No abstract

oped and developing countries (Algozin et al., 1988; Bowen and Wilkens, 1998; Jagtap et al., 1993;Lal et al., In low-input systems, where most nutrients become available from Singh et al., 1993;Thornton and Wilkens, 1998). soil organic matter (SOM) and residue turnover, the applicability of DSSAT (Decision Support System for Agrotechnology Transfer) crop DSSAT version 3.5 incorporates 16 crops {maize (Zea simulation models is limited because (i) it recognizes only one type mays L.), wheat (Triticum aestivum L.), rice (Oryza of SOM (i.e., humus) and recently added, but not yet humified, resisativa L.), sorghum [Sorghum bicolor (L.) Moench], mildue; (ii) it does not recognize a residue layer on top of the soil; (iii) let [Pennisetum typhoides (Burm.) Stapf & Hubb.], barnewly formed humus is given a fixed C/N ratio of 10; (iv) only one ley (Hordeum vulgare L.), bean (Phaseolus vulgaris L.), litter pool is recognized for N although three are recognized for C; soybean [Glycine max (L.) Merr.], peanut (Arachis hy-(v) for residues with C/N ratios Ͻ25, the three litter pools for C pogaea L.), chickpea (Cicer arietinum L.), cassava (Mandecompose at a rate that is independent of the residue's N concentraihot esculenta Crantz), potato (Solanum tuberosum L.), tion; and (vi) SOM and residue flows are independent of soil texture. sugarcane (Saccharum officinarum L.), tomato [Lyco-A SOM-residue module from the CENTURY model was incorpoersicon lycopersicum (L.) Karsten], bahiagrass (Pasrated in the DSSAT crop simulation models, and a residue layer was added on top of the soil. Modifications were also made in the palum notatum Fluegge), and sunflower (Helianthus ansenescence module of CROPGRO, a model within DSSAT, so that nuus L.)}, with several more under development. The senesced material is now added daily to the soil. Evaluation of the model handles management strategies that involve crop model, using a data set of 40 yr of bare fallow, showed an excellent rotations, irrigation, fertilization, and organic applicafit [product moment correlation coefficient (r ) of 0.983] between tions. Although crops (or cultivars) and crop managesimulated and measured values for SOM-C. Soil N from decomposing ment (e.g., mechanization) may differ from country to SOM and residues was evaluated with data from a Brazilian expericountry or even from village to village, the effect of ment with seven leguminous residue types. By incorporating the CENfertilizer or irrigation on crop production is likely to TURY SOM-residue module, DSSAT crop simulation models have follow similar biophysical and biochemical pathways. become more suitable for simulating low-input systems and conduct-However, there is an important difference between ing long-term sustainability analyses.

No abstract

Cropping systems models have evolved over the last four decades in response to the demand for modeling to address more complex questions, including issues on sustainable production, climate change, and environmental impacts. Early models, which were used primarily for yield gap analysis, have increased in complexity to include not only nutrient and water deficiencies, but also pest and disease damage and processes affecting soil nutrient dynamics. This is the case in the Cropping System Model (CSM) within Decision Support System for Agrotechnology Transfer (DSSAT). This package was developed from various models of individual crops beginning about 25 years ago into one that now has over 25 crops integrated into one program that share many components in a modular format. This modular structure was intended to facilitate incorporation of new components to address those more complex issues. A recent example of this continuing progression is that the CENTURY soil organic matter model was adapted for the DSSAT-CSM modular format in order to better model the dynamics of soil organic nutrient processes. This capability is particularly important to enable CSM to be used for predicting yields in low input cropping systems where soils tend to be deficient in organic matter and nutrients. Organic matter processes are also critical when analyzing the dynamics of cropping systems over long periods of time such as for climate change scenarios. The addition of this more detailed organic matter module provided opportunities to also improve existing components of the model, including energy balance at the soil-plant-atmosphere interface and surface water runoff computations. Conversely, the more detailed organic matter module required additional inputs from existing model components, which were not previously used. Thus, addition of this one new model capability both required and allowed further modifications throughout CSM in order to improve model predictions. This paper provides a brief overview of the DSSAT-CSM model architecture and the DSSAT-CENTURY module and details the changes made to accommodate and take advantage of the more complex soil organic matter modeling capability.

oped and developing countries (Algozin et al., 1988; Bowen and Wilkens, 1998; Jagtap et al., 1993;Lal et al.,