Bias Correction of GCM Data Using Quantile Mapping Technique

Climate scenarios generated from GCMs taken as input to various climate change impact assessment studies. These GCMS data have lots of bias which may result in uncertain predictions of various impact studies. So, it is necessary to provide bias correct data as input to reduce the uncertainties in these studies. Therefore, this study is compared with two methods of quantile mapping bias correction technique: linear and tricube for Delhi location over 30 years’ (1976–2005) period for temperature and rainfall. GCM(CSIRO-MK3) model simulated climate data were bias corrected using both approaches. These bias-corrected data were compared yearly, and also, its parameterwise statistics (RMSE) were calculated. Based on the analysis, this study indicates that both the methods (linear and tricube) have the ability to remove the bias from climate data. But tricube method performed better among both the methods. Tricube method along with the wet_false combination can remove bias more near to the observed data as compared to combining with wet_true condition.