RELATING MODIS NDVI AND EVI TO RICE YIELDS OF TERAI DISTRICTS OF NEPAL

Rice is an important crop in Nepal. Despite being an important crop, several challenges have led to the increasing rice imports in the country. A constraint still remains in the proper rice yield estimation mechanisms still unapplied. Remote sensing technologies have been successfully deployed in the yield prediction where the satellite collected spectral information is used to derive the vegetative indices. Such vegetative indices are empirically related to the yield. This research correlates the vegetation indices (NDVI and EVI) derived from the MODIS satellite data with the district wise rice yield of Terai region of Nepal. The utilisation of MODIS cropland cover as rice masks also presents its usability when long term rice mapping is unavailable. The results showed a significant positive correlation with rice yield and vegetation indices in almost all districts, except some districts showing no such positive association. The applied method of computing vegetation indices can be applied in future rice yield prediction. With the growing applications of satellite based remote sensing, this research presents an example of rice remote sensing in Nepal.