Modelling the Impact of Climate Change and Associated Uncertainties on Future Water Availability of a Drought Prone Watershed in Central India

Climate change (CC) is a major concern for future water availability and agricultural water management. Accurate prediction of impact of CC at different spatial and temporal scales has engaged researchers for long. However, CC impact assessment at watershed scale is limited due to uncertainties associated with different resolution input datasets. The goal of this research is to investigate different sources of uncertainties associated with modelling the impact of CC on water availability at a watershed-scale in Central India using Soil and Water Assessment Tool (SWAT) model. Towards this goal, 25 Coupled Model Intercomparison Project Phase 6 (CMIP6) Global Climate models (GCMs) were screened to discard poor performing models based on a custom developed criterion. The screened models were downscaled using three statistical downscaling techniques for minimum temperature, maximum temperature, and precipitation followed by identification of uncertainties associated with these variables. The downscaled variables were used as an input in the SWAT model to evaluate impact of uncertainties in future runoff projection. In addition, the impact of CC on surface runoff was studied under two global warming scenarios - Shared Socio-Economic Pathways (SSP245 and SSP585). Based on the screening 8 CMIP6 models - CanESM5, ACCESS-CM2, ACCESS-ESM5, INMCM5-0, UKESM-1-0-11, CNRM-CM6-HR, MPI-ESM-LR and CMCC-CM2-SR5 were discarded. Results suggest that uncertainty associated with GCM variables for the period 2015-2100 ranged from -2oC to 4oC, -2oC to 6oC, and -10 to 30% for daily maximum temperature, minimum temperature, and precipitation, respectively. Results of uncertainty associated with future water availability over the watershed under changing climate will be presented. This study will provide an integrated modelling approach using GCMs and SWAT model for assessing potential impacts of CC at watershed scale.