Assessment of Impact of Drought on Desert Vegetation through Remote Sensing

Abstract

Arid vegetation shows monsoonal growth. When rains fail in the event of drought, regenerated annuals and perennials turn dry. Since NDVI senses greenness of vegetation, such dry and pale vegetation in the desert needed to be assessed by some other way. We have therefore used another index, PD-54, based on cover of dried vegetation and compared its results with those from NDVI in real time in five grazinglands in Shergarh tehsil of Jodhpur district in Rajasthan. Satellite data of IRS 1C/1D/P6 with LISS 3 sensor for different seasons were acquired for three contrasting wet-dry season events. After calibrating these radiometrically and registering geometrically, index of vegetation cover PD54 as well as NDVI were calculated. PD54 is a perpendicular vegetation index based on the green and red spectral band width. Ground radiometric observations were also used to calculate the PD54 and NDVI and were related to vegetation cover measured on ground in permanent plots. This confirmed superiority of PD54 for estimating cover in arid dry grasslands. Sequential trends in cover of ground vegetation in a protected and nearby unprotected, open grazinglands from a good rainfall year to drought year were related with satellite data. A grass cover of 14.9% by wheel point method in the protected area (compared to zero per cent grass cover in unprotected site) in October drought event could not be detected by NDVI while PD 54 successfully captured it. Total vegetation cover of 17.4% as measured on ground (i.e., total of forbs, grass and woody perennials) in unprotected site after drought in October was assessed more by NDVI though it was much less than that at protected site (39.8%). Reverse was true for PD54. Thus drought impacted vegetation in arid grazinglands could be more accurately assessed through satellite data using PD54. Such estimate on real time scale helps to prepare plans for drought preparedness, pastoralism and transhumance. Key words: Grazingland, arid zone, spectral reflectance, vegetation index, degradation, NDVI.