Crop growth models for decision support systems

Studies on crop production are traditionally carried out by using conventional experience-based agronomic research, in which crop production functions were derived from statistical analysis without referring to the underlying biological or physical principles involved. The weaknesses and disadvantages of this approach and the need for greater in-depth analysis have long been recognized. Recently, application of the knowledge-based systems approach to agricultural management has been gaining popularity because of our expanding knowledge of processes that are involved in the growth of plants, coupled with the availability of inexpensive and powerful computers. The systems approach makes use of dynamic simulation models of crop growth and of cropping systems. In the most satisfactory crop growth models, current knowledge of plant growth and development from various disciplines, such as crop physiology, agrometeorology, soil science and agronomy, is integrated in a consistent, quantitative and process-oriented manner. After proper validation, the models are used to predict crop responses to different environments that are either the result of global change or induced by agricultural management and to test alternative crop management options. Computerized decision support systems for field-level crop management are now available. The decision support systems for agrotechnology transfer (DSSAT) allows users to combine the technical knowledge contained in crop growth models with economic considerations and environmental impact evaluations to facilitate economic analysis and risk assessment of farming enterprises. Thus, DSSAT is a valuable tool to aid the development of a viable and sustainable agricultural industry. The development and validation of crop models can improve our understanding of the underlying processes, pinpoint where our understanding is inadequate, and, hence, support strategic agricultural research. The knowledge-based systems approach offers great potential to expand our ability to make good agricultural management decisions, not only for the current climatic variability, but for the anticipated climatic changes of the future.