Crop Models and Applications


Crop models calculate expected growth and development based on equations that describe how a crop, as community of plants, respond to soil and weather conditions. At their simplest level of interpretation, the equations used in a model are a set of differential equations representing rates of growth or development. Numerical integration over time, typically with daily or hourly time steps, allows estimation of growth, development, and water and nutrient levels. The equations are based on information from crop physiology, soil science, meteorology and other fields.

The models provided in DSSAT deal primarily with annual crops including wheat, rice, maize and various grain legumes but also include herbaceous perennials such as forage legumes and grasses. Besides crop growth and development, the models simulate water and nutrient dynamics in the soil and crop, so processes such as leaching, organic matter decomposition, and runoff are also considered. The level of process details varies greatly, and in many cases, users may select among model options, allowing the user to assess how different assumptions affect simulations.

Model applications range from real-time decision support for crop management to assessing the potential impact of climate change on global food security. Crop models are also invaluable as heuristic devices that help identify research problems where our current knowledge has limits and further research is needed. The ability of crop models to simulate how different weather years or soil conditions affect crop performance make models especially useful in research involving climatic uncertainty or geospatial variation. Recent advances in field phenomics and crop genomics are opening opportunities for crop models to support research in fundamental plant science.

Because the quality of simulation results depends heavily on the data inputs, DSSAT includes tools to assist modelers in organizing input data for crop management, soils and weather. An especially challenging set of inputs are the genotype-specific parameters (GSPs) used to quantify how one cultivar differs from another. GSPs are most often estimated through calibration to measurements from field trials, and DSSAT provides tools both to organize data used for calibration and to estimate required GSPs.