Extreme precipitation events simulation under plausible scenarios of climate change in Satluj River basin, India

Abstract

A K-nearest neighbour weather-generating model is applied to simulate: 1) total precipitation during extreme precipitation events; 2) duration of extreme wet spells; 3) duration of extreme dry spells in the Satluj River basin - a key basin in the central Himalayas under climate change scenarios projected by 15 different combinations of global climate models (GCMs) and emission scenarios. The intent is to assess the vulnerability of the basin to extreme hydrological events, including floods and droughts, resulting from extreme precipitation events. The driving data set for the K-NN model was obtained by applying the precipitation change fields obtained from different combinations of the GCM and emission scenarios to the observed data. Results indicate that several extreme wet and dry spells that were more severe than the observed were simulated. Results of frequency analysis of storm depths indicated that ECHAM5 model under A2 scenario produced the highest one-day precipitation maxima for each return period considered.