A framework to identify droughts using temporal satellite data - a study of upper brantas watershed in east java of Indonesia

Abstract

dentifying drought-affected regions is usually done using ground based data collection such as rainfall or field V\veys. However in data scarce regions or of large spatial extent, such assessment is difficult and time consuming. Remote using data provides opportunities to assess the status of vegetation and these can be effectively used to assess drought conditions. NDVI values computed with NOAA AVHRR data are shown to reflect the drought condition. Vegetation response to rainfall indicated an approximately one-month time lag. In this study a comparison of a drought year and an adjacent year which did not experience a drought enabled the identification of a framework for drought monitoring using remote sensing data. In this study a NDVI value of less than 0.3 was recognized as an appropriate threshold for the identification of drought affected areas. Landsat TM data was used for land cover classification.