Dynamic Downscaling & Hydrologic Modelling of the Ganges Basin to Assess its Response to the Climate Change

Abstract:

Climate variability greatly alters the hydrologic regime but its impact analysis is hampered by disparity in the resolutions of GCMs & meteorological forcings required for catchment-scale simulation as the GCM outputs are too coarse to capture the local climate. Hence, dynamic downscaling was employed to represent the regional heterogeneity. Daily climatic data from HadCM3 at 2.5°x3.75° was downscaled to generate 3-hourly outputs at 0.25° for the Ganges basin using Advanced WRF model. Domain specific geographic data (SRTM 30m DEM, NRSC LU/LC & NBSSLUP soil) for the Indian region was ingested into the model, along with the global landuse & terrestrial data (GTOPO 1km DEM, USGS LU/LC etc.) for other regions, to aid regional meteorological prediction. Appropriate physics options were applied to mimic the interactions between atmosphere & land surface processes. The resultant downscaled precipitation & temperature products were calibrated & bias corrected using NCEP Reanalysis data & IMD gridded dataset. The model was able to reconstruct the orographic rainfall & extreme precipitation events that occurred in different years over Leh, Rudraprayag, Uttarkashi, Manali, Kedarnath, etc. The model was also compiled with varying spatial, temporal & hardware configurations to optimize the computational cost. A physically based semi distributed VIC model was forced using the downscaled product, & calibrated-validated using observed flow data to assess the impact of climate change on the hydrologic responses of the basin. The results suggested an early onset of monsoon with heavy rainfall in the early months of monsoon. While the number of rainy days in monsoon season decreased, intensity of rainfall increased. Subsequently, the runoff intensity increased with an overall decrease in total runoff in the future scenario. This study demonstrated the efficacy of dynamic downscaling & hydrologic models for climate change studies.