PREDICTION OF PROCESSING TOMATO YIELD USING A CROP GROWTH MODEL AND REMOTELY SENSED AERIAL IMAGES.

Remote sensing using aerial images is very useful in obtaining large amounts of data in a short period of time. Vegetation indices derived from remotely sensed images often correlate well with plant characteristics. In our previous research, we found a relationship between modified normalized difference vegetation index (NDVI) and leaf area index (LAI). In this study, we have developed a model to predict the processing tomato yield based on soil, crop, and environmental parameters. LAI derived from modified NDVI and photosynthetically active radiation (PAR) are the two major inputs to this yield prediction model. The model was calibrated and validated using yield data gathered from a load cell based tomato yield monitor during the 2000 crop growing season. Although the actual and predicted yield maps did not have a very high correlation, the two maps showed similar yield patterns. The RMS error in yield prediction was about 6%. [ABSTRACT FROM AUTHOR]