Basso et al. (2011). Procedures for Initializing Soil Organic Carbon Pools in the DSSAT‐CENTURY Model for Agricultural Systems

SSSAJ: Volume 75: Number 1 • January–February 2011

69

Soil Sci. Soc. Am. J. 75:69–78
Posted online 30 Nov. 2010
doi:10.2136/sssaj2010.0115
Received 11 Mar. 2010.
*Corresponding author (bruno.basso@unibas.it).
© Soil Science Society of America, 5585 Guilford Rd., Madison WI 53711 USA
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by
any means, electronic or mechanical, including photocopying, recording, or any information storage
and retrieval system, without permission in writing from the publisher. Permission for printing and for
reprinting the material contained herein has been obtained by the publisher.

Procedures for Initializing Soil Organic Carbon Pools in

the DSSAT-CENTURY Model for Agricultural Systems

Soil Biology & Biochemistry

S

oil organic C is recognized as a heterogeneous mixture of organic substances,
but for experimental and modeling purposes it is oft en conceptually divided into

discrete C pools for interpretation of decomposition kinetics, chemical and physical
attributes (e.g., C/N ratios or solubility), and stabilization mechanisms (Jenkinson,
1981; Elliott et al., 1996). Th

e SOC pools or fractions have been determined by

many diff erent procedures (Olk and Gregorich, 2006). Chemical-based extractions
were dominant until recent decades, when physically based fractionations have be-
come increasingly popular, and more elaborate analyses for microbial characteristics
have been developed. Comprehensive reviews are available for extractions involving
particulate organic matter and the light fraction (von Lützow et al., 2007; Gregorich
and Janzen, 1996; Wander, 2004), particle size fractions, aggregate classes (Six et al.,
2004), and chemical extractants (Hayes, 1985; Stevenson, 1994). In addition, SOC
pool diff erentiation has also been attempted on the basis of mid-infrared spectra
(Janik et al., 2007). At present, however, there is no single, agreed-on biological,
physical, or chemical fractionation technique that adequately describes the continu-
um of SOC from the standpoint of decomposition and stabilization kinetics.

Process-based SOC models are widely used for simulating, monitoring, and

verifying soil C change. Modeling is a necessary tool for extrapolating the fi ndings
on C changes obtained from individual experiments to larger areas or regions and
varying time scales to assess future changes (Paustian et al., 1997; Falloon et al.,
2002; Senthilkumar et al., 2009; Ogle et al., 2010). To extend models beyond a sin-
gle fi eld plot (e.g., a well characterized experimental fi eld), a number of soil, plant,
climate, and management-related parameters and driving variables are needed that

Bruno Basso*

Dep. of Crop, Forest and Environ. Sci.
Univ. of Basilicata
Viale Ateneo Lucano 10
85100, Potenza, Italy
and
W.K. Kellogg Biological Station
Michigan State Univ.
Hickory Corner, MI
and
Institute for Sustainable Resources
Queensland Univ. of Technology
Brisbane, QLD, Australia

Osvaldo Gargiulo

Dep. of Agricultural & Biological Eng.
Univ. of Florida
Gainesville, FL 32611

Keith Paustian

Natural Resource Ecology Lab.
and
Dep. of Soil and Crop Sciences
Colorado State Univ.
Fort Collins, CO 80523-1499

G. Philip Robertson

W.K. Kellogg Biological Station
Michigan State Univ.
Hickory Corner, MI
and
Dep. of Crop and Soil Sciences
Michigan State Univ.
East Lansing, MI 48824

Cheryl Porter

Dep. of Agricultural & Biological Eng.
Univ. of Florida
Gainesville, FL 32611

Peter R. Grace

WK. Kellogg Biological Station
Michigan State Univ.
Hickory Corner, MI
and
Institute for Sustainable Resources
Queensland Univ. of Technology
Brisbane, QLD, Australia

James W. Jones

Dep. of Agricultural & Biological Eng.
Univ. of Florida
Gainesville, FL 32611

Process-based soil organic C (SOC) models are widely used for simulating, monitoring, and verifying soil C change.
In such models, determining the initial distribution of SOC across multiple pools is oft en not well defi ned, yet
pool initialization can strongly infl uence the subsequent SOC dynamics. We developed a model-based procedure
to initialize SOC fractions that uses site-specifi c soil, climatic, and land use history information, along with mea-
sured initial total SOC, to estimate SOC distribution across pools. Th e procedure consists of creating sets of SOC
fractions for scenarios including diff erent fi eld histories, soil texture, and management practices for medium- and
long-term simulations to provide a broad spectrum of conditions, using data from an experimental site in Georgia.
Tables created using this procedure can help model users select the SOC fractions needed to properly initialize the
model. Th e model is executed for the duration of the prior land use history time period, and the simulated fi nal
total SOC is compared with the soil C measurement at the beginning of the subsequent experimental time period.
If these values are equal, the fi nal SOC fractions from the land use history simulations are used with the measured
soil C to start the simulation of the cropping system being studied. If these values are not equal, then the procedure
is repeated iteratively until the measured value of soil C is adequately predicted. We demonstrated the improved pre-
diction accuracy by using the proposed procedure in simulating soil C for a long-term fi eld experiment in Michigan.

Abbreviations:

CTOT, total soil organic carbon; KBS, Kellogg Biological Station; SOC, soil organic carbon.

Published January, 2011

70

SSSAJ: Volume 75: Number 1 • January–February 2011

span the range of spatial and temporal conditions being consid-
ered (Paustian et al., 1995). Predicted changes in SOC should
also be compared with measured data that represent the spatial
and temporal range of model inferences to assess uncertainty and
bias (Falloon and Smith, 2003; Ogle et al., 2007).

Models can be used to gain an understanding of the process-

es and control involved in nutrient cycles and to generate data on
the size of various pools and the rates at which nutrients are trans-
formed. Most models of SOC dynamics include multiple pools
or fractions of soil C that are governed by fi rst-order decay rates
(Paustian, 1994; McGill, 1996; Falloon and Smith, 2000). Th

ese

models begin by modeling the decay of litter at the soil surface.

Th

e division of SOM into various pools is based on vari-

ous criteria such as stabilization mechanisms, bioavailability,
and kinetic parameters. Aside from litter pools, several models
subdivide SOC into three main pools: active, slow, and resistant
(van Veen and Paul, 1981; van Veen et al., 1984; Parton et al.,
1988a). Th

e active pool is characterized by a rapid turnover rate,

while the larger size pools have a turnover rate that ranges from
a few decades (the slow pool) to thousands of years (the resistant
pool). Th

e active pool is associated with measureable entities

such as microbial biomass, while the slow and resistant pools are
more conceptual (Plante and Parton, 2007).

A particular challenge for regional model applications is to

derive estimates for the initial SOC conditions (i.e., attributes at
the start of the simulation). Th

e importance of initializing dif-

ferent pools of C has not been emphasized enough, especially in
agricultural systems in which crops depend on nutrients mineral-
ized from organic matter and in which land use and management
practices change frequently with time. Th

e availability of reliable

measurements of total SOC may not be suffi

cient to properly

initialize soil C models. Depending on the partitioning of soil C
into the diff erent pools, decomposition of SOC during a growing
season may result in very low or very rapid rates of mineralization
and supply of nutrients. Th

us, for the same initial total SOC val-

ue, the model may simulate vastly diff erent yields under identical
environmental and management conditions depending on how
the SOC is partitioned into diff erent pools. Typical values of the
C fractions in each pool may be provided by model developers,
but caution should be used because such information may prove
to be unreliable for the soil and cropping system being simulated.
In early attempts to initialize the DSSAT-CENTURY model to
simulate soil C dynamics and crop production in Africa, Bostick et
al. (2007) found that neither crop yield nor soil C dynamics were
adequately described. Th

e simulated yields, soil C loss, and N up-

take were typically higher than those measured in crop rotation ex-
periments, such as in an 11-yr study in Burkina Faso (Bado, 2002).
Simulated mineralization of N from soil organic matter provided
relatively high amounts of N, whereas in reality, the crop growth
was severely limited by a lack of N. Such inconsistency may be due
to the fact that the DSSAT-CENTURY model was implemented
with two default SOC fractions that varied only as a function of
prior land use (grassland or cultivated soils) (Gijsman et al., 2002).

A common approach to initializing soil C pools in dynamic

SOC models is based on the use of soil C steady-state conditions,
obtained by simulating a long time period of 100 to 10,000 yr
(Parton et al., 1988b). One of the limitations of this method is
that agricultural systems oft en change during a relatively short
period of time (i.e., every 3–10 yr), which does not allow the soil
C levels to reach steady state. A preliminary sensitivity analysis
of initial SOC pool size and mineralization using the DSSAT-
CENTURY model clearly demonstrated that the initial parti-
tioning of measured soil C into pools using reported “typical”
values was inadequate for production systems with little or no ex-
ternal nutrient inputs. An additional set of analyses showed that
initializing the soil C pool fractions using long-term simulations
(i.e., >1000 yr) aft er achieving steady state was also inadequate
( J.W. Jones, personal communication, 2010).

Th

e literature on this important issue is limited, even though it

is well known and accepted that model initializations can infl uence
the behavior of soil organic C dynamics. Conceptual model pools
in many cases do not correspond to measurable fractions (Smith et
al., 2002) and thus the initial conditions for individual pools can-
not be directly determined from measurements. Th

e inappropriate

initialization of SOC pools can also lead to inaccurate assessment
of interannual variability (Yeluripati et al., 2009). In addition to the
conventional procedure of spin-up simulation for hundreds or thou-
sands of years to fi nd a soil C equilibrium or assume the equilibrium
conditions, recently Yeluripati et al. (2009) assessed the potential of
Bayesian calibration to initialize soil C pools and reduced the uncer-
tainties in the initial C distribution. Th

is methodology overcomes

some of the diffi

culties with the long-term spin-up but it is compu-

tationally intensive. Even though the Bayesian procedure was tested
with some CO

2

measurements—the temporal variability associated

with CO

2

output within and between days—it did not provide a

comprehensive test of the initialization of pools for predicting lon-
ger term SOC dynamics in agricultural systems.

In this study, we hypothesized that initial SOC pools in the

model can be estimated if total SOC and soil texture are mea-
sured and the cropping history is known. Precise initialization
of SOC fractions by simulating the past cropping history at a
site allows a better comparison of measured and estimated SOC
trends. Large relative diff erences in fraction sizes may occur in
response to fi eld management history even if the total SOC is
the same at Time 0 for a given soil. Th

e primary objective of this

study was to develop a model-based procedure to initialize SOC
fractions that would closely match the measured value of SOC
using site-specifi c soil, climatic, and land management condi-
tions. Th

e procedure produces initialization data sets of SOC

fractions for a wide range of scenarios that take into account site
agronomic history, soil texture, and climate to accommodate a
broad spectrum of conditions. We further aimed to demonstrate
the improved accuracy obtained using the proposed procedure
in simulating SOC for an experimental site in Michigan.

14350661, 2011, 1, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.2136/sssaj2010.0115 by Hayley Peter-Contesse - Oregon State University , Wiley Online Library on [19/10/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

SSSAJ: Volume 75: Number 1 • January–February 2011

71

METHODS AND PROCEDURES

Model Description

Th

e DSSAT-CENTURY model was developed by Gijsman et

al. (2002) by incorporating the widely used CENTURY model with
the DSSAT cropping system model (Hoogenboom et al., 2004). Th

e

CENTURY model (Parton et al., 1987, 1988a; Parton and Rasmussen,
1994; Metherell et al., 1993) is an ecosystem biogeochemistry model that
simulates the dynamics of soil organic C and N for a homogeneous land
area. Th

e DSSAT-CENTURY model calculates the initial amount of or-

ganic C (kg ha

−1

) in each pool using initial SOC fractions, bulk density,

and the thickness of soil layers (Gijsman et al., 2002). Th

e model has three

SOC pools (active, intermediate, and passive) in each of several soil layers,
above- and belowground litter pools, and a surface microbial C pool that is
associated with decomposing surface litter. Th

e pools are conceptually de-

fi ned (i.e., they don’t correspond directly to a specifi c measured SOC pool),
in which the decomposition rate constants and other parameters aff ecting
decay and stabilization processes have been estimated from long-term fi eld
experiments and laboratory incubations (Parton et al., 1987, 1988b). Th

e

daily total C change per layer is determined from the amount of C in each
pool, a decomposition factor calculated as a function of soil temperature
and water content, the maximum specifi c decomposition rate for each pool,
and a cultivation factor that aff ects the mineralization process.

Th

e active pool (SOC1) in CENTURY, including DSSAT-

CENTURY (Gijsman et al., 2002), represents soil microbes and microbial
products. It has a turnover time of a few months, depending on the environ-
ment and soil texture. Th

e intermediate pool (SOC2) includes resistant plant

material derived from structural plant tissue and soil-stabilized products de-
rived from the active and surface microbial pools. It has a turnover rate (resi-
dence time) of around 5 to 20 yr depending on soil texture, soil disturbance,
and environmental conditions. Th

e passive pool (SOC3) is resistant to de-

composition and includes physically and chemically stabilized SOC with a
turnover time of 200 yr or longer depending on environmental conditions.
Th

ese SOC fractions vary with time and are aff ected by soil texture and man-

agement, which in turn can determine the long-term trends of organic C.

Model-Based Procedure for Initializing
Soil Organic Carbon Pools

We have developed an approach for estimating SOC pools in soil

where crops have been grown under rotation systems or changing manage-
ment for a short period of time relative to the turnover rate of the passive
soil C pool (i.e., time periods of <500 yr). Th

e procedure can be used to

produce initialization values that approximate relative pool fractions, or it
can be used directly to initialize pool fractions for a particular experimen-
tal fi eld. Th

e rationale for producing look-up tables is that some users may

not be able to simulate the actual history for a fi eld, and the tables provide
useful approximations to those values using minimal information on land
use history. In both cases, the procedure requires a measurement of initial
total organic soil C (CTOT

t

) at time

t

= 0 when the simulation is to start.

Th

e procedure does not attempt to estimate total SOC at time

t

= 0.

Creating Tables of Intermediate Soil Carbon Fractions

To produce tables of SOC fractions, we focused on the intermediate

C pool (SOC2) in that the changes in SOC are mostly due to this pool
when the simulation duration is <50 yr; during that time, changes to the

C in the passive pool (SOC3) are relatively small. Th

e SOC2 fraction is

unknown at time

t

= 0, but if it can be estimated, then with the measure-

ment of total SOC, the other SOC pool fractions (FC

i

, for

i

= 1–3) can

be estimated. If we have an estimate of soil C in the intermediate pool
(SOC2), we can compute the fraction of soil C (%) in that pool (FC

2,0

) as

0

2,0

0

SOC2

FC

CTOT

=

[1]

Furthermore, because the active C pool fraction is small (typically on
the order of 1 to 5% of total C), an approximation of the initial C frac-
tion in this pool is

1,0

FC

0.02

=

[2]

where FC

1,0

is the fraction of C in the active pool at time

t

= 0. Th

is

fraction size is arbitrary; because the active pool is small and has a rapid
turnover rate, it will quickly adjust and come into equilibrium within the
fi rst simulated year. Finally, the fraction of the passive pool at time

t

= 0

(FC

3,0

) is approximated by

3,0

2,0

FC

1.00 0.02 FC

[3]

Th

ese estimated FC

i

,0

values (where

i

is the pool fraction and 0 is

the time) are then used, along with the measured total soil organic C
(CTOT

0

), to initialize all of the C pools. Th

e tables only need FC

2,0

,

and Eq. [1–3] are used to approximate the other pool fractions using
the measured CTOT

0

. If the cropping system and its management has

been practiced consistently for an extended period of time (e.g., 100 yr
or more), then it could be assumed that this intermediate soil C pool is
near its steady-state value for the climate and soil at this site. Th

erefore,

the previous cropping system could be simulated for 100 yr and the fi nal
value of simulated SOC

2,100

used in Eq. [2] along with the measured

CTOT

0

to initialize the simulation. If the cropping system has been in

place for only a few years, however, the intermediate pool would prob-
ably not be at its steady-state value. In this study, we estimated transient
SOC2 values using preliminary simulation results based on soil texture,
years during which the prior cropping system had been in place, and an
estimate of fi eld management prior to that crop.

For example, a fi eld with a known measured SOC value may have

been under maize (

Zea mays

L.) cropping for the past 8 yr, and before

that the fi eld was in fallow. To use the model to estimate the SOC frac-
tions for this example, the cropping system grown during the last 8 yr
needs to be simulated, using initial conditions estimated for the fi eld
when that cropping system was fi rst started, with the objective of match-
ing the current measured SOC value (aft er 8 yr). Although the initial
values of the soil C fractions 8 yr ago may not be accurate, the eff ects of
using incorrect estimates of the intermediate soil C pool fraction will
diminish with time as SOC2 transitions toward a new steady state.

Direct Use of Procedure for a Simulation Study

Tables created using the above procedure can aid model users in se-

lecting the most suitable SOC fraction distribution to properly initialize
the model. If users do not have tables created for their climate and crop-
ping systems, then they can use the procedure to initialize the SOC pools

14350661, 2011, 1, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.2136/sssaj2010.0115 by Hayley Peter-Contesse - Oregon State University , Wiley Online Library on [19/10/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

72

SSSAJ: Volume 75: Number 1 • January–February 2011

for any particular study. In this case, the model is initialized to simulate
the cropping history in an iterative way to obtain the measured CTOT
at the current time. To do this, SOC fractions from a table created for a
similar climate and soil are used to initialize the soil for the beginning of

the time period considered for the known cropping history (e.g., 20 yr
ago). Because total SOC at the beginning of this period is unknown, a
fi rst guess approximation is made based on soil texture and climate. Th

e

model is then executed for the duration of the cropping history time pe-
riod—20 yr in this example. Th

e simulated fi nal total SOC is compared

with the current SOC value, and if these are equal, the fi nal distribution
of SOC fractions is used with the measured SOC to restart the simula-
tion of the cropping system being studied. If the measured and simulated
values are not equal, then the procedure is repeated by changing the initial
CTOT. Th

is is repeated iteratively until the measured value of SOC at

the start of the experiment (i.e., the end of the antecedent simulation) is
adequately predicted. At this point, the intermediate (SOC2) and stable
organic C (SOC3) fractions can be noted and used to initialize the model.

Simulation Scenarios for Estimating
Initial Soil Organic Carbon Pool Fractions

We defi ned fi ve scenarios for estimating soil C pool fractions and

created tables to demonstrate the approach: (i) previously cultivated
soil under good management during the cropping history time period;
(ii) previously cultivated soil under poor management during the crop-
ping history period; (iii) a previous grassland soil under good man-
agement during the cropping history period; (iv) a previous grassland
soil under poor management during the cropping history period; and
(v) initially degraded soil under good management during the crop-
ping history period. Th

e simulation analysis was performed using the

DSSAT-CENTURY version 4.02 model (Hoogenboom et al., 2004)
on 12 soil texture classes (Table 1) to demonstrate the procedure and
results for diff erent combinations of soil texture. We simulated a maize
crop planted on 4 April of each year. In the fi rst scenario, we used a soil
that had been cultivated before the known cropping history time period
and imposed a nonlimiting fully irrigated crop management with high
N supply (306 kg N ha

−1

) (good management) to eliminate variations

with time associated with N stress and rainfall variability. Th

e manage-

ment used for the second scenario had no N or irrigation applications
(poor management). Th

e third and fourth scenarios had a grassland soil

cover before the time when the cropping history started, and crop man-
agement during the cropping history period used the same good or poor
management as the fi rst two scenarios. Th

e initial SOC used in these

scenarios was higher due to expected higher organic matter residues
from roots and plant litter present in the soil. Th is set of simulations was
performed with a one-time addition on the starting date of 4.0 Mg ha

−1

of fresh organic matter residues to approximate the conditions encoun-
tered when a grassland is converted to a cultivated fi eld. For the fi nal
scenario, a previously degraded soil with a low starting SOC2 fraction
was used and good management was imposed to simulate the eff ect of
this management combination on the dynamics of the SOC pool.

It is important to emphasize that in this study, we did not create

universal tables of SOC fractions but illustrated a generic approach to
SOC pool initialization based on climate, soil type, and management.
For this reason, we selected one high-quality climate data set and various
soil types and management strategies to investigate the sensitivity of the
SOC pool dynamics. It was our intent to illustrate the procedure and
how users can potentially adopt it to create site-specifi c SOC fractions
based on the soil and climatic conditions of their locations. For the mod-

Table 1. Physical and chemical characteristics of the soils used
in the simulation scenarios.

Parameter

Concentration per layer

5 cm 15 cm 30 cm 45 cm 60 cm 90 cm 120 cm150 cm

—————————— % ——————————

Sand

SOC

0.29 0.29 0.28 0.24 0.24 0.18 0.11

0.05

Clay

3

Silt

5

Loamy sand

SOC

0.7

0.66 0.58 0.58 0.43 0.26

0.12

Clay

6

Silt

12

Sandy loam

SOC

0.7

0.66 0.58 0.58 0.43 0.26

0.12

Clay

10

Silt

32

Silt loam

SOC

1.16 1.16 1.1

0.97 0.97 0.72 0.43

0.2

Clay

13

Silt

70

Silt

SOC

1.16 1.16 1.1

0.97 0.97 0.72 0.43

0.2

Clay

5

Silt

85

Loam

SOC

1.16 1.16 1.1

0.97 0.97 0.72 0.43

0.2

Clay

18

Silt

39

Sandy clay loam

SOC

1.75 1.75 1.6

1.45 1.45 1.1

0.65

0.3

Clay

27

Silt

15

Silty clay loam

SOC

1.75 1.75 1.6

1.45 1.45 1.1

0.65

0.3

Clay

34

Silt

56

Clay loam

SOC

1.75 1.75 1.6

1.45 1.45 1.1

0.65

0.3

Clay

34

Silt

34

Sandy clay

SOC

1.75 1.75 1.6

1.45 1.45 1.1

0.65

0.3

Clay

42

Silt

6

Silty clay

SOC

1.75 1.75 1.6

1.45 1.45 1.1

0.65

0.3

Clay

47

Silt

47

Clay

SOC

1.75 1.75 1.6

1.45 1.45 1.1

0.65

0.3

Clay

58

Silt

20

14350661, 2011, 1, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.2136/sssaj2010.0115 by Hayley Peter-Contesse - Oregon State University , Wiley Online Library on [19/10/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

SSSAJ: Volume 75: Number 1 • January–February 2011

73

el simulations, 20 yr of observed daily weather data from Tift on, GA,
were used as inputs. Monthly average air temperature (°C), precipitation
(mm d

−1

), and solar radiation (MJ m

−2

d

−1

) are summarized in Table 2,

computed using the 20 yr of historical daily weather data for this site. Th

e

initial soil C fractions for these 100-yr simulations were set to the values
for cultivated and grassland suggested by Gijsman et al. (2002).

Th

ese represent only a small fraction of cases that may be needed

for practical use of this approach. If simulations are performed continu-
ously for long periods of time, say 1000 or more yr, total SOC and pool
fractions can be very diff erent from those obtained using this approach,
and this procedure would not be needed. Th e procedures described
above are generally applicable, although the pool fractions obtained in
this study are specifi c to the particular site and management practices
and are not intended to be universal.

Validation on Long-Term Crop Rotation Experiment

Th

e study site selected to evaluate the proposed procedure was the

W.K. Kellogg Biological Station (KBS) Long-Term Ecological Research
(LTER) site (42°24´ N, 85°24´ W) in southwest Michigan (Grandy et
al., 2006a). Tillage treatments were established in 1989 in six replicated,
1-ha plots in a randomized complete block design. Th

e soils present on

the LTER site are Kalamazoo (a fi ne loamy, mixed, semiactive, mesic
Typic Hapludalf ) and Oshtemo (a coarse-loamy, mixed, active, mesic
Typic Hapludalf ). Th

e site was cleared of trees (roots were left in place

to decompose) in 1956 and thereby converted from a northern hard-
wood forest to a mid-successional grassland community that has since
been mowed each fall, with mown biomass left in place. Th

e site was

principally in continuous corn production for >20 yr before the estab-
lishment of the LTER project in 1989.

Th

e crop rotation before 1995 consisted of maize followed by soy-

bean [

Glycine max

(L.) Merr.]. In 1995, wheat (

Triticum aestivum

L.)

was planted aft er soybean, initiating a maize–soybean–wheat rotation.
Primary tillage consisted of spring moldboard plowing from 1989 to
1998 and spring chisel plowing from 1999 to 2002. Secondary tillage
consisted of disking before planting. Fertilizer was broadcast on maize
and wheat. Soybean did not receive N, P, or K fertilization (Grandy
et al., 2006b). Detailed information about annual applications can be

found on the KBS LTER website (lter.kbs.msu.edu/about/experimen-
tal_design.php; verifi ed 17 Nov. 2010).

For this study, we used the treatment defi ned as T1 (conventional till-

age, high external chemical input). We built a management sequence fi le on
the basis of the T1 treatment and executed the model for 13 yr using DSSAT
version 4.02. Th

e sequence fi le included a maize–soybean–wheat rotation,

conventional tillage, N fertilization, and crop residue incorporation.

RESULTS AND DISCUSSIONS

Tabular Results

Th

e simulated results of the SOC2 fraction for the 0- to

20-cm soil layer with time for the fi ve scenarios and the 12 soil
types are reported in Tables 3 to 6. Table 3 shows the changes in
the 0- to 20-cm depth of a fi eld that had been cultivated, under
good or poor management practices, before these simulations
for 12 soil types at Tift on, GA. Th

e results of the fi rst two sce-

narios demonstrated the sensitivity of changes in the SOC2 frac-
tion with time to management and soil type. Th

e initial SOC2

fraction used to initialize the soils for simulating the fi rst two
scenarios were typical values for a cultivated soil (FC

1

= 0.01%,

FC

2

= 0.54%, FC

3

= 0.44%) (A.J. Gijsman, personal communi-

Table 2. Weather data for Tifton, GA, expressed as monthly
averages.

Month

Temperature

Rain

Radiation

°C

mm

MJ m

−2

d

−1

January

9.2

4.2

February

11.5

3.7

March

14.9

4.5

April

18.2

2.2

May

22.7

2.0

June

25.6

3.6

July

27.0

3.8

August

26.6

3.2

September

24.2

3.4

October

19.4

2.5

November

14.9

2.6

December

10.4

2.7

Table 3. Changes in the intermediate soil organic C pool (SOC2) as a percentage of the total soil organic C with time in a culti-
vated soil (0–20 cm) under good and poor management practices for 12 soil types at Tifton, GA.

Soil

SOC2 pool

0 yr

5 yr

10 yr

20 yr

60 yr

100 yr

Poor

Good

Poor

Good

Poor

Good

Poor

Good

Poor

Good

Poor

Good

—————————————————————————— % ——————————————————————————

Sand

0.54

0.45

0.50

0.41

0.51

0.30

0.58

0.10

0.68

0.06

0.70

Loamy sand

0.54

0.43

0.44

0.40

0.39

0.37

0.13

0.39

0.05

0.39

Sandy loam

0.54

0.46

0.49

0.45

0.47

0.45

0.49

0.23

0.55

0.10

0.56

Sandy clay loam

0.54

0.43

0.42

0.37

0.34

0.32

0.27

0.18

0.24

0.14

0.24

Sandy clay

0.54

0.44

0.43

0.39

0.35

0.34

0.29

0.20

0.25

0.16

0.25

Loam

0.54

0.46

0.47

0.44

0.43

0.42

0.29

0.43

0.19

0.44

Clay loam

0.54

0.45

0.41

0.40

0.37

0.35

0.26

0.31

0.20

0.31

Clay

0.54

0.45

0.46

0.41

0.40

0.36

0.24

0.32

0.19

0.32

Silt loam

0.54

0.47

0.50

0.46

0.50

0.46

0.52

0.34

0.55

0.25

0.56

Silt

0.54

0.48

0.50

0.47

0.51

0.47

0.53

0.35

0.56

0.26

0.56

Silty clay loam

0.54

0.46

0.47

0.43

0.44

0.40

0.41

0.28

0.39

0.22

0.39

Silty clay

0.54

0.47

0.48

0.44

0.45

0.42

0.43

0.33

0.42

0.26

0.42

14350661, 2011, 1, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.2136/sssaj2010.0115 by Hayley Peter-Contesse - Oregon State University , Wiley Online Library on [19/10/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

74

SSSAJ: Volume 75: Number 1 • January–February 2011

cation, 2002). Even though the same initial SOC fractions were
used for all soils, the eff ects of fi eld history and management var-
ied signifi cantly with soil type and with poor and good manage-
ment. Th

e SOC2 fraction in the sandy soil increased under good

management, from 0.54 to 0.70% aft er 100 yr of simulation, while
the stable organic fraction, SOC3, decreased to a value of 29%.
Poor management in this scenario showed a continuous decrease
in the SOC2 fraction, as expected due to its lower biomass pro-
duction and amount of fresh organic matter returned to the soil.

In the silty clay loam soil, the SOC2 fraction varied during

100 yr under good management, from 0.54 to 0.39%, but signifi -
cantly decreased under poor management, reaching a fi nal value of
0.22 aft er 100 yr. Th

e clay soil showed a SOC2 fraction decrease

from 0.54 to 0.32% under good management but the SOC3 frac-
tion increased as a consequence of the slow mineralization and the
stronger organic C protection. In the cultivated scenarios, poor
management showed an exponential decay function for all the soils.

Table 4 shows the eff ects of poor and good management in

a fi eld that had been a grassland before the start of the simula-
tion study. Th

e initial SOC2 fraction used for this scenario was

0.64% for all soils. Th

e SOC2 fractions at the end of the 100-yr

period under poor management were similar to the cultivated
scenarios, meaning that near-steady-state SOC2 fractions were
reached in both cases. Th

e good management treatment in the

previous-grassland fi eld showed the same range of variation as
for the previously cultivated soils, demonstrating that soil texture
and management have greater impacts on SOC3 aft er 20 yr or
more than the initial SOC2 fractions.

Th

e changes in the SOC2 fractions with time in an ini-

tially degraded soil with good management practices are shown
in Table 5. In this scenario, the SOC2 fraction for 0- to 20-cm
depth increased with time from a minimum value of 0.05 in the
loamy sand at

t

= 0 to a maximum of 0.70% in the sand soil aft er

100 yr. Good management increased SOC2 fractions in all soils
even though diff erent initial values were used for each soil type.

Further illustrations of how changes in initial SOC fractions,

soil texture, and history aff ect changes in soil C mass with time are
shown in Fig. 1 to 6. Figure 1a shows the changes in soil organic
C mass with time for the 0- to 20-cm soil layer in a 100-yr maize
monoculture under poor management following grassland in a clay
soil. Th

e total SOC content changed with an exponential decline

from the initial value of 1.6 to 0.87% aft er 100 yr. Th

is change was

mainly due to decomposition of the SOC2 fraction while the stable
C (SOC3) remained constant. Figure 1b shows the SOC fraction
changes for the same scenario. Th

e SOC2 fraction changed from

an initial value of 0.64% to a value of 0.32%. Th

e SOC1 fraction

changed very little, with an approximate value of 1%. Th

e dynamic

of CTOT and the SOC fractions for the silt soil for the scenario be-
ginning with grassland under poor management is shown in Fig. 2.
Th

e silt soil showed a slower decline from its initial CTOT of 1.08 to

0.64%, and the SOC2 fraction increased to about 0.80 aft er 100 yr.
Th

e CTOT in the sand soil changed from 0.26 to 0.11% (Fig. 3a),

and the SOC2 fraction changed from 0.64 to 0.06%.

Table 4. Changes in the intermediate soil organic C pool (SOC2) as a percentage of total soil organic C with time in a grassland
soil (0–20 cm) under good and poor management practices for 12 soil types at Tifton, GA.

Soil

SOC2 pool

0 yr

5 yr

10 yr

20 yr

60 yr

100 yr

Poor

Good

Poor

Good

Poor

Good

Poor

Good

Poor

Good

Poor

Good

—————————————————————————— % ——————————————————————————

Sand

0.64

0.53

0.56

0.48

0.56

0.37

0.60

0.12

0.68

0.06

0.70

Loamy sand

0.64

0.50

0.45

0.43

0.41

0.39

0.14

0.39

0.05

0.39

Sandy loam

0.64

0.53

0.55

0.51

0.49

0.51

0.24

0.55

0.10

0.56

Sandy clay loam

0.64

0.50

0.49

0.42

0.38

0.34

0.29

0.18

0.24

0.12

0.24

Sandy clay

0.64

0.51

0.49

0.43

0.40

0.36

0.30

0.20

0.25

0.14

0.25

Loam

0.64

0.53

0.49

0.47

0.46

0.44

0.29

0.43

0.17

0.43

Clay loam

0.64

0.52

0.46

0.44

0.40

0.37

0.26

0.31

0.19

0.31

Clay

0.64

0.52

0.46

0.45

0.39

0.38

0.24

0.32

0.17

0.32

Silt loam

0.64

0.54

0.57

0.52

0.54

0.50

0.54

0.34

0.55

0.23

0.56

Silt

0.64

0.55

0.57

0.52

0.55

0.50

0.55

0.35

0.56

0.24

0.56

Silty clay loam

0.64

0.53

0.54

0.48

0.43

0.29

0.39

0.21

0.39

Silty clay

0.64

0.54

0.50

0.45

0.46

0.33

0.42

0.25

0.42

Table 5. Changes in the intermediate soil organic C pool
(SOC2), as a percentage of total soil organic C, with time in
an initially degraded soil under good management practices
for 12 soil types at Tifton, GA.

Soil

SOC2 pool

0 yr

5 yr

10 yr

20 yr

60 yr

100 yr

————————— % —————————

Sandy

0.06

0.17

0.30

0.48

0.67

0.70

Loamy sand

0.05

0.10

0.17

0.27

0.37

0.39

Sandy loam

0.10

0.15

0.25

0.39

0.54

0.56

Sandy clay loam

0.14

0.11

0.15

0.20

0.23

0.24

Sandy clay

0.16

0.12

0.16

0.21

0.25

Loam

0.19

0.16

0.24

0.33

0.43

Clay loam

0.20

0.15

0.20

0.26

0.31

Clay

0.19

0.15

0.21

0.27

0.32

Silt loam

0.25

0.21

0.31

0.43

0.54

0.56

Silt

0.26

0.22

0.32

0.45

0.56

Silty clay loam

0.22

0.17

0.24

0.32

0.38

0.39

Silty clay

0.26

0.19

0.26

0.34

0.42

14350661, 2011, 1, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.2136/sssaj2010.0115 by Hayley Peter-Contesse - Oregon State University , Wiley Online Library on [19/10/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

$Figures and page graphics, page 7$

SSSAJ: Volume 75: Number 1 • January–February 2011

75

Th

e eff ects of good management on the changes in the

CTOT and SOC fractions in the maize rotation with the prior
grassland fi eld history are illustrated in Fig. 4 to 6. Th

e CTOT

changed from 1.60 to 1.25% aft er 100 yr, while the SOC2 frac-
tion decreased from 0.64 to 0.32% and the SOC3 fraction in-
creased from 0.34 to 0.67% (Fig. 4a–4b). Th

e SOC2 fraction

in the silty soil slightly decreased; consequently the stable pool
increased from 0.34 to 0.43% (Fig. 5a–5b). Figure 6 shows the
dynamics of the CTOT (Fig. 6a) and SOC fractions (Fig. 6b)
in the sand soil. For this case, the CTOT increased from 0.25 to
0.30% and the stable C fraction decreased from 0.34 to 0.29%.

Direct Use of the Procedure

Table 6 illustrates the iterative procedure more clearly us-

ing an example from a silty clay loam soil where the total SOC
(CTOT, %) and SOC fraction changes were simulated for 18 yr
to reach a measured target value of 2.06% equivalent to a total
amount of 44,525 kg ha

−1

. Th

e simulated total SOC varied af-

ter 18 yr based on the diff erent initial organic C used iteratively
in the procedure. Th

e diff erence in organic C between the initial

organic C and the organic C used in the simulation can be added
to (Table 6, column ΔIOC) or subtracted from, as in the case of
the sand (Table 7, column ΔIOC), the initial soil C to adequately
predict the target organic C value. In our simulations, we changed
CTOT by 0.01% increments until the simulated and measured

Table 6. Iterative procedure to predict the measured soil C and intermediate soil organic C fraction (SOC2) as a percentage of the
total soil organic C, for a cultivated silty clay loam soil under good management.

Target

Total organic C after 18 yr

∆IOC†

IOC + ∆IOC

Error

SOC2 after 18 yr

SOC2

%

kg ha

−1

%

kg ha

−1

%

kg ha

−1

%

kg ha

−1

%

1.71

44,525

1.53

39,809

1

1.73

4716

10.59

18,902

0.42

1.71

44,525

1.69

43,990

16

1.99

535

1.20

19,984

0.45

1.71

44,525

1.70

44,258

17

2.00

267

0.60

20,056

0.45

1.71

44,525

1.71

44,516

18

2.02

9

0.02

20,123

0.45

1.71

44,525

1.72

44,785

19

2.04

−260

−0.58

20,195

0.45

1.71

44,525

1.73

45,041

20

2.06

−516

−1.16

20,262

0.46

1.71

44,525

1.74

45,310

21

2.07

−785

−1.76

20,332

0.46

1.71

44,525

1.75

45,568

22

2.09

−1043

−2.34

20,400

0.46

1.71

44,525

1.76

45,836

23

2.11

−1311

−2.95

20,471

0.46

1.71

44,525

1.77

46,094

24

2.12

−1569

−3.52

20,539

0.46

† ∆IOC, difference from initial organic C.

Fig. 1. Changes in (a) soil organic C (OC) and (b) the active soil
organic C fraction (SOC1), the intermediate soil organic C fraction
(SOC2), and the passive soil organic C fraction (SOC3) during 100 yr
of a maize rotation following grassland in a clay soil (0

–

20 cm) under

poor management.

Fig. 2. Change in (a) soil organic C (OC) and (b) soil organic matter
(SOM) fractions during 100 years of a maize rotation following
grassland in a silt soil (0

–

20 cm) under poor management.

14350661, 2011, 1, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.2136/sssaj2010.0115 by Hayley Peter-Contesse - Oregon State University , Wiley Online Library on [19/10/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

76

SSSAJ: Volume 75: Number 1 • January–February 2011

Fig. 3. Change in (a) soil organic C (OC) and (b) the active soil organic C
fraction (SOC1), the intermediate soil organic C fraction (SOC2), and the
passive soil organic C fraction (SOC3) during 100 years of a maize rotation
following grassland in a sand soil (0–20 cm) under poor management.

Fig. 4. Change in (a) soil organic C (OC) and (b) the active soil organic
C fraction (SOC1), the intermediate soil organic C fraction (SOC2),
and the passive soil organic C fraction (SOC3) during 100 years of
a maize rotation following grassland in a clay soil (0–20 cm) under
good management.

Fig. 5. Change in (a) soil organic C (OC) and (b) the active soil organic C
fraction (SOC1), the intermediate soil organic C fraction (SOC2), and the
passive soil organic C fraction (SOC3) during 100 years of a maize rotation
following grassland in a silt soil (0–20 cm) under good management.

Fig. 6. Change in (a) soil organic C (OC) and (b) the active soil organic C
fraction (SOC1), the intermediate soil organic C fraction (SOC2), and the
passive soil organic C fraction (SOC3) during 100 years of a maize rotation
following grassland in a sand soil (0–20 cm) under good management.

14350661, 2011, 1, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.2136/sssaj2010.0115 by Hayley Peter-Contesse - Oregon State University , Wiley Online Library on [19/10/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

SSSAJ: Volume 75: Number 1 • January–February 2011

77

target values were closest in magnitude. Th

e smallest error be-

tween measured and simulated target CTOT values in this case
was reached aft er changing our initial guess of CTOT by 0.18%
aft er 18 iterations. Th

e corresponding SOC2 fraction at the end

of the 18 yr, to be used for further simulations, was 0.45%.

Table 7 shows the amount of SOC to be added to the initial

estimate in this example to closely predict the target SOC value.
Also shown are the SOC2 fractions to be used for future simula-
tions. Note that to match the SOC under good management in
a sand soil, the change in organic C was −0.08% in the sand soil
and +0.27% in the clay soil.

Validation of Procedure Using
a Long-Term Rotation Experiment

As mentioned above, Tables 4 to 6 developed for Tift on, GA,

cannot be directly used for initializing SOC fractions in places
with diff erent climatic conditions. To evaluate the procedure pre-
sented here and to demonstrate estimation of SOC fractions in
places other than Tift on, we used 13 yr of SOC values measured in
the KBS study. We ran a maize rotation for 13 yr using an SOC2
value of 0.54%, which was found to be the best SOC2 fraction to
minimize the error between the measured and simulated results
aft er 13 yr (Table 8). Th

e model performed well when compared

with the measured values of SOC, with a RMSE of 0.023%. Using
our procedure, compared with the Gijsman et al. (2002) partition-
ing, the improved initial values resulted in a reduction of 10% in
the error between the model and measured values during the 18 yr.

Th

e CTOT percentage and SOC2 fractions for the KBS

site (Fig. 7) show that the SOC2 at the end of the 13 yr changed
from 0.54 to 0.52%. Th

e latter value is the one to use for further

simulations if the fate of soil C for this location were further in-
vestigated. Th

e value of SOC2 for a cultivated loam soil under

good management (like the conditions at the KBS site during
13 yr), for Tift on, GA, is around 0.47%. Th

e values of SOC2 for

Tift on and KBS are diff erent, which demonstrates the interaction
of climate with soils on SOC dynamics, in this case a faster overall
turnover time in SOC in the warmer soils of Tift on. Th

is suggests

caution in using tables for climatically distinct locations, diff erent
from where they were created, unless they are only used to provide
an indication of the magnitude of the SOC fraction to be used to
initialize the iterative procedure proposed in this study.

CONCLUSIONS

We used an approach to demonstrate the eff ects and interac-

tions of soil type and management on the variations in SOC pools.
We used the DSSAT-CENTURY version 4.02 model and SOC
equations to estimate the changes in SOC fractions under diff erent

Table 8. Measured and simulated organic C (0–20 cm) at
Kellogg Biological Station using intermediate soil organic C
fraction of the total soil organic C (SOC2) = 0.54%.

Year Measured organic C Simulated organic C

Error

——————————— % ———————————

1988

0.85

0

1989

0.84

0

1990

0.85

0.83

−0.2

1991

0.82

0.83

0.1

1992

0.81

0.83

0.2

1993

0.85

0.83

−0.2

1994

0.83

0.82

−0.1

1995

0.86

0.85

−0.1

1996

0.85

0.83

−0.2

1997

0.89

0.84

−0.5

1998

0.87

0.86

−0.1

1999

0.82

0.84

0.2

2000

0.81

0.85

0.4

2006

0.82

0.83

0.1

Table 7. Iterative procedure results for intermediate soil organic C fraction (SOC2) estimation for different soil types.

Texture

Initial organic C

Target

SOC2 fraction

∆IOC†

%

kg ha

−1

%

kg ha

−1

%

kg ha

−1

%

Sand

0.27

8799

0.288

9546

0.559

5341

−8

Silt

1.12

30763

1.145

31373

0.535

16798

−2

Silt Clay Loam

2.02

52551

1.713

44525

0.452

20123

18

Clay

2.18

56567

1.713

44525

0.415

18475

27

† ∆IOC, difference from initial organic C.

Fig. 7. Change in (a) soil organic C (OC) and (b) the active soil organic C
fraction (SOC1), the intermediate soil organic C fraction (SOC2), and the
passive soil organic C fraction (SOC3) during 100 years of a maize rotation
following grassland in a loam soil (0–20 cm) under good management.

14350661, 2011, 1, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.2136/sssaj2010.0115 by Hayley Peter-Contesse - Oregon State University , Wiley Online Library on [19/10/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

78

SSSAJ: Volume 75: Number 1 • January–February 2011

soil and management conditions with a high-quality climate data
set recorded for Tift on, GA. We have shown how the approach can
be used to simulate the cropping history in an iterative way to ob-
tain the measured CTOT at the starting year for a fi eld experiment
in order to estimate the relative distribution of SOC across the
model pools for subsequent modeling of the experimental period.

Th

e advantages of using the proposed procedure are intui-

tive because it provides help for the user facing the diffi

cult task of

choosing the proper SOC fractions to initialize the dynamic mod-
el. Th

is procedure should be adopted to create site-specifi c SOC

fractions based on the soil and climatic conditions of a specifi c lo-
cation. Th

e model provided excellent results when compared with

measured soil C data using the SOC fraction that minimized the
error between measured and simulated results aft er the model was
executed with the iterative procedure described in here.

REFERENCES

Bado, B.V. 2002. Rôle des légumineuses sur la fertilité des sols ferrugineux

tropicaux des zones guinéenne et soudanienne du Burkina Faso. (In French,
with English abstract.) Ph.D. diss. Univ. Laval, Québec City, QC, Canada.

Bostick, W.M., V.B. Bado, A. Bationo, C.T. Soler, G. Hoogenboom, and J.W. Jones.

2007. Soil carbon dynamics and crop residue yields of cropping systems in
the Northern Guinea Savanna of Burkina Faso. Soil Tillage Res. 93:138–151.

Elliott, E.T., K. Paustian, and S.D. Frey. 1996. Modeling the measurable or measuring

the modelable: A hierarchical approach to isolating meaningful soil organic
matter fractions. p. 161–179.

In

D.S. Powlson et al. (ed.) Evaluation of soil

organic matter models using existing, long-term datasets. Springer-Verlag, Berlin.

Falloon, P.D., and P. Smith. 2000. Modelling refractory soil organic matter. Biol.

Fertil. Soils 30:388–398.

Falloon, P., and P. Smith. 2003. Accounting for changes in soil carbon under

the Kyoto Protocol: Need for improved long-term data sets to reduce
uncertainty in model projections. Soil Use Manage. 19:265–269.

Falloon, P., P. Smith, J. Szabo, and L. Pasztor. 2002. Comparison of approaches

for estimating carbon sequestration at the regional scale. Soil Use Manage.
18:164–174.

Gijsman, A.J. 1996. Soil aggregate stability and soil organic matter fractions under

agropastoral systems established in native savanna. Aust. J. Soil Res. 34:891–907.

Gijsman, A.J., G. Hoogenboom, W.J. Parton, and P.C. Kerridge. 2002. Modifying

DSSAT crop models for low-input agricultural systems using a soil organic
matter–residue module from CENTURY. Agron. J. 94:462–474.

Grandy, A.S., G.P. Robertson, and K.D. Th

elen. 2006a. Do productivity and

environmental trade-off s justify periodically cultivating no-till cropping
systems? Agron. J. 98:1377–1383.

Grandy, A.S., D.L. Terrance, S. Parr, and G.P. Robertson. 2006b. Long-term

trends in nitrous oxide emissions, soil nitrogen, and crop yields of till and
no-till cropping systems. J. Environ. Qual. 35:1487–1495.

Gregorich, E.G., and H.H. Janzen. 1996. Storage of soil carbon in the light

fraction and macro-organic matter. p. 167–190.

In

M.R. Carter and B.A.

Stewart (ed.) Structure and organic matter storage in agricultural soils.
CRC Press, Boca Raton, FL

Hayes, M.H.B. 1985. Extraction of humic substances from soil. p. 329–362.

In

G.R.

Aiken et al. (ed.) Humic substances in soil, sediment, and water: Geochemistry,
isolation, and characterization. John Wiley & Sons, New York.

Hoogenboom, G., J.W. Jones, P.W. Wilkens, C.H. Porter, W.D. Batchelor, L.A.

Hunt, et al. 2004. Decision Support System for Agrotechnology Transfer,
version 4.0 [CD-ROM]. Univ. of Hawaii, Honolulu.

Janik, L.J., J.O. Skyemstad, K.D. Shepherd, and L.R. Spouncer. 2007. Th

e

prediction of soil carbon fractions using mid-infrared-partial least square
analysis. Aust. J. Soil Res. 45:73–81.

Jenkinson, D.S. 1981. Th

e fate of plant and animal residues in soil. p. 505–561.

In

D.J. Greenland and M.H.B. Hayes (ed.) Th

e chemistry of soil processes.

John Wiley & Sons, Chichester, UK.

McGill, W.B. 1996. Review and classifi cation of ten soil organic matter (SOM)

models. p. 111–133.

In

D.S. Powlson et al. (ed.) Evaluation of soil organic

matter models using existing, long-term datasets. Springer-Verlag, Berlin.

Metherell, A.K., L.A. Harding, C.V. Cole, and W.J. Parton. 1993. CENTURY

manual: CENTURY soil organic matter model environment technical
documentation. Agroecosystem Version 4.0. Tech. Rep. 4. USDA-ARS
Great Plains Syst. Res. Unit, Fort Collins, CO.

Ogle, S.M., F.J. Breidt, M. Easter, S. Williams, K. Killian, and K. Paustian. 2010.

Scale and uncertainty in modeled soil organic carbon stock changes for U.S.
croplands using a process-based model. Global Change Biol. 16:810–822.

Ogle, S.M., F.J. Breidt, M. Easter, S. Williams, and K. Paustian. 2007. An

empirically-based approach for estimating uncertainty associated with
modeling carbon sequestration in soils. Ecol. Modell. 205:453–463.

Olk, D.C., and E.G. Gregorich. 2006. Overview of the symposium proceedings,

“Meaningful pools in determining soil carbon and nitrogen dynamics.

”

Soil Sci. Soc. Am. J. 70:967

–974

.

Parton, W.J., C.V. Cole, J.W.B. Stewart, D.S. Ojima, and D.S. Schimel. 1988a.

Simulating regional patterns of soil C, N, and P dynamics in the U.S.
Central Grasslands region. Biogeochemistry 5:109–131.

Parton, W.J., and P.E. Rasmussen. 1994. Long-term eff ects of crop management

in wheat–fallow: II. CENTURY model simulations. Soil Sci. Soc. Am. J.
58:530–536.

Parton, W.J., D.S. Schimel, C.V. Cole, and D. Ojima. 1987. Analysis of factors

controlling soil organic matter levels in Great Plains grasslands. Soil Sci.
Soc. Am. J. 51:1173–1179.

Parton, W.J., J.W. Stewart, and C.V. Cole. 1988b. Dynamics of C, N, P and S in

grassland soils: A model. Biogeochemistry 5:109–131.

Paustian, K. 1994. Modelling soil biology and biogeochemical processes for

sustainable agriculture research. p. 182–196.

In

C. Pankhurst et al. (ed.)

Management of soil biota in sustainable farming systems. CSIRO Publ.,
Melbourne, VIC, Australia.

Paustian, K., E.T. Elliott, H.P. Collins, C.V. Cole, and E.A. Paul. 1995. Use of a

network of long-term experiments for analysis of soil carbon dynamics and
global change: Th

e North America model. Aust. J. Exp. Agric. 35:929–939.

Paustian, K., E. Levine, W.M. Post, and I.M. Ryzhova. 1997. Th

e use of models

to integrate information and understanding of soil C at the regional scale.
Geoderma 79:227–260.

Plante, A.F., and J.W. Parton. 2007. Th

e dynamics of soil organic matter and

nutrient cycling. p. 433–464.

In

E.A. Paul (ed.) Soil microbiology, ecology

and biochemistry. 3rd ed. Academic Press, San Diego.

Senthilkumar, S., B. Basso, A. N. Kravchenko, G. P. Robertson. 2009.

Contemporary evidence of soil carbon loss in the U.S. Corn Belt. Soil Sci.
Soc. Am. J. 73: 6 2078–2086

Six, J., H. Bossuyt, S. Degryze, and K. Denef. 2004. A history of research on

the link between (micro)aggregates, soil biota, and soil organic matter
dynamics. Soil Tillage Res. 79:7–31.

Smith, J.U., P. Smith, R. Monaghan, and A.R.J. Macdonald. 2002. When is a

measured soil organic matter fraction equivalent to a model pool? Eur. J.
Soil Sci. 53:405–416.

Stevenson, F.J. 1994. Humus chemistry: Genesis, composition, reactions. 2nd ed.

John Wiley & Sons, New York.

van Veen, J.A., J.N. Ladd, and M.J. Frissel. 1984. Modeling C and N turnover

through the microbial biomass in soil. Plant Soil 76:257–274.

van Veen, J.A., and E.A. Paul. 1981. Organic carbon dynamics in grassland soils: 1.

Background information and computer simulation. Can. J. Soil Sci. 61:185–201.

von Lützow, M., I. Kögel-Knabner, K. Ekschmitt, H. Flessa, G. Guggenberger, E.

Matzner, and B. Marschner. 2007. SOM fractionation methods: Relevance
to functional pools and to stabilization mechanisms. Soil Biol. Biochem.
39:2183–2207.

Wander, M. 2004. Soil organic matter fractions and their relevance to soil

function. p. 67–102.

In

F. Magdoff and R.R. Weil (ed.) Soil organic matter

in sustainable agriculture. CRC Press, Boca Raton, FL.

Yeluripati, J.B., M. van Oijen, M. Wattenbach, A. Neft el, A. Ammann, W.J.

Parton, and P. Smith. 2009. Bayesian calibration as a tool for initialising the
carbon pools of dynamic soil models. Soil Biol. Biochem. 41:2579–2583.

14350661, 2011, 1, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.2136/sssaj2010.0115 by Hayley Peter-Contesse - Oregon State University , Wiley Online Library on [19/10/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

Basso et al. (2011). Procedures for Initializing Soil Organic Carbon Pools in the DSSAT‐CENTURY Model for Agricultural Systems

Page 1

Page 2

Page 3

Page 4

Page 5

Page 6

Page 7

Page 8

Page 9

Page 10