Satellite Based Data Mining to Support Egypt's Agriculture

Agriculture in developing countries might benefit from advanced satellite based data mining approaches. The objective of the current study was to evaluate the added value of highabove coarseresolution satellite imagery in crop yield forecasting. This study focused on a coarse-resolution pixel in the Nile Delta in Egypt. Within this coarse-resolution pixel, 256 high-resolution (15 m) ASTER pixels are present, with wheat and berseem being the main crops. A crop-water model was used to simulate crop yields for the coarse-resolution pixel and for each of the high-resolution pixels. The model was driven by remotely sensed LAIs; one time-series for the coarse-resolution run, and 256 time-series for the high-resolution runs. The model was calibrated with SEBAL retrieved ETa. It was concluded that with the use of coarseresolution remote sensing, yields were overestimated between 9-26%, while high-resolution remote sensing resulted in errors below 3%.