Evaluating regional climate model simulations at Wadi El Abid catchment (northeastern Tunisia) using HBV rainfall-runoff model.

This study aims to assess climate model performance at Wadi El Abid basin (northeastern Tunisia), using Hydrologiska Byråns Vattenbalansavdelning (HBV)-light rainfall-runoff model (RRM). We evaluate time series of daily precipitation and mean daily temperature simulated by eleven couple of regional climate models (RCMs) downscaling five general circulation models (GCMs), provided by the Coordinated Downscaling Experiment–European Domain (EURO-CORDEX), over 1970–2000 historical period. The hydrological performance of regional climate models forced by Global Circulation Models (GCM-RCMs) is achieved using six discharge metrics, evaluating the runoff simulated by RRM forced by climate models' data. Quantile Delta Mapping bias correction technique was used to correct the distribution of precipitation and temperature time series simulated by GCM-RCMs. HBV-light model performed highly over the historical period with Kling-Gupta Efficiency around 0.80 and bias almost null. Quantile Delta Mapping (QDM) reduces greatly the bias in precipitation simulated by GCM-RCMs; nevertheless, disagreement between observed and bias-corrected precipitation is still present at monthly time scale after the application of QDM. This disagreement was transmitted to simulated runoff with further accentuation due to the high elasticity of runoff to precipitation. This finding shows the capacity of the hydrological evaluation of the climate model to highlight the hydrological consequences of certain weaknesses of the GCM-RCM simulations that are not well detectable by the separate evaluation of the climate variables. The hydrological rating of GCM-RCMs shows that IPSL-IPSL-CM5A-MR-IPSL-INERIS-WRF331F was low performing even though it was after bias correction. It is recommended to a modeler, to build their impact studies over the study catchment on the ensemble projection from the remaining ten best-performing GCM-RCMs. [ABSTRACT FROM AUTHOR]