validation of global climate models (gcms) temperature and rainfall simulation in kermanshah, ravansar and west islamabad stations

the objective of this study was to evaluate the ability of ten gcms and three emission scenarios to reproduce the rainfall and temperature parameters in kermanshah, ravansar and west islamabad stations located inside qaresou basin in iran using the average weighting method, over 1989-2008 period. at first, the models output and the scenarios were provided for this period. then, the models validation was carried out using mean absolute error (mae), root mean square error (rmse), model bias, and nash – sutcliffe criteria. moreover, the models and scenarios uncertainty analysis was assessed by the weighting method. based on these criteria, hadgem1 and bcm2 had the best efficiency in kermanshah station for mean temperature and rainfall simulation respectively. for ravansar station the best models were echo-g and hadcm3, while for west islamabad was echo-g. the overall results of this analysis showed that the best applied gcms and emission scenarios for simulation of temperature and rainfall in the study area are echo-g (a2) and hadcm3 (a1b), respectively. the results also indicated that these models and scenarios performance for simulation of temperature and rainfall are varied and thus a gcm need to be calibrated for different regions.