Assessing the Applicability of Variable Infiltration Capacity (VIC) Model using Remote Sensing Products for the Analysis of Water Balance: Case Study of the Tons River Basin, India

Assessment of accurate water balance components depends on a comprehensive performance evaluation of the hydrological model. In this study, the variable infiltration capacity (VIC) model was evaluated (calibration: 1992-1996; validation: 1997-1999) using ground-based observed and remote sensing-derived datasets, and the evaluated model was used for long-term simulations (1990-2016) to assess the spatial-temporal distributions of water balance components for the Tons river basin (drainage area of 17,440 km2) in India. A good agreement was found between simulated and observed discharge on a daily scale (coefficient of determination, R2 estimated as 0.74 and 0.75 during calibration and validation, respectively) and on a monthly scale (R2 estimated as 0.90 and 0.91 during calibration and validation, respectively). To substantiate the model prediction reliability, the simulated evapotranspiration (ET) and soil moisture were validated with the Moderate Resolution Imaging Spectroradiometer (MODIS-MOD16A2) and Advanced SCATterometer (ASCAT) datasets, respectively. A good match was found between VIC-simulated ET and MODIS-derived ET (R2 = 0.78), and VIC-simulated soil moisture and ASCAT-derived soil moisture (R2 = 0.60). The long-term simulations were run to estimate the average annual water balance components (rainfall, ET, runoff, and baseflow) and their monthly spatial variations. ET was found to be predominant in the basin, accounting for about 59% of annual precipitation (average annual ET was found to be 579.62 mm). Overall, quantifications of hydrological water balance components and their spatial-temporal variations could help decision-makers to effectively plan and manage the river basin.