Decisions based solely upon mean climatic data are likely to be of limited use for at least two reasons. The first is concerned with definition of success and the second with averaging and time scale. In planning and analyzing agricultural systems it is essential not only to consider variability, but also to think of it in terms directly relevant to components of the system. Such analyses may be relatively straightforward probabilistic analyses of particular events, such as the start of cropping seasons in West Africa and India. The principal effects of weather on crop growth and development are well understood and are predictable. Crop simulation models can predict responses to large variations in weather. At every point of application weather data are the most important input. The main goal of most applications of crop models is to predict commercial out-put (Grain yield, fruits, root, biomass for fodder etc.). In general the management applications of crop simulation models can be defined as: 1) strategic applications (crop models are run prior to planting), 2) practical applications (crop models are run prior to and during crop growth) and 3) forecasting applications (models are run to predict yield both prior to and during crop growth).
Crop simulation models are used in USA and in Europe by farmers, private agencies, and policy makers to a greater extent for decision making. Under Indian and African climatic conditions these applications have an excellent role to play. The reasons being the dependence on monsoon rains for all agricultural operations in India and the frequent dry spells and scanty rainfall in crop growing areas in Africa. Once the arrival of monsoon is delayed the policy makers and agricultural scientists in India are under tremendous pressure. They need to go for contingency plans. These models enable to evaluate alternative management strategies, quickly, effectively and at no/low cost. To account for the interaction of the management scenarios with weather conditions and the risk associated with unpredictable weather, the simulations are conducted for at least 20-30 different weather seasons or weather years. If available, the historical weather data, and if not weather generators are used presently. The assumption is that these historical data will represent the variability of the weather conditions in future. Weather also plays a key role as input for long-term crop rotation and crop sequencing simulations.
Was this article helpful?