资源环境科技发展态势分析平台

With the objective to provide a basis for regional climate models (RCMs) selection and ensemble generation for climate impact assessments, we perform the first ever analysis of climate projections for Western Nepal from 19 RCMs in the Coordinated Regional Downscaling Experiment for South Asia (CORDEX-SA). Using the climate futures (CF) framework, projected changes in annual total precipitation and average minimum/maximum temperature from the RCMs are classified into 18 CF matrices for two representative concentration pathways (RCPs: 4.5/8.5), three future time frames (2021-2045/2046-2070/2071-2095), three geographic regions (mountains/hills/plains) and three representative CF (low-risk/consensus/high-risk). Ten plausible CF scenario ensembles were identified to assess future water availability in Karnali basin, the headwaters of the Ganges. Comparison of projections for the three regions with literature shows that spatial disaggregation possible using RCMs is important, as local values are often higher with higher variability than values for South Asia. Characterization of future climate using raw and bias-corrected data shows that RCM projections vary most between mountain and Tarai plains with increasing divergence for higher future and RCPs. Warmer temperatures, prolonged monsoon and sporadic rain events even in drier months are likely across all regions. Highest fluctuations in precipitation are projected for the hills and plains while highest changes in temperature are projected for the mountains. Trends in change in annual average discharge for the scenarios vary across the basin with both precipitation and temperature change influencing the hydrological cycle. CF matrices provide an accessible and simplified basis to systematically generate application-specific plausible climate scenario ensembles from all available RCMs for a rigorous impact assessment.