Small farmers in developing countries, unlike their counterparts in the industrialized world, cannot afford to apply much chemical fertilizer. Instead, they use 'low-input' agricultural systems, which draw nutrients for crop growth mostly from the decomposition of soil organic matter and plant residues. In the 2002 May-June issue of the Agronomy Journal, scientists report the development of a computer-based simulation model that greatly simplify the evaluation of such systems in an article titled, 'Modifying DSSAT Crop Models for Low-Input Agricultural Systems Using a Soil Organic Matter-Residue Module from CENTURY.' This is good news for scientists working to improve low-input agriculture and thus increase food security, raise farm incomes, and halt the rampant decline of soil fertility in the tropics.
Field studies of low-input agriculture are being conducted at many locations in Africa, Latin America, and Asia, but these studies cannot cover all the possible combinations of soil, crops, weather, and management factors that merit evaluation. Simulation models, because they can include innumerable combinations, have powerful potential for aiding the evaluation of different low-input systems at specific sites.
Various agricultural simulation models are available, but they are tailored mainly to the requirements of industrialized-country agriculture. Scientists with the International Center for Tropical Agriculture (CIAT) in Colombia asked whether these same models could be applied to low-input systems. In seeking an answer, they applied a simulation model called the 'Decision Support System for Agrotechnology Transfer,' or DSSAT, to hillside agriculture in Honduras and Nicaragua. Developed by a consortium of North American universities and research institutes, DSSAT is the most widely used agricultural model in the world. But since it assumes heavy use of chemical fertilizers, the model's soil organic matter section is weak. To overcome this limitation, scientists linked DSSAT to CENTURY, one of the leading models for soil organic matter.
The combined DSSAT-CENTURY model was tested with 40 years of data from a typical field under low-input agriculture. The field had not been under mechanized cropping, and no fertilizer had been applied for a long time. "While the original DSSAT model did a poor job of simulating the strong decline of soil organic matter, said Arjan Gijsman, senior author of the paper, "the new DSSAT-CENTURY model performed very well in this respect. And that means it probably also simulated quite well the release of nutrients from the organic matter."
"As a result," add coauthors Gerrit Hoogenboom of the University of Georgia, William J. Parton of Colorado State University, and CIAT's Peter C. Kerridge, "the DSSAT-CENTURY should prove to be of great practical value for studying small-scale agriculture in developing countries. Moreover, the model's improved section on soil organic matter makes it more suitable for long-term simulations. For example, the model could be used to estimate carbon sequestration in soil organic matter under different systems, which is one approach scientists are exploring to mitigate the effects of global warming on agriculture."
The combined DSSAT-CENTURY model will be part of the new version of DSSAT, to be released later this year.
