Climate change signals based on different bias-corrected EURO-CORDEX simulations over Hungary

Climate change is one of the major challenges that humanity faces today. The latest IPCC report highlights that an increase in the frequency and intensity of extreme weather and climate events are likely to occur in the future. Global and regional climate models (GCMs and RCMs, respectively) are key tools for climate research which provide valuable information about future climate change. However, model simulations are subjects to uncertainties of various origins. In order to quantify and reduce these uncertainties it is recommended to evaluate several RCM simulations together, as members of an ensemble, and/or apply a bias-correction method. The implementation of a bias-correction procedure has become a standard step in the process of using climate model outputs, since unbiased data are required for impact studies. During this procedure, the raw simulated meteorological variables are adjusted to measurements, so the use of a high-quality, observation-based dataset as a reference is a crucial point for all bias-correction methods, in addition, the choice of the calibration period is also an important factor.

In our research an ensemble of 5 RCMs (CCLM, HIRHAM, RACMO, RCA, REMO) driven by 4 different GCMs are investigated from the framework of EURO-CORDEX with the finer (0.11°) horizontal resolution for the historical (1976–2005) and the scenario (2006–2099) time periods under two radiative forcing scenarios (RCP4.5 and RCP8.5). Our aim is to investigate how the choice of the reference dataset and different calibration periods affects the results of the expected changes based on the different bias-corrected datasets, focusing on Hungary. For this purpose, a bias-adjustment was carried out by applying the internationally widely used percentile-based quantile mapping method on a monthly level for 4 variables: daily minimum- and maximum temperature, mean temperature and precipitation. The most accurate, measurement-based and quality controlled HuClim dataset was used as a reference. Two calibration periods were chosen from HuClim for the bias-correction: an earlier (1976–2005) and a more recent (1993–2022) one, thus creating two, different bias-corrected databases. Beside them, a third bias-corrected database was also analyzed, namely, the same RCMs provided by the EURO-CORDEX community and adjusted using the MESAN reanalysis data based on the 1989–2010 calibration period. In the present study, we analyze and show the results of the different bias-adjusted RCM simulations for two future periods (2021–2050; 2070–2099) focusing on the detected change of climate indices.