Paper No. 19-11 POTENTIAL DELINEATION OF GROUNDWATER STORAGE USING SATELLITE-BASED VEGETATION INDEX OVER PARTS OF THE INDIAN SUBCONTINENT



Paper No. 19-11
Presentation Time: 10:45 AM

POTENTIAL DELINEATION OF GROUNDWATER STORAGE USING SATELLITE-BASED VEGETATION INDEX OVER PARTS OF THE INDIAN SUBCONTINENT


PAPER WITHDRAWN
Potentiality of using satellite-based vegetation cover as indicator of groundwater storage (GWS) is studied for the first time over the parts of the densely populated Indian subcontinent where groundwater availability is a critical issue for human sustenance. Satellite derived Normalized Difference Vegetation Index (NDVI) is efficiently used to represent vegetation cover, globally. We used NDVI data files from the database of Moderate Resolution Imaging Spectroradiometer (MODIS). We also acquired water level data from a dense network (n> 13,000) of ground based monitoring locations and calculated GWS change (ΔGWSobs) using screened and culled water level change and aquifer specific yield information. Liquid water equivalent thickness (LWET) data files of the Gravity Recovery and Climate Experiment (GRACE) program from the National Aeronautics and Space Administration’s (NASA) Jet Propulsion Laboratory (JPL) archive were also used here for comparison. Simulated GWS change (ΔGWSsim) was calculated using LWET, and soil moisture, surface runoff and snow cover data. We observed high correlation (r > 0.6) for NDVI with ΔGWSobs and ΔGWSsim in areas of dense natural vegetation cover in the central study region. However, moderate to low correlation (r < 0.4) was found between ΔGWS and NDVI in areas with sparse vegetation in northern and southern study regions. Correlation analyses indicate lower values in the areas subjected to human interference in the form of intense irrigational groundwater withdrawal in northern and eastern study region. Trees in the forest area have widespread root systems, thus they are directly depended on GWS fluctuations. On the other hand, growth of grass or shrubs is not directly linked with GWS fluctuations owing to their shallow root zones. Consequently, correlation between soil moisture and NDVI exhibit higher values (r > 0.6) in areas of sparse vegetation. Therefore, our observation suggests that vegetation cover can be used as a proxy of groundwater storage in the naturally vegetated areas particularly forests.