Assessment of agricultural drought using MODIS derived normalized difference water index

Abstract

Remote sensing index NDVI or its derivatives are used for agricultural drought monitoring and early warning at regional scale worldwide. Studies have shown that NDVI has lagged response to rainfall deficit. Moreover the red band used in NDVI is highly absorbed by crop canopy in comparison to short infrared which has high penetration so thus there remains a discrepancy between the levels of penetration in crop canopy. In contrast, Normalized Difference Water Index (NDWI) uses both the bands in near infrared region and is very sensitive to liquid water content of vegetation canopy and so rainfall. So this study was conducted to evaluate the sensitivity of NDWI in detecting and monitoring the agricultural drought in comparison with NDVI. In the study three indices of NDVI, NDWI5 and NDWI6 were computed using MODIS 09A1 surface reflectance product from June to October of 2002 (drought year) and 2003 (normal year) for the state of Rajasthan. NOAA Climate Prediction Centre (CPC) rainfall product was used and averaged at district level. The NDWI5 showed very strong relation with current rainfall than NDWI6 and weakest was shown by NDVI. The relation of NDVI with lagged rainfall was much better than with current rainfall. The spatial comparison of changes in NDVI and NDWI5 between the drought year (2002) and normal year (2003) for each 8 days composite showed that NDWI5 very well picks up the intensity and extent of drought. Study also showed that NDWI5 is more sensitive to agricultural drought than NDWI6. The study recommends use of NDWI5 for better early detection and monitoring of agricultural drought in operational drought management programmes.