Prototype geographic information system for agricultural water quality management

Abstract

A prototype raster geographic information system (GIS) for agricultural water quality analysis was developed considering the farm as an aggregation of spatial units with homogeneous physical and management characteristics. A crop model that simulates the farm and environment response to different management scenarios was integrated with the GIS. The integrated GIS-model is then run on each homogeneous area. The results of crop yield and chemical leaching are geographically referenced for further display and analysis, and to serve as an input to the decision model. A decision model based on maximization of expected utility (MEU) was also integrated to help assess and evaluate the impacts of fertilizer application on the faun system and the environment. By using utilities for both crop yield and chemical leaching the model circumvents the issue of assigning a monetary value to the environment. Accommodating both the farmers' goals, in terms of higher yield and the well being of the environment, in terms of lower chemical leaching, the model computes the expected utility of each management scenario. The management practice with the maximum expected utility is then recommended. The integrated model was tested with an example of lettuce production in Arizona. Results were compared to published field reports, the model recommendation matched well with the field results. The prototype model was simple to use, and very well integrated, which makes it an alternative to the more complex and expensive coupling of commercial GIS and simulation models.