ABSTRACT

The ability to predict yield in advance of harvest would be of great benefit to producers in terms of risk management and overall profitability. Crop growth models have proven to be useful, but they are not suited to use on a site-specific basis. Remote sensing provides site-specific data, but alone it is lacking in utility for producers. However, together these two technologies can conceivably provide for site-specific prediction of yield. A computer interface between the Arcview GIS (geographic information system) and the cotton growth model, Gossym, was developed to allow the model to accept inputs and provide predictive outputs on a site-specific basis. The relationship between remote-sensing data and cotton plant height was studied, and a means to input remote-sensing-based plant-height estimates to Gossym was implemented in the GIS interface. Based on this research, it appears to be reasonable to use remote-sensing data to estimate plant height, and then to use the estimate as real-world feedback to improve Gossym’s yield predictions. The results of this work facilitate the entire process of combining remote-sensing data with the cotton growth model, Gossym, in a GIS framework.