Estimation of land-cover linkage to trends in hydrological variables of river basins in the Indian sub-continent using satellite observation and model outputs

The Indian subcontinent suffers from a decline in the per capita of water resources in the course of recent decades due to exponential population growth. In India, the majority of the population is reliant on agribusiness which is in turn primarily dependent on water from monsoons. Regions that get lower precipitation than needed experience water deficits which impact agriculture. However, investigation of linkages between the trends in water balance components and land-cover distribution has not been performed. In our work, we focus on the water balance for the major river basins in India utilizing the following satellite and model-based datasets: Terrestrial Water Storage Anomaly (TWSA) from Gravity Recovery and Climate Experiment (GRACE), Precipitation (P) from Tropical Rainfall Measuring Mission (TRMM), Evapotranspiration (ET) from the Moderate Resolution Imaging Spectroradiometer (MODIS), and Total Runoff (surface overflow and baseflow) (R) from the NASA Global Land Data Assimilation System (GLDAS). We assessed the seasonal spatio-temporal changes in the water balance from 2002 to 2019 and observed that `Agriculture' and `Urban' are the two most disrupted land-cover types displaying a monotonic increasing/decreasing trend in the components of water balance. We also observed from the SVD (Singular Value Decomposition) analysis that the inherent spatial variability between P-ET-R and TWSA do not correlate well. We also found that North-Eastern India and regions in Southern India along the west coast show large negative trends of P (-15 to -20 mm) and R (-10 to -15 mm) in the monsoon season, indicating that these regions suffer from drier monsoon seasons over the study period (2002-2019). From the land-cover linkages, it was found that 74% of monotonic trends observed in several of river basins were linked to 'Agricultural' land cover type and 19% were linked to `Urban' land cover type. These linkages suggest that agricultural lands are more vulnerable to changes in the components of the water balance.