 t km 2 yr 1
173 43 t km 2 yr 1
755 263 t km 2 yr 1
0.069
0.053
0.144
0.629
19 5 t km 2 yr 1
3 7 t km 2 yr 1
11 5 t km 2 yr 1
0.016
0.003
0.009
129 12 t km 2 yr 1

93 11 t km 2 yr 1
86 12 t km 2 yr 1
0.108
0.078
0.072
0.000
16 7 t km 2 yr 1
2 6 t km 2 yr 1
0.013
0.0
127 12
117 12 t km 2 yr 1
0.106
0.098
0 3 t km 2 yr 1
7 3 t km 2 yr 1
15 5 t km 2 yr 1
16 5 t km 2 yr 1
24 7 t km 2 yr 1
0.0
0.006
0.013
0.013
0.020
132
106
144
104
117
14 t km 2 yr 1
11 t km 2 yr 1
15 t km 2 yr 1
11 t km 2 yr 1
12 t km 2 yr 1
0.110
0.088
0.120
0.087
0.098

(Continued)
Table A1, part 2
(Continued)
Reference
ID
Soil
depth
(cm)
Dixon et al.
(2009)e
LD-0
75.4
75.4
66.2
141.5
LD-1
LD-2
LD-3
LD-4
LD-5
LD-6
Mean
WB-0
WB-1
WB-2
WB-3
WB-4
WB-5
WB-6
WB-7
WB-8
Mean
BM-2006-1
BM-2006-2
BM-2006-3
BM-2006-4
Mean
25
27
40
45
53
75
0
24.4
34.9
38.5
22.7
5.5
18.0 6.1
13.0
10.9
0.0
31.3
5.1
21.2
11.4
28.3
5.7
14.1 3.6
0
25.2
17.4
31.3
24.4 7.9
104.8
34.4
3.8
27.5
19.7
53.4
45.6 13.2
39.5
40.7
43.1
0
25.4
15.9
6.4
35.1
18.0
24.9 5.3
87.0
66.5
98.1
53.9
69.0 10.6
104.8
58.7
38.6
66.1
42.4
58.9
63.6 8.4
52.5
51.7
43.1
31.3
30.6
37.1
17.8
63.4
23.7
39.0 5.0
87.0
91.7
115.5
85.1
93.4 5.6
126.4
80.5
34.3
54.8
37.1
45.7
63.6 12.1
57.5
54.2
34.2
34.8
35.1
27.5
23.6
43.2
34.3
38.3 3.8
89.5
103.3
131.1
86.2
104.4 8.1
231.2
139.3
73.0
120.8
79.5
104.6
127.1 20.1
110.0
105.8
77.4
66.1
65.6
64.6
41.4
106.5
57.9
77.3 8.2
176.5
195.1
246.6
171.3
197.8 13.3
53
64
70
110
75
75
60
90
80
107
29
102
127
CWRsoil
(t km 2 yr 1)
CWRsaprolite
(t km 2 yr 1)
CWRtotal
(t km 2 yr 1)
Erosion rate
(t km 2 yr 1)
Total
denudation rate
(t km 2 yr 1)
Dixon et al.
(2009)e
PC-1
PC-2
PC-2006-1
PC-2006-2
PC-2006-3
PC-2006-4
Mean
KR-1
KR-2
KR-2006-1
KR-2006-3
Mean
LD-0
80
110
107
91
61
71
4.2
13.2
0
0
0
21.2
6.4 4.0
N.A.
N.A.
6.5
5.4
6.0 0.5
0
158.0
67.6
124.1
131.3
172.6
76.1
121.6 19.0
4.8
0
17.7
16.7
9.8 4.4
0.034
162.2
80.8
124.1
131.3
172.6
97.3
128.1 15.9
N.A
N.A.
24.3
22.1
23.2 1.1
0.034
131.8
59.4
82.0
92.3
122.7
63.2
91.9 13.4
N.A.
N.A.
52.1
41.2
46.7 5.4
0.030
294.0
140.2
206.1
223.6
295.4
160.4
220.0 29.2
36.0
77.4
76.4
63.3
63.3 9.6
0.064
LD-1
LD-2
LD-3
LD-4
LD-5
LD-6
25
27
40
45
53
75
0
0.020
0.029
0.032
0.019
0.005
0.048
0.016
0.002
0.013
0.009
0.024
0.048
0.036
0.031
0.045
0.028
0.029
0.057
0.037
0.016
0.025
0.017
0.021
0.105
0.073
0.046
0.069
0.045
0.050
33
25
64
64
(Continued)
Table A1, part 2
Reference
a
b
c
d
e
(Continued)
ID
Soil
depth
(cm)
CWRsoil
(t km 2 yr 1)
CWRsaprolite
(t km 2 yr 1)
CWRtotal
(t km 2 yr 1)
Erosion rate
(t km 2 yr 1)
Total
denudation rate
(t km 2 yr 1)
WB-0
WB-1
WB-2
WB-3
WB-4
WB-5
WB-6
WB-7
WB-8
BM-2006-1
BM-2006-2
BM-2006-3
BM-2006-4
PC-1
PC-2
PC-2006-1
PC-2006-2
PC-2006-3
PC-2006-4
KR-1
KR-2
KR-2006-1
KR-2006-3
53
64
70
110
75
75
60
90
80
107
29
102
127
80
110
107
91
61
71
33
25
64
64
0.011
0.009
N.A.
0.026
0.004
0.018
0.010
0.024
0.005
0
0.021
0.015
0.026
0.004
0.011
0
0
0
0.018
N.A.
N.A.
0.005
0.005
0.018
0.019
0.020
0
0.012
0.007
0.003
0.016
0.008
0.040
0.030
0.045
0.025
0.072
0.031
0.056
0.060
0.078
0.035
0.002
N.A.
0.008
0.008
0.029
0.028
0.020
0.026
0.016
0.025
0.012
0.040
0.013
0.040
0.051
0.059
0.051
0.075
0.042
0.056
0.060
0.078
0.052
N.A.
N.A.
0.013
0.012
0.026
0.025
0.016
0.016
0.016
0.013
0.011
0.020
0.016
0.041
0.047
0.060
0.039
0.060
0.027
0.037
0.042
0.056
0.029
N.A.
N.A.
0.024
0.019
0.055
0.052
0.035
0.042
0.032
0.037
0.023
0.059
0.029
0.080
0.098
0.119
0.090
0.135
0.069
0.094
0.102
0.134
0.081
N.A.
N.A.
0.037
0.031
Total denudation rates were taken from Heimsath et al. (2005).

Total denudation rates were taken from Heimsath et al. (2006).
Riebe et al. (2003) used amalgamated soil samples to determine chemical weathering rates and total denudation rates.
Riebe et al. (2004b) used amalgamated soil samples and parent material (49 soil samples, 28 samples of parent material) to determine chemical weathering rates
and total denudation rates.
Erosion rates equal rates of soil production, here total denudation rates are calculated as the sum of total chemical weathering and erosion rates.
Table A2
Characteristics of study sites in, and outside, Australia
Reference
Study site
Elevation
Climate
Geology
Study sites in Australia

Green et al. (2006)
Bega Valley, NSW
200 m
Warm temperate
Rainfall 910 mm yr 1
Warm temperate
Granite and granodiorite of

the Bega Batholith
Ordovician metasediments
and Devonian granites
Mixed-conifer to
Subalpine
Rainfall 330
1200 mm yr 1
Mediterranean
Tropical
Mean annual
temperatures: 225 C
Mean annual rainfall: 220
4200 mm yr 1
Mean annual temperatures
and rainfall from 3.6 C,
650 mm yr 1 (base) to
0.4 C, 850 mm yr 1
(top)
Granitic rocks
Burke et al. (2009)
Snug (lowland), Brown

Mountain (escarpment
crest), NSW
Study sites outside Australia
Dixon et al. (2009)
Sierra Nevada Mountains,
California, USA
Burke et al. (2007)

Riebe et al. (2003)
Point Reyes, California,

USA
Rio Icacos, Puerto Rico
Riebe et al.
(2004a)
42 different study sites
Riebe et al.
(2004b)
Santa Rosa Mountains,

Nevada, USA, Transect
200 m
1000 m
11862991 m
150 m
700 m
20902750 m
Granitic rocks (quartz

diorite, granodiorite)
Quartz diorite
Granite
Granodiorite
Table A3
Soil mixing rates
Reference
Location
Soil-mixing rates
(published rates)
Source: modified after Lobry de Bruyn and Conacher, 1990

Williams (1968)
Northern
Australia
Lee and Wood
Northern
470 g m 2 yr 1
(1971a)
Australia
Holt et al. (1980) Northern
Australia
Spain and
Northern
McIvor (1988)
Australia
Nye (1955)
West Africa
1.25 t ha 1 yr 1
Maldague (1964) Africa
Lepage (1972,
Africa
1973)
Lepage (1974)
Africa
0.670.90 t ha 1 yr 1
Nel and Malan
Africa
0.35 t ha 1 yr 1
(1974)
Pomeroy (1976)
Africa
Wood and Sands Africa
0.3 t ha 1 yr 1
(1978)
Aloni et al.
Africa
4.0 t ha 1 yr 1
(1983)
Akamigbo (1984) Africa
Bagine (1984)
Africa
1.06 t ha 1 yr 1
Lepage (1984)
Africa
Soil-mixing
rate (mm yr 1)
Soil mass moved to surface

(mm yr 1)
Moving agent
Termites
0.39
5.4
Termites
0.083
Termites
0.9172.17
Termites
1.47
2
Termites
termites
Termites
0.104
0.0560.075
0.029
Termites
Termites
0.025
Termites
Termites
0.333
Termites
0.308
0.088
Termites
Termites
Termites
Gupta et al.
(1981)
Salick et al.
(1983)
Aloni and Soyer
(1987)
Nutting et al.
(1987)
Greenslade
(1974)
Humphreys
(1981)
Briese (1982)
Cowan et al.
(1985)
Talbot (1953)
Lyford (1963)
Salem and Hole
(1968)
Rogers (1972)
Wiken et al.
(1976)
Levan and Stone
(1983)
India
15.9 g m 2 day 1
4.8
Termites
South America
0.02 and 0.78

t ha 1 yr 1
3 t ha 1 yr 1
0.0020.065
Termites
0.07 and 0.56

t ha 1 yr 1
400 cm3 ha 1 yr 1
0.006 and
0.047
8.41 t ha 1 yr 1
0.701
Ants
0.350.42 t ha 1 yr 1
0.029 and
0.035
< 0.004
Ants
Africa
North America
South Australia
Eastern
Australia
South-eastern
Australia
Eastern
Australia
Michigan
America
America
Wisconsin
America
America
America
Wisconsin
America
< 0.05 t ha 1 yr 1
0.25
0.25
Termites
Termites
Ants
0.6 t ha 1 yr 1

11.36 t ha 1 yr 1
0.05
0.947
11.36 t ha 1 yr 1
0.947
Ants
0.07
Ants
20
Ants
Ants
0.00020.0006
Ants
Ants
0.133
Ants
(Continued)
Table A3
(Continued)
Reference
Location
England
Waloff and
Blackith
(1962)
Sudd (1969)
England
Argentina
Bucher and
Zuccardi
(1967)
N/A
Lobry de Bruyn
and Conacher
(1990)
Source: listed publications
Carson and
N/A
Kirkby (1972)
Carson and
N/A
Kirkby (1972)
Soil-mixing rates
(published rates)
8.24 t ha
11 t ha
1
1
yr
yr
1
1
3 t ha 1 yr 1
Soil-mixing
rate (mm yr 1)

(mm yr 1)
0.687
0.917
Ants
40400 mg nest 1 day 1
0.007
0.250
0.10 cm2 yr 1
N/A
0.25 cm2 yr 1
Carson and
Kirkby (1972)
N/A
0.15 cm2 yr 1
Muller-Lemans
and van Dorp
(1996)
Europe
1.7
Ants
Ants
Soil erosion by
rainsplash
0.003 cm2 yr 1
Carson and
Kirkby (1972)
Moving agent
1.67
Wedging by grass
roots
Rabbits,
burrowing and
refilling
Worms,
distribution of
surface casts
Worms,
burrowing and
refilling
Earthworms
Gabet et al.
(2003)
Amundson
(2004) (cited
from Darwin)
Buchan (2010)
Wilkinson et al.
(2009)
Gabet et al.
(2003)
Wilkinson et al.
(2009)
Gabet et al.
(2003)
Paton et al.
(1995)
Wilkinson and
Humphreys
(2005)
Wilkinson and
Humphreys
(2005)
Wilkinson et al.
(2009)
5.4 10 40.01

m yr 1
0.02 cm 2 yr 1
0.5410
N/A
90 t ha 1 yr 1
7.5
Various
10100 t ha 1 yr 1
0.8338.33
Earthworms
(digestion)
Earthworms
N/A
4.5 10 6
1.8 10 3 m yr 1
510 t ha 1 yr 1
0.0051.8
Ants
0.4170.833
Ants
1.3 10 5
4.1 10 4 m yr 1
15 kg m 2 yr 1
0.0130.41
Termites
0.8334.167
Ants, termites,
earthworms
Earthworms and
ants
N/A
Various
N/A
N/A
Earthworms
Earthworms
South-eastern
Australia
10.6
South-eastern
Australia
0.4280.714
Ants, termites and

cicades
0.0830.417
Ants, termites,
vertebrates
Various
15 t ha 1 yr 1
(Continued)
Table A3
(Continued)
Reference
Location
Soil-mixing rates
(published rates)
2
Soil-mixing
rate (mm yr 1)
1
Yoo et al. (2005)
N/A
50140 cm yr
Hole (1981)
Heimsath et al.
(2002)
Kaste et al. (2007)
Arctic climates
South-eastern
Australia
South-eastern
Australia
Temperate
maritime
climates
Temperate
continental
climates
Tropical
climates
Polar and
montane
climates
18 t ha 1 yr 1
1040 cm kyr 1
1.5
0.10.4
1020 cm kyr 1
0.10.2
Saunders and
Young (1983)
Saunders and
Young (1983)
Saunders and
Young (1983)
Saunders and
Young (1983)

(mm yr 1)
Moving agent
Pocket gopher
(simulation
model)
Ground squirrel
Soil creep
0.52
Diffusion-like
mixing rates
Soil creep
210
Soil creep
45
Soil creep
1300
10100
Solifluction
Note: Soil-mixing rates recorded in mass per area per year were converted to mm yr 1 in assuming a soil bulk density of 1200 kg m 3 in the conversion processes.
Source: modified after Lobry de Bruyn and Conacher, 1990.
65
REFERENCES
Ahnert, F. (1977). Some comments on the quantitative formulation of geomorphological
processes in a theoretical model. Earth Surf. Process. 2, 191201.
Akamigbo, F. (1984). The role of the nasute termites in the genesis and fertility of Nigerian
soils. Pedologie 36, 179189.
Alexander, E. B. (1985). Rates of soil formation from bedrock or consolidated sediments.
Phys. Geography 6, 2542.
Alexander, E. B. (1988). Rates of soil formation: Implications for soil-loss tolerance. Soil Sci.
145, 3745.
Aloni, K., and Soyer, J. (1987). Cycle des materiaux de construction des termitie`res
dhumivores en savane au Shaba meridional (Zare). Revue de zoologie et de botanique
africaines 101, 329358.
Aloni, K., Malaisse, F., and Kapinga, I. (1983). Roles des termites dans la decomposition du
bois et le transfert de terre dans une foret claire zambezienne In New Trends in Soil
Biology, pp. 600602. Dieu Brichart, Louvain-la-Neuve.
Amundson, R. (2004). Soil formation. In Treatise on Geochemistry (H. D. Holland and
K. K. Turekian, Eds.), Vol. 5, pp. 135. Elsevier Press, Amsterdam.
Amundson, R. (2005). Jenny, Hans. In Encyclopedia of Soils in the Environment
(H. Daniel, Ed.), pp. 293300. Elsevier, Oxford.
Amundson, R., and Jenny, H. (1997). On a state factor model of ecosystems. Bioscience 47,
536543.
Amundson, R., Richter, D. D., Humphreys, G. S., Jobbaggy, E. G., and Gaillardet, J.
(2007). Coupling between biota and earth materials in the critical zone. Elements 3,
327332.
Anderson, J. M. (1988). Section 1. Interactions between invertebrates and microorganisms in
organic-matter decomposition, energy flux and nutrient cycling in ecosystems. Invertebrate-mediated transport processes in soils. Agric. Ecosyst. Environ. 24, 519.
Anderson, S. P., von Blanckenburg, F., and White, A. F. (2007). Physical and chemical
controls on the critical zone. Elements 3, 315319.
Bagine, R. K. N. (1984). Soil translocation by termites of the genus Odontotermes (Holgren)
(Isoptera: Macrotermitinae) in an arid area of northern Kenya. Oecologia 64, 263266.
Blum, W. E., and Ganssen, R. (1972). Bodenbildende Prozesse der Erde, ihre Erscheinungsformen und diagnostischen Merkmale in tabellarischer Darstellung. Die Erde.
Zeitschrift der Gesellschaft fur Erdkunde zu Berlin 103, 720. (in German).
Bockheim, J. G., and Gennadiyev, A. N. (2000). The role of soil-forming processes in the
definition of taxa in Soil Taxonomy and the World Soil Reference Base. Geoderma 95,
5372.
Bockheim, J. G., and Gennadiyev, A. N. (2009). The value of controlled experiments in
studying soil-forming processes: A review. Geoderma 152, 208217.
Brantley, S. L. (2010). Weathering: Rock to regolith. Nat. Geosci. 3, 305306.
Brantley, S. L., White, T. S., White, A. F., Sparks, D., Richter, D., Pregitzer, K., Derry, L.,
Chorover, J., Chadwick, O., April, R., Anderson, S., and Amundson, R. (2006).
Frontiers in Exploration of the Critical Zone: A Workshop Sponsored by the National
Science Foundation (NSF). Newark, DE.
Brantley, S. L., Goldhaber, M. B., and Ragnarsdottir, K. V. (2007). Crossing disciplines and
scales to understand the critical zone. Elements 3, 307314.
Briese, D. T. (1982). The effects of ants on the soil of a semi-arid salt bushland habitat. Insectes
Sociaux 29, 375386.
Buchan, G. (2010). Ode to soil. J. Soil Water Conserv. 65, 48A54A.
66
Bucher, E. H., and Zuccardi, R. B. (1967). Signification de 10s hormigueros de Arta

vouenweideri Forel como alteradores del suelo en la Provinces de Tucuman. Acta
Zoologica Lillona 23, 8396.
Burke, B. C., Heimsath, A. M., and White, A. F. (2007). Coupling chemical weathering
with soil production across soil-mantled landscapes. Earth Surf. Process. Landforms 32,
853873.
Burke, B. C., Heimsath, A. M., Dixon, J. L., Chappell, J., and Yoo, K. (2009). Weathering
the escarpment: Chemical and physical rates and processes, south-eastern Australia. Earth
Surf. Process. Landforms 34, 768785.
Carson, M. A., and Kirkby, M. J. (1972). Hillslope Form and Process. Cambridge University
Press, Cambridge.
Churchman, G. J. (2010). The philosophical status of soil science. Geoderma 157, 214221.
Cleaves, E. T. (1993). Climatic impact on isovolumetric weathering of a coarse-grained
schist in the northern Piedmont Province of the central Atlantic states. Geomorphology 8,
191198.
Colman, S. M. (1981). Rock-weathering rates as functions of time. Quaternary Res. 15,
250264.
Colman, S. M., and Dethier, D. P. (1986). An overview of rates of chemical weathering.
In Rates of Chemical Weathering of Rocks and Minerals (S. M. Colman and
D. P. Dethier, Eds.), pp. 118. Academic Press, Orlando.
Cowan, J. A., Humphreys, G. S., Mitchell, P. B., and Murphy, C. L. (1985). An assessment
of pedoturbation by two species of mound-building ants, Camponotus intrepidus (Kirby)
and Iridomyrmex purpureus (F. Smith). Aust. J. Soil Res. 22, 95107.
Dethier, D. P. (1986). Weathering rates and the chemical flux from catchments in the Pacific
Northwest, USA. In Rates of Chemical Weathering of Rocks and Minerals
(S. M. Colman and D. P. Dethier, Eds.), pp. 503530. Academic Press, Orlando.
Dietrich, W. E., Reiss, R., Hsu, M.-L., and Montgomery, D. R. (1995). A process-based
model for colluvial soil depth and shallow landsliding using digital elevation data. Hydrol.
Process. 9, 383400.
Dixon, J. L., Heimsath, A. M., and Amundson, R. (2009). The critical role of climate and
saprolite weathering in landscape evolution. Earth Surf. Process. Landforms 34, 15071521.
Dokuchaev, V. V. (1886). Key points in the history of land evaluation in the European
Russia, with classification of Russian soils. Materials for Land Evaluation of the Nizhny
Novgorod Governorate. Natural and Historical Part: Report to the Nizhny Novgorod
Governorate Zemstvo, Vol. 1. Nizhny Novgorod Governorate Zemstvo, St. Petersburg
(in Russian).
Dokuchaev, V. V. (1899). Report to the Transcaucasian statistical committee on soil
taxonomy in general and especially for the Transcaucasia. Horizontal and vertical soil
zones. Office Press of the Civilian Affairs Commander-in-Chief in the Caucasus, Tifilis
(in Russian).
Dosseto, A., Turner, S. P., and Chappell, J. (2008). The evolution of weathering profiles
through time: New insights from uranium-series isotopes. Earth Planet. Sci. Lett. 274,
359371.
Edmond, J. M., Palmer, M. R., Measures, C. I., Grant, B., and Stallard, R. F. (1995). The
fluvial geochemistry and denudation rate of the Guayana Shield in Venezuela, Colombia,
and Brazil. Geochim. Cosmochim. Acta 59, 33013325.
Finke, P. A., and Hutson, J. L. (2008). Modelling soil genesis in calcareous loess. Geoderma
145, 462479.
Florinsky, I. V. (2011a). Digital terrain analysis in soil science and geology. Academic
Press, Amsterdam et al. (to be published).
Florinsky, I. V. (2011b). The soil formation equation: Imaginary scientific priority of Hans
Jenny. Pedometron (to be published).
67
Furbish, D. J., and Fagherazzi, S. (2001). Stability of creeping soil and implications for
hillslope evolution. Water Resources Res. 37, 26072618.
Gabet, E. J., Reichman, O. J., and Seabloom, E. W. (2003). The effects of bioturbation on
soil processes and sediment transport. Annu. Rev. Earth Planet. Sci. 31, 249274.
Gilbert, G. K. (1877). Report on the Geology of the Henry Mountains (Utah). United States
Geological Survey, Washington, DC.
Gill, R. A., and Jackson, R. B. (2000). Global patterns of root turnover for terrestrial
ecosystems. New Phytol. 147, 1331.
Graham, R. C., Rossi, A. M., and Hubbert, K. R. (2010). Rock to regolith conversion:
Producing hospitable substrates for terrestrial ecosystems. GSA Today 20, 49.
Granger, D. E., Riebe, C. S., Kirchner, J. W., and Finkel, R. C. (2001). Modulation of
erosion on steep granitic slopes by boulder armoring, as revealed by cosmogenic 26Al and
10
Be. Earth Planet. Sci. Lett. 186, 269281.
Green, E. G., Dietrich, W. E., and Banfield, J. F. (2006). Quantification of chemical
weathering rates across an actively eroding hillslope. Earth Planet. Sci. Lett. 242, 155169.
Greenslade, P. J. N. (1974). Some relations of the meat ant, Iridomyrmex purpureus
(Hymenoptera: Formicidae) with soil in South Australia. Soil Biology and Biochemistry 6,
714.
Gupta, S. R., Rajvanshi, R., and Singh, J. S. (1981). The role of the termite Odontotermes
gurdaspurensis (Isoptera: Termitidae) in plant decomposition in a tropical grassland.
Pedobiologia 22, 254261.
Heimsath, A. M., Dietrich, W. E., Nishiizumi, K., and Finkel, R. C. (1997). The soil
production function and landscape equilibrium. Nature 388, 358361.
Heimsath, A. M., Dietrich, W. E., Nishiizumi, K., and Finkel, R. C. (1999). Cosmogenic
nuclides, topography, and the spatial variation of soil depth. Geomorphology 27, 151172.
Heimsath, A. M., Chappell, J., Dietrich, W. E., Nishiizumi, K., and Finkel, R. C. (2000).
Soil production on a retreating escarpment in southeastern Australia. Geology 28,
787790.
Heimsath, A. M., Chappell, J., Dietrich, W. E., Nishiizumi, K., and Finkel, R. C. (2001a).
Late Quaternary erosion in southeastern Australia: A field example using cosmogenic
nuclides. Quaternary Int. 8385, 169185.
Heimsath, A. M., Dietrich, W. E., Nishiizumi, K., and Finkel, R. C. (2001b). Stochastic
processes of soil production and transport: Erosion rates, topographic variation and
cosmogenic nuclides in the Oregon Coast Range. Earth Surf. Process. Landforms 26,
531552.
Heimsath, A. M., Chappell, J., Spooner, N. A., and Questiaux, D. G. (2002). Creeping soil.
Geology 30, 111114.
Heimsath, A. M., Furbish, D. J., and Dietrich, W. E. (2005). The illusion of diffusion: Field
evidence for depth-dependent sediment transport. Geology 33, 949952.
Heimsath, A. M., Chappell, J., Finkel, R. C., Fifield, K., and Alimanovic, A. (2006).
Escarpment erosion and landscape evolution in southeastern Australia. Geological Society
of America 398, 173190.
Heimsath, A. M., Fink, D., and Hancock, G. R. (2009). The humped soil production
function: Eroding Arnhem Land, Australia. Earth Surf. Process. Landforms 34, 16741684.
Hole, F. D. (1981). Effects of animals on soil. Geoderma 25, 75112.
Holt, J. A., Coventry, R. J., and Sinclair, D. F. (1980). Some aspects of the biology and
pedological significance of mound-building termites in a red and yellow earth landscape
near Charters Towers, North Queensland. Aust. J. Soil Res. 18, 97109.
Hoosbeek, M. R., and Bryant, R. B. (1992). Towards the quantitative modeling of
pedogenesisA review. Geoderma 55, 183210.
Huggett, R. J. (1975). Soil landscape systems: A model of soil genesis. Geoderma 13, 122.
68
Huggett, R. J. (1998). Soil chronosequences, soil development, and soil evolution: A critical
review. CATENA 32, 155172.
Humphreys, G. S. (1981). The rate of ant mounding and earthworm casting near Sydney,
NSW. Search 12, 129131.
Humphreys, G. S., and Wilkinson, M. T. (2007). The soil production function: A brief
history and its rediscovery. Geoderma 139, 7378.
Jenny, H. (1941). Factors of Soil Formation. A System of Quantitative Pedology. McGrawHill Book Company, New York, London.
Johnson, D. L. (1990). Biomantle evolution and the redistribution of earth materials and
artifacts. Soil Sci. 149, 84102.
Johnson, D. L., and Watson-Stegner, D. (1987). Evolution model of pedogenesis. Soil Sci.
143, 349366.
Kaste, J. M., Heimsath, A. M., and Bostick, B. C. (2007). Short-term soil mixing quantified
with fallout radionuclides. Geology 35, 243246.
Kirkby, M. J. (1977). Soil development models as a component of slope models. Earth Surf.
Process. Landforms 2, 203230.
Kirkby, M. J. (1985). A basis for soil profile modelling in a geomorphic context. J. Soil Sci.
36, 97121.
Lasaga, A. C., Soler, J. M., Ganor, J., Burch, T. E., and Nagy, K. L. (1994).
Chemical weathering rate laws and global geochemical cycles. Geochim. Cosmochim.
Acta 58, 23612386.
Lee, K. E., and Wood, T. G. (1971a). Physical and chemical effects on soils of some
Australian termites and their pedological significance. Pedobiologia 11, 376409.
Lepage, M. (1972). Recherches ecologiques sur une savane sahelienne du Ferlo Septentrional, Senegal: donnees preliminaires sur Iecologie des termites. Terre et la Vie 26,
384409.
Lepage, M. (1973). Recherches, ecologiques sur une savane shaelienne du Senegal Septentrional. Termites: repartition, biomasse et recolte de nourriture. Annales des lUniversite d
Abidjan 6, 136145.
Lepage, M. (1974). Les termites dune savane sahelienne (Ferlo Septentrional, Senegal):
peuplements, consommation, role dans Iecosysteme. University of Dijon, Dijon.
Lepage, M. (1984). Distribution, density and evolution of Macrotermes bellicosus nests
(Isoptera: Macrotermitinae) in the north-east of Ivory Coast. J. Anim. Ecol. 53, 107118.
Levan, M. A., and Stone, E. L. (1983). Soil modification by colonies of Black Meadow ants
in a New York old field. Soil Science Society of America Proceedings 47, 11921195.
Lobry de Bruyn, L. A., and Conacher, A. J. (1990). The role of termites and ants in soil
modification: A review. Aust. J. Soil Res. 28, 5593.
Lyford, W. H. (1963). Importance of ants to brown podzolic soil genesis in New England.
Harvard Forest Bulletins and Papers 7.
Maldague, M. E. (1964). Importance des populations de termites dans les sols equatoriaux.
The Transactions of the 8th International Congress of Soil Science. Bucharest 3, 743751.
McBratney, A. B., Mendonca Santos, M. L., and Minasny, B. (2003). On digital soil
mapping. Geoderma 117, 352.
McKean, J. A., Dietrich, W. E., Finkel, R. C., Southon, J. R., and Caffee, M. W. (1993).
Quantification of soil production and downslope creep rates from cosmogenic 10Be
accumulations on a hillslope profile. Geology 21, 343346.
Minasny, B., and McBratney, A. B. (1999). A rudimentary mechanistic model for soil
production and landscape development. Geoderma 90, 321.
Minasny, B., and McBratney, A. B. (2001). A rudimentary mechanistic model for soil
formation and landscape development: II. A two-dimensional model incorporating
chemical weathering. Geoderma 103, 161179.
69
Minasny, B., and McBratney, A. B. (2006). Mechanistic soil-landscape modelling as an

approach to developing pedogenetic classifications. Geoderma 133, 138149.
Minasny, B., McBratney, A. B., and Salvador-Blanes, S. (2008). Quantitative models for
pedogenesisA review. Geoderma 144, 140157.
Muller-Lemans, H., and van Dorp, F. (1996). Bioturbation as a mechanism for radionuclide
transport in soil: Relevance of earthworms. J. Environ. Radioact. 31, 720.
NAS (2010). Landscapes on the edge. New horizons for research on Earths surface. The
National Academy Press, Washington, DC.
Nel, J. J. C., and Malan, E. M. (1974). The distribution of the mounds of Trinervitermes
trinervoides in the central Orange Free State. J. Entomol. Soc. South Af. 37, 251256.
Nye, P. H. (1955). Some soil forming processes in the humid tropics. IV. The action of soil
fauna. J. Soil. Sci. 6, 7383.
Nutting, W. L., Haverty, M. I., and La Fage, J. P. (1987). Physical and chemical alteration of
soil by two subterranean termite species in Sonoran Desert grassland. J. Arid Environ. 12,
233239.
Owens, L. B., and Watson, J. P. (1979). Landscape reduction by weathering in small
Rhodesian watersheds. Geology 7, 281284.
Paces, T. (1986). Rates of weathering and erosion derived from mass balance in small
drainage basins. In Rates of Chemical Weathering of Rocks and Minerals
Paton, T. R., Humphreys, G. S., and Mitchell, P. B. (1995). Soils: A new global view.
University College London Press, London.
Pavich, M. J. (1986). Processes and rates of saprolite production and erosion on a foliated
granitic rock of the Virginia Piedmont. In Rates of Chemical weathering of Rocks and
Minerals (S. M. Colman and D. P. Dethier, Eds.), pp. 552590. Academic Press,
Orlando.
Pavich, M. J. (1989). Regolith residence time and the concept of surface age of the Piedmont
Peneplain. Geomorphology 2, 181196.
Pelletier, J. D., and Rasmussen, C. (2009). Geomorphically based predictive mapping of soil
thickness in upland watersheds. Water Resources Res. 45, doi:10.1029/2008WR007319.
Pomeroy, D. E. (1976). Some effects of mound-building termites on soils in Uganda. J. Soil
Sci. 27, 377394.
Regan, E. J. (1977). The Natural Energy Basis for Soils and Urban Growth in Florida.
University of Florida, Florida.
Riebe, C. S., Kirchner, J. W., and Finkel, R. C. (2003). Long-term rates of chemical
weathering and physical erosion from cosmogenic nuclides and geochemical mass balance. Geochim. Cosmochim. Acta 67, 44114427.
Riebe, C. S., Kirchner, J. W., and Finkel, R. C. (2004a). Erosional and climatic effects on
long-term chemical weathering rates in granitic landscapes spanning diverse climate
regimes. Earth Planet. Sci. Lett. 224, 547562.
Riebe, C. S., Kirchner, J. W., and Finkel, R. C. (2004b). Sharp decrease in long-term
chemical weathering rates along an altitudinal transect. Earth Planet. Sci. Lett. 218,
421434.
Rogers, L. E. (1972). The ecological effects of the western harvester ant (Pogonomyrmex
occidentalis) in the shortgrass plain ecosystem. USA/BP Grassland Biome, Technical
Report No206.
Runge, E. C. A. (1973). Soil development sequences and energy models. Soil Sci. 115,
183193.
Ruxton, B. P. (1968). Measures of the degree of chemical weathering of rocks. J. Geol. 76,
518527.
Salem, M., and Hole, F. D. (1968). Ant (Formica exsectoides) pedoturbation in a forest soil.
Soil Science Society of America Proceedings 32, 563567.
70
Salick, J., Herrera, R., and Jordan, C. F. (1983). Termitaria: nutrient patchiness in nutrient
deficient rain forests. Biotropica 15, 17.
Salvador-Blanes, S., Minasny, B., and McBratney, A. B. (2007). Modelling long-term in situ
soil profile evolution: Application to the genesis of soil profiles containing stone layers.
Eur. J. Soil Sci. 58, 15351548.
Samedov, P. A., and Nadirov, F. T. (1990). Effect of earthworms and woodlice on the
physicochemical and surface properties of soils. Soil Biol. 22, 4855.
Saunders, I., and Young, A. (1983). Rates of surface processes on slopes, slope retreat and
denudation. Earth Surf. Process. Landforms 8, 473501.
Schaetzel, R., and Anderson, S. (2005). Soils. Genesis and Geomorphology. Cambridge
University Press, New York.
Schaetzel, R. J., Barrett, L. R., and Winkler, J. A. (1994). Choosing models for soil
chronofunctions and fitting them to data. Eur. J. Soil Sci. 45, 219232.
Shaw, C. F. (1930). Potent factors in soil formation. Ecology 11(XI), 239245.
Simonson, R. W. (1959). Modern concepts of soil genesis. Outline of a generalized theory of
soil genesis. Soil Science Society Proceedings 152156.
Sommer, M., Gerke, H. H., and Deumlich, D. (2008). Modelling soil landscape genesisA
"time split" approach for hummocky agricultural landscapes. Geoderma 145, 480493.
Spain, A. V., and McIvor, J. G. (1988). The nature of herbaceous vegetation associated with
termitaria in north-eastern Australia. J.Ecol. 76, 181191.
Sudd, J. H. (1969). The excavation of soil by ants. Zeitschrift fur Tierpsychologie 26, 257276.
Talbot, M. (1953). Ants of an old field community on the Edwin S. George Reserve,
Livingston County, Michigan. Contributions from the Laboratory of Vertebrate Biology.
University of Michigan 69, 19.
Taylor, G., and Eggleton, R. A. (2001). Regolith Geology and Geomorphology. John
Wiley & Sons Ltd., Chichester.
Tonneijck, F. H., and Jongmans, A. G. (2008). The influence of bioturbation on the vertical
distribution of soil organic matter in volcanic ash soils: A case study in northern Ecuador.
Eur. J. Soil Sci. 59, 10631075.
Velbel, M. A. (1986). The mathematical basis for determining rates of geochemical and
geomorphic processes in small forested watersheds by mass balance: Examples and
implications. In Rates of Chemical Weathering of Rocks and Minerals
Volobuyev, V. R. (1974). Main concepts of ecology. Geoderma 12, 2733.
Volobuyev, V. R. (1984). Two key solutions of the energetics of soil formation. Soviet Soil
Sci. 16, 18.
Volobuyev, V. R., and Ponomarev, D. G. (1977). Some thermodynamic characteristics of
the mineral associations of soils. Soviet Soil Sci.Genesis Geography of Soils 111.
Volobuyev, V. R., Ponomarev, D. G., and Mikailov, F. D. (1980). Relation between the
thermodynamic functions of soils, their mineral composition and infiltration capacity.
Soviet Soil Sci.Soil Phys. 210212.
Wakatsuki, T., and Rasyidin, A. (1992). Rates of weathering and soil formation. Geoderma
52, 251263.
Waloff, N., and Blackith, R. E. (1962). The growth and distribution of the mounds of
Lasius flavus (F.) (Hymenoptera: Formicidae) in Silwood Park, Berkshire. J.Anim.Ecol.
31, 421437.
White, A. F., and Blum, A. E. (1995). Effects of climate on chemical weathering in watersheds. Geochim. Cosmochim. Acta 59, 17291747.
White, A. F., and Brantley, S. L. (2003). The effect of time on the weathering of silicate
minerals: Why do weathering rates differ in the laboratory and field? Chem. Geol. 202,
479506.
71
Wiken, E. B., Boersma, L. M., Lavukulich, L. M., and Farstead, L. (1976). Biosynthetic
alteration in a British Columbia soil by ants (Formica fusca, Linne). Soil Science Society of
America Proceedings 40, 422426.
Wilkinson, M. T., and Humphreys, G. S. (2005). Exploring pedogenesis via nuclide-based
soil production rates and OSL-based bioturbation rates. Aust. J. Soil Res. 43, 767779.
Wilkinson, M. T., Chappell, J., Humphreys, G. S., Fifield, K., Smith, B., and Hesse, P.
(2005). Soil production in heath and forest, Blue Mountains, Australia: Influence of
lithology and palaeoclimate. Earth Surf. Process. Landforms 30, 923934.
Wilkinson, M. T., Richards, P. J., and Humphreys, G. S. (2009). Breaking ground:
Pedological, geological, and ecological implications of soil bioturbation. Earth Sci. Rev.
97, 257.
Willgoose, G. (2005). Mathematical modeling of whole landscape evolution. Annu. Rev.
Earth Planet. Sci. 33, 443459.
Williams, M. A. J. (1968). Termites and soil development near Brocks creek, Northern
Territory. Aust. J .Earth Sci. 31, 153154.
Wood, T. G., and Sands, W. A. (1978). The role of termites in ecosystems. In Production
Ecology of Ants and Termites (M. V. Brian, Ed.), pp. 245292. Cambridge University
Press, London.
Yaalon, D. H. (1975). Conceptual models in pedogenesis: Can soil-forming functions be
solved? Geoderma 14, 189205.
Yoo, K., Amundson, R., Heimsath, A. M., and Dietrich, W. E. (2005). Process-based model
linking pocket gopher (Thomomys bottae) activity to sediment transport and soil thickness. Geology 33, 917920.
Yoo, K., Amundson, R., Heimsath, A. M., Dietrich, W. E., and Brimhall, G. H. (2007).
Integration of geochemical mass balance with sediment transport to calculate rates of soil
chemical weathering and transport on hillslopes. J. Geophys. Res. 112.
Zakharov, S. A. (1927). A course of soil science. Gosizdat, Moscow (in Russian).

80 Pedogenesis Review

Uploaded by

Copyright:

Available Formats

80 Pedogenesis Review

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

80 Pedogenesis Review

Uploaded by

Copyright:

Available Formats

See

discussions, stats, and author profiles for this publication at:

Available from: Budiman Minasny

Faculty of Agriculture, Food

Advances in Agronomy, Volume 113

2011 Elsevier Inc.

Uta Stockmann et al.

Quantifying Processes of Pedogenesis

Water flow paths

Uta Stockmann et al.

BOX 1 The critical zone

Quantifying Processes of Pedogenesis

However, there has been a shift of interest toward mechanistic modeling.

2. Conceptual Models of Soil Formation

Uta Stockmann et al.

BOX 2 Background: The equation of soil-forming factors

Quantifying Processes of Pedogenesis

the nature of the parent rock

Here, Dokuchaevs hypothesis is only published as a sentence. However,

p f M:G:P:; P::OpG:; :; Bop: cTp:; P-:

of time (T). In addition to these soil-forming factors, he also included the

Although presented as a mathematical equation, Shaws soil-formation

Uta Stockmann et al.

The most well-known conceptual model of soil formation is Jennys

where cl is the climate, o are the organisms, r is the topography, p is the

Quantifying Processes of Pedogenesis

Figure 2 Soil profile evolution as a function of additions, removals, translocations, and

and transformations (i.e., primary minerals into secondary minerals) as

The soil-forming processes approach by Simonson (1959) can be seen as

Uta Stockmann et al.

where P stands for progressive pedogenic conditions, including processes

where S is the soil, o is the organic matter production (renewing factor), w is

Quantifying Processes of Pedogenesis

are obviously underestimated, but nevertheless this model can be seen as a

where Q is the expenditure of energy on soil formation, R is the energy of

Uta Stockmann et al.

16,000 14,000 12,000 10,000 8000

where Q is the (annual) expense of energy on soil-forming processes, R is

Quantifying Processes of Pedogenesis

3. Soil Weathering and Production

3.1. Production of soil from parent materials

Uta Stockmann et al.

climate and vegetation cover. New views in evolutionary pedogenesis tried

Further, statistical models can be applied to express chronosequence data

Figure 4 Types of mathematical functions commonly used in chronofunctions

Quantifying Processes of Pedogenesis

Demonstrated in Fig. 4 chronofunctions might be modeled using:

simple linear behavior Y a bt

Relative production rate

Uta Stockmann et al.

(1) The exponential soil-production model

where P0 ([L T 1], mm kyr 1) is the rate of weathering of bedrock at

Quantifying Processes of Pedogenesis

major factor in this phenomenon by Carson and Kirkby (1972). Water is

where P0 ([L T 1], mm kyr 1) represents the rate of weathering of

An empirical parameterization of the humped model from field data is

Uta Stockmann et al.

3.2. Chemical weathering of bedrock to soil

Quantifying Processes of Pedogenesis

P0 expk1 h expk2 h Pa Quantitative-mechanistic Dietrich et al. (1995), Furbish