A New Model of Spatial Prediction on Drought Prone Risk Areas

This research aims to identify and predict Central Java areas that are potentially droughts. It uses data extraction from the Normalized Difference Vegetation Index) (NDVI), Normalized Difference Water Index (NDWI), Difference Vegetation Index (DVI), and Difference Water Index (DWI) on Landsat 8 OLI image year 2015-2017. NDVI and DVI show the vegetation level, while NDWI and DWI show the humidity level. Drought potential areas prediction is made by kriging interpolation. This research shows that the highest drought occurred on July 16, 2015, and August 22, 2017. The kriging test on January 10, 2017, shows a match rate of 76.75% on Magelang, Purworejo, and Semarang Regencies. The testing on the kriging result for August 22, 2017, shows a match rate of 64.25% on Magelang, Purworejo Regencies, and part of Semarang Regency.