Biophysical agricultural assessment and management models for developing countries

Scientists in the developed world, in an effort to develop tools that will support informed land use decision making, have created a number of very different land use analysis and agricultural production computer models. Such models should be increasingly useful in many parts of Latin America, where issues of sustainable land use are approaching critical dimensions. These include the need to maximize agricultural production while minimizing erosion, to use limited land areas in the most efficient manner, to reduce pollution pressure on the landscape, and to reduce deforestation—as well as optimize these goals. The models reviewed in this chapter are derived quite independently from one another and for very different purposes: ALES (Automated Land Evaluation System) is a computerized framework for allowing expert judgment to estimate crop and consequent economic production, and DSSAT (Decision Support System for Agrotechnology Transfer) is an explicit suite of models of crop plant production. The development of an agroecological modeling approach for land use planning in Costa Rica faced two major problems: a dispersed and not very accurate database and an official land use classification scheme that had little previous impact on decision making. There is an inherent social difficulty in applying even perfectly accurate models, especially where models show that a certain agricultural use might not be a good use for a particular land.