Multi-model assessment of regional climate simulations over Pakistan using SMHI RCA4 downscaled global climate models under the CORDEX framework

Abstract

We present an evaluation and future projections of ten Global Climate Models (GCMs) dynamically downscaled using Rossby Centre Regional Atmospheric model version 4 (RCA4) under the Coordinated Regional Climate Downscaling Experiment (CORDEX) framework. Evaluations were done with Climate Research Unit (CRU) time-series (TS) CRU TS v4.09 datasets (baseline: 1975–2005) for temperature and precipitation to identify the most reliable models, in the construction of multi-model ensembles (MMEs) for robust future climate projections and risk assessment over Pakistan. Findings showed that all models produced 100% wet bias in eastern parts of HKH. Below HKH, CNRM produced the lowest dry bias while ICHEC produced the lowest wet bias. Highest model correlations and root mean square error (RMSE) for precipitation were found in eastern parts of Pakistan depicted by CNRM. CNRM, NCC and NOAA most realistically reproduced precipitation trends. Hotter temperature trends over south-western and north-east Pakistan were most realistically captured with ICHEC. CNRM is the only model which realistically captures the long-term spatial trends of both precipitation and temperature. Owing to this, multi-model ensembles (MMEs) were created as a holistic approach to study two Representative Concentration Pathways (RCPs) namely RCP 4.5 and 8.5 for three future time periods (F1:2016–2035, F2:2046–2065 and F3:2081–2100). MMEs showed a change of 6o C under RCP 8.5 over Pakistan while RCP 4.5 stabilized at 3o C by the end of the century. Future climate anomalies depict that HKH is experiencing an increase in temperature with a decline in precipitation indicating higher melting, while below HKH, a higher increase in precipitation than temperature is observed indicating increased future precipitation intensities.