Section IV Soil Carbon Dynamics and Farming/Cropping Systems © 2005 by Taylor & Francis Group, LLC 407 16 Soil Carbon Sequestration: Understanding and Predicting Responses to Soil, Climate, and Management JAMES W. JONES, VALERIE WALEN, MAMADOU DOUMBIA, AND ARJAN J. GIJSMAN CONTENTS 16.1 Introduction 408 16.2 Combining Models and Data to Assess Options for Soil C Sequestration 411 16.2.1 Model Adaptation to Local Conditions 412 16.2.1.1 Soil Data 413 16.2.1.2 Weather Data 414 16.2.1.3 Agronomic Experiment Data 415 16.2.2 Simulation of Soil C Sequestration Potential, RT vs. CT 416 © 2005 by Taylor & Francis Group, LLC 408 Jones et al. 16.3 Combining Measurements and Models for Estimating SOC Sequestration 418 16.3.1 Soil Carbon Model 419 16.3.2 Soil Carbon Measurements 420 16.3.3 The Ensemble Kalman Filter: Combining Model and Measurements 421 16.3.4 Example Results 423 16.4 Discussion 427 Acknowledgments 429 References 429 16.1 INTRODUCTION Managing agricultural lands to increase soil organic carbon (SOC) could help counter the rising atmospheric CO 2 concen- tration as well as reduce soil degradation and improve crop productivity. However, soils, climate, and management prac- tices vary over space and time, creating an almost infinite combination of factors that interact and influence how much carbon is stored in soils. Thus, quantifying soil carbon seques- tration under widely varying conditions is complicated. Fur- thermore, SOC changes slowly over time; experiments for quantifying carbon gain under different practices must be conducted over a number of years. Due to the human and financial resources and time needed to conduct such experi- ments, it may not be practical to rely on this approach alone to provide needed information. Further complicating the pic- ture is climate change. As temperature and atmospheric CO 2 increase and rainfall changes, new combinations of factors will occur that have not been studied. For these reasons, models are needed to complement information gained from experiments to help understand and predict SOC and food production responses to soil, climate, and management com- binations. Biophysical models integrate crop, soil, weather, and management practice information and predict the consequent biomass and yield components as well as changes in soil nutrients and carbon (Cole et al., 1987; Moulin and Beckie, © 2005 by Taylor & Francis Group, LLC Soil Carbon Sequestration 409 1993; Singh et al., 1993; Probert et al., 1995; Gijsman et al., 2002a; Jones et al., 2003). By simulating responses for a number of years, it is possible to estimate potential changes in productivity and SOC. Through a series of computer exper- iments, using models along with local soil and climate infor- mation, one could identify cropping systems that would meet productivity and SOC sequestration goals. However, there are a number of uncertainties associated with models and their use, and if one does not adequately address these uncertain- ties, simulated results will be meaningless. These uncertain- ties are due to the fact that models are simplifications of reality, there are uncertainties in model parameters, and there are uncertainties in inputs used in computer experi- ments. Thus, work is needed to ensure that models can repro- duce responses measured in real experiments. This may require one to estimate crop and soil parameters (e.g., Mavro- matis et al., 2001; Gijsman et al., 2002b), to adjust other relationships to adapt the model for the region in which it is to be used (e.g., du Toit et al., 1998), and possibly conduct new research to evaluate predictions. If the model accurately describes yield and SOC responses measured in real experi- ments in the region, one will have more confidence in its ability to predict responses under other combinations of soil, weather, and management practices. Biophysical models may also be useful in monitoring SOC changes over time and space to fulfill carbon contract require- ments. Although this is not a common use of agricultural models, methods developed in other fields of science and engi- neering can be applied to help quantify and verify soil carbon sequestration. Once models have been adapted for a region, they can be used to predict changes in soil C under weather conditions that occur each year and for management practices actually used at lower cost than empirical research (Bationo et al., 2003). However, model predictions are uncertain, even if inputs are accurate. Spatial variability of inputs adds to the uncertainties outlined above, which results in propagation of prediction errors over space and time. Measurements of carbon also are uncertain and costly; errors may be much larger than annual changes in SOC. Thus, by combining © 2005 by Taylor & Francis Group, LLC 410 Jones et al. measurements with model predictions, more accurate esti- mates of SOC can be obtained (Jones et al., 2004; Koo et al., 2003; Bostick et al., 2003). Existing models are useful tools for understanding and predicting SOC changes if they are combined with measure- ments and used carefully. Our objective is to demonstrate the use of biophysical models in combination with data for two different types of uses. In the first demonstration, we explore options for increasing yield and SOC in a maize farming system in Mali. West and Post (2002) found a global average C sequestration rate of 570 kg ha −1 year −1 for no-till vs. con- ventional tillage when they analyzed data from 67 long-term experiments from around the world. In a 10-year study in Burkina Faso, soil C increase averaged 116 and 377 kg ha −1 year −1 for treatments with low and high levels of both inor- ganic fertilizer and manure, respectively (Pichot et al., 1981). Lal (2000) observed annual rates of soil C increase under no- till management ranging from 363 kg ha −1 year −1 to more than 1000 (for one severely depleted soil) over a 3-year experiment aimed at restoring soil carbon in western Nigeria. Because soil C in western African soils is known to be depleted (Bationo et al., 2003), the hypothesis used to guide our study was that ridge tillage (RT) combined with manure, nitrogen fertilizer applications, and residue management will increase soil carbon by 0.20% in 10 years (about 500 kg ha −1 year −1 ) relative to levels under conventional tillage (CT) manage- ment. The DSSAT-CENTURY model is used to simulate annual maize growth and yield as well as changes in SOC for 10 years. But first, care is taken to adapt the model to maize cultivars, soil, management, and climate conditions of Mali using available, although limited, data. In the second demon- stration, the hypothesis is that model predictions of soil car- bon can be combined with in situ measurements to improve estimates of soil carbon sequestration. An ensemble Kalman filter approach is used to assimilate observations over time into a simple model to increase accuracy of SOC estimates and to improve future predictions for specific fields. © 2005 by Taylor & Francis Group, LLC Soil Carbon Sequestration 411 16.2 COMBINING MODELS AND DATA TO ASSESS OPTIONS FOR SOIL C SEQUESTRATION Two of the biggest constraints for improving household food security in West Africa are retention of rainwater in the field and improvement of soil quality (Kaya, 2000; Lal, 1997a, 1997b; Ringius, 2002; Bationo et al., 2003). The practice known as ridge tillage or aménagement en courbes de niveau (Gigou et al., 2000) was designed to address these issues concurrently, and is thought to have potential for sequestering SOC. This is logical since ridge tillage increases crop biomass production and grain yield in Mali (Gigou et al., 2000). Unfor- tunately, data for evaluating SOC sequestration potential in West Africa are scarce (Lal, 1997a, 1997b; Pieri, 1992; Ring- ius, 2002). Estimates of SOC sequestration potential are needed to help guide research and to give donors confidence that their investments will succeed. Soil C measurements taken by Yost and colleagues (Yost et al., 2002; Neely and Uehara, 2002) show that SOC levels in Mali are very low (ranging from 0.13% to 0.88% of soil mass) in the top 20 cm, and that fields that have been under ridge tillage for several years tend to have higher SOC levels than fields under con- ventional tillage. However, few measurements have been made to date, and thus no conclusions can be made regarding how much SOC will increase under RT, nor how long it might take to achieve that increase. Our hypothesis was that RT, coupled with other soil management practices, could increase soil C in the top 20 cm of soil by 5 metric tons ha −1 over 10 years. Objectives were (1) to adapt the DSSAT-CENTURY maize model for simulating RT vs. CT management systems in Mali using available data, and (2) to conduct a 10-year computer experiment to make preliminary estimates of poten- tial SOC sequestration amounts under CT vs. RT manage- ment systems. This study demonstrates the adaptation of a cropping system model for studying management options for increasing soil carbon in Oumarbougou, Mali (Lat 12.18 N, Long 5.14 © 2005 by Taylor & Francis Group, LLC 412 Jones et al. W). Rainfall in the region is 900 to 1000 mm per year, falling unimodally from June to October (Roncoli et al., 2002). The cultivated soils in the area are characterized as red sandy soils (bogo bile), generally alfisols with high sand content and low organic C and N. The area is highly prone to runoff and erosion, as is the case in much of West Africa (Bielders et al., 1996; Daba, 1999; Rockstrom et al., 1998; Zhang and Miller, 1996). 16.2.1 Model Adaptation to Local Conditions The Decision Support System for Agrotechnology Transfer (DSSAT), with its suite of CERES- and CROPGRO-based crop models, was developed to help researchers understand crop responses to various management options, soils, and weather conditions (Tsuji et al., 1998; Jones et al., 2003). The CEN- TURY soil organic matter model, originally developed to sim- ulate soil C dynamics in temperate grasslands (Parton et al., 1987), has since been used in a wide range of conditions including tropical systems (Paustian et al., 1992; Parton et al., 1988, 1994; Woomer, 1993; Anderson and Ingram, 1993; International Centre for Research in Agroforestry, 1994). Recently, the CENTURY model was linked with the DSSAT cropping system model to improve capability for simulating cropping systems with low inputs (Gijsman et al., 2002a; Jones et al., 2003). This linked DSSAT–CENTURY model was used in this study. Answers are sought to the following questions: (1) Does the model adequately simulate growth and yield of the crops under the soils, climate, and management conditions being considered? (2) Does the model adequately simulate changes in soil processes (including SOC) under those same condi- tions? One can be relatively sure that existing models, even robust, widely used models like DSSAT, will not perform well in a new location unless an effort is made to adapt them to local conditions. Adapting the model requires: (1) assembly of local data on soil, weather, and crop performance under field conditions, (2) estimation of crop model parameters for local cultivars, (3) estimation of critical soil parameters not © 2005 by Taylor & Francis Group, LLC Soil Carbon Sequestration 413 normally measured (particularly soil hydraulic properties), and (4) evaluation of model ability to simulate crop (e.g., phenological development, yield, and biomass) and soil (e.g., water, SOC, and N) responses under local conditions. Although appropriate data for this area were limited (i.e., no long-term experiments with observed SOC changes), avail- able data were used to simulate these preliminary estimates of SOC sequestration. In this study, continuous use of maize was assumed to demonstrate the approach. The first step was to adapt the model for simulating maize cultivars normally grown in Mali agronomic experiments. The second step was to adjust runoff characteristics for CT vs. RT so that published differences in runoff and crop yield between these two systems were correctly simulated. The final step was to adjust initial C fractions so that SOC under CT was at steady state. These procedures allowed us to confirm that the model correctly simulates absolute yield levels as well as differences between the two systems that are being compared. 16.2.1.1 Soil Data Soil samples collected by Mamadou Doumbia and Russ Yost in March 2002 were used to develop necessary soil profile inputs to the model. A composite of soils sampled from the fields of Zan Diarra, (Lat 12.55 N, Long 6.47 W) and of Yaya Diassa (Lat 11.14 N, Long 5.35 W) was used to create a soil input file with parameters listed in Table 16.1. Soil water- Table 16.1 Selected Soil Inputs in Zan Diarra Samples and Yaya Diassa Soils Soil Depth (cm) SOC (%) Sand (%) Silt (%) Clay (%) pH Wilting Point a (cm 3 cm −3 ) Field Capacity b (cm 3 cm −3 ) Bulk Density (g cm −3 ) 0–20 0.24 a 72.4 21.4 6.2 5.34 0.069 0.176 1.44 20–40 0.22 52.9 25.3 21.8 4.93 0.213 0.297 1.49 a Initial soil C was assumed to be 7016 kg ha −1 in the top 20 cm. b Calculated using the method described by Jagtap et al. (2004). Source: From M. Doumbia, personal communication, 2003. © 2005 by Taylor & Francis Group, LLC 414 Jones et al. holding characteristics were estimated from soil texture using the nearest neighbor method of Jagtap et al. (2004). SOC composition in the DSSAT–CENTURY model is initialized by partitioning total C into three pools based on rates of decom- position: microbial, slow, and stable, with default fractions for grassland and previously-cultivated soils of 02:64:34 and 02:54:44, respectively. Since we had no measurements that would allow us to estimate these fractions directly, we assumed that soil C under CT was at a steady state. Thus, we varied these fractions for CT simulations until we achieved a steady-state level of SOC. When fractions of 02:41:57 were used for Mali soil, climate, and CT management, SOC remained at 0.24% for the 10-year period of simulations (see results for CT in Figure 16.1). 16.2.1.2 Weather Data Historical daily weather data are needed for simulating exper- iments conducted in the past and evaluating model predic- tions vs. observations. Observed daily weather data were obtained in order to compare simulated maize results with Figure 16.1 Plot of annual change in soil organic carbon (SOC) over 10 years under conventional tillage and fully implemented ridge tillage using initial SOC composition of calibrated stability (02:41:57). Soil Organic Carbon 0-20 cm 0.2 0.25 0.3 0.35 0.4 Time, yrs CT RT All 0246810 % 50C © 2005 by Taylor & Francis Group, LLC Soil Carbon Sequestration 415 those obtained by Coulibaly (i.e., Table 16.2). We also gener- ated 10 years of daily weather data by interpolation between nearest existing weather stations using MarkSim, version 1 (P. Jones et al., 2002). The generated daily data include rain- fall, maximum temperature, minimum temperature, and solar radiation. Small amounts of N (13 kg ha −1 100 cm −1 infiltrated rainfall) (Campbell, 1978; Pieri, 1992; Vitousek et al., 1997) were applied to all simulated crops according to infiltration of rainfall. 16.2.1.3 Agronomic Experiment Data Agronomic yield trial data for a 3-year maize study were obtained from Njti Coulibaly in Mali, including soil, weather, and management of the crops in each year. That experiment was simulated using the DSSAT CERES-Maize model, and genetic coefficients were estimated using measured anthesis dates and yields for the 3 years (Jones et al., 2002a). Data in Table 16.2 demonstrate that the model describes anthesis dates and yields across the 3 years with errors less than 10%. Although additional tests are desirable, this exercise demon- strated that the model can simulate growth and yield responses to typical growing conditions in Mali under conven- tional management. Detailed measurements from experiments comparing RT vs. CT were not available. Thus, a computer experiment was conducted over a 10-year period: (1) to adjust field runoff parameters for RT vs. CT, and (2) to compare predicted grain and biomass yield values for RT and CT with those responses Table 16.2 Calibration of Local Maize Variety Sotubaka: Three Years of Observed and Simulated Grain Yield and Days to Anthesis Time to Anthesis (days) Maize Grain Yield (kg ha −1 ) 1999 2000 2001 1999 2000 2001 Simulated 62 61 57 5486 4138 5514 Observed 63 61 58 5100 3900 6070 Source: From Coulibaly, Ntji, personal communication, 2002. © 2005 by Taylor & Francis Group, LLC [...]... water, nitrogen, and crop yield for a long-term fallow management experiment Aust J Exp Agric., 35: 941 –950 Ringius, L 2002 Soil carbon sequestration and the CDM: Opportunities and challenges for Africa Climatic Change, 54: 471 49 5 Rockstrom, J., P.-E Jansson, and J Barron 1998 Seasonal rainfall partitioning under runon and runoff conditions on sandy soil in Niger On-farm measurements and water balance... Limits, and Tradeoffs JOHN M DUXBURY CONTENTS 17.1 Conceptual Basis for Carbon Sequestration in Soils 43 6 17.2 Opportunities for Carbon Sequestration in Soils 44 0 17.3 Impact of Tillage Management on Greenhouse Gas Fluxes 44 0 17 .4 GWP Analysis for Conventional and No-Tillage Maize Production in the United States 44 1 17.5 Long-Term GWP Effects of Changing Tillage Practice 44 5... in Agroforestry 19 94 Slash-andBurn: Update on Alternatives Vol 1, no 2 International Centre for Research in Agroforestry, Nairobi, Kenya Jagtap, S.S., U Lall, J.W Jones, A.J Gijsman, and J.T Ritchie 20 04 A dynamic nearest neighbor method for estimating soil water parameters Trans ASAE, 47 (5): 143 7– 144 4 Jones, J.W., K Boote, and G Hoogenboom 2002a Crop Modeling Team Trip Report: Mali and Ghana NASA Carbon... properties Soil Tillage Res., 42 :161–1 74 Lal, R 2000 Land use and cropping system effects on restoring soil carbon pool of degraded alfisols in Western Nigeria In Lal, R., J.M Kimble, and B.A Stewart, Eds Global Change and Tropical Ecosystems Lewis Publishers, Boca Raton, FL, pp 157–165 Margulis, S.A., D McLaughlin, D Entekhabi, and S Dunne 2002 Land data assimilation and soil moisture estimation using... al 2000 1500 1000 500 29 27 25 23 21 19 17 15 13 11 -5 00 9 7 5 3 0 1 Annual changes in SOC, kg ha-1 42 6 Year -1 000 -1 500 Measured EnKF Estimates True Values Figure 16.3 Annual changes in soil organic carbon comparing EnKF estimates with measured and true values (Modified from Jones, J.W., W.D Graham, D Wallach, W.M Bostick, and J Koo 20 04 Trans ASAE, 47 (1):331–339.) Meas = 1:1 Year Meas = 1:2 Years Meas... Gijsman, and J.B Naab 2003 Estimating soil carbon in agricultural using ensemble Kalman filter and DSSAT-CENTURY ASAE Paper 033 041 American Society of Agricultural Engineers, St Joseph, MO Lal, R 1997a Long-term tillage and maize monoculture effects on a tropical alfisol in Western Nigeria I Crop yield and soil physical properties Soil Tillage Res., 42 : 145 –160 Lal, R 1997b Long-term tillage and maize... Changing Tillage Practice 44 5 17.6 Conclusions 44 6 Acknowledgments 44 7 References 44 7 43 5 © 2005 by Taylor & Francis Group, LLC 43 6 Duxbury Carbon sequestration in soils is a land-based option to reduce the greenhouse warming potential (GWP) of the atmosphere It has the additional benefits of improving soil quality and the sustainability of agriculture This chapter discusses... better understand and predict SOC sequestration The importance of both data and models needs to be recognized as efforts are made to improve knowledge and tools for use in science and policymaking ACKNOWLEDGMENTS This research is supported by the Soil Management Collaborative Research Program (SM CRSP) through a grant (LAGG-0 0-9 7-0 000 2-0 0) from the U.S Agency for International Development and by a grant... with mean values of 0 .4 to 0.5 metric tons C/ha/year for moist temperate and tropical environments West and Marland (2002) estimate an average rate of 0. 34 ± 0.1 metric tons C/ha/year for cropland in the United States, and West and Post (2002) provide a global average of 0.57 ± 0. 14 metric tons C/ha/year by considering 67 long-term experiments around the world However, the rate of OC accumulation depends... dividing point between large and small aggregates, and between the physically protected and passive organic matter pools, is most likely in the 5 0- to 25 0- m size range The total OC content of soils increases with the ability to form aggregates, and hence with increasing clay content Figure 17.1 shows how tillage and soil texture relate to soil aggregation and carbon sequestration Sandy soils with little . pools based on rates of decom- position: microbial, slow, and stable, with default fractions for grassland and previously-cultivated soils of 02: 64: 34 and 02: 54: 44, respectively. Since we had. understand and predict SOC and food production responses to soil, climate, and management com- binations. Biophysical models integrate crop, soil, weather, and management practice information and. Soils Soil Depth (cm) SOC (%) Sand (%) Silt (%) Clay (%) pH Wilting Point a (cm 3 cm −3 ) Field Capacity b (cm 3 cm −3 ) Bulk Density (g cm −3 ) 0–20 0. 24 a 72 .4 21 .4 6.2 5. 34 0.069 0.176 1 .44 20 40 0.22 52.9 25.3 21.8 4. 93