On the need for bias-adjusted CORDEX data: present day and future temperature characteristics over the Carpathians at regional and local level

Present research focuses on temperature change signals over the Carpathian Basin with a special focus on selected lowland and mountainous sub-regions. Under the framework of the international initiative called COordinated Regional Downscaling EXperiment (CORDEX): EURO-CORDEX and Med-CORDEX provide regional climate model (RCM) simulations targeting Europe. Simulation of near-surface air temperature from a mini-ensemble of high-resolution (0.11°) EURO and Med-CORDEX RCM climate change experiments are analyzed based on raw and bias-adjusted data. The mini-ensemble consists of 8 RCM simulations driven by 5 different general circulation models (GCMs) for the period 1976–2099 under the high-end RCP8.5 (representative concentration pathway) scenario. The high-resolution, homogenized and quality controlled CARPATCLIM was used as a reference dataset during the bias-adjustment procedure. The selected sub-regions cover 8 municipalities located at diverse altitudes: Bratislava, Budapest, Brassov, Debrecen, Hoverla, Novi Sad, Pécs and Poprad. The following near-surface temperature related climate indices assessed over the region of interest: summer days (SU), ice days (ID), frost days (FD), tropical nights (TR), the coldest day (TXn), the warmest day (TXx), the coldest night (TNn) and the warmest night (TNx). In general, TXx, TNx and TNn were overestimated by the raw simulations by 1–4 °C in the central part of the region, but an underestimation was found in the case of TNn and TXn in the Carpathians for the reference period (1976–2005), whilst bias-adjusted RCM data showed almost perfect match with observations. Accordingly, no best performing RCM is found for all indices. The ensemble mean of the bias-adjusted RCM simulations project an increase (decrease) in the annual number of SU and TR (FD and ID) for the near future (2021–2050) and for the far future (2070–2099). Profound warming manifests in the increase of TXx of up to 2–3 °C by the near future and of 5–7 °C by the end of the 21st century. Our results also highlight the need on bias-adjusted data applied by different sectors (e.g. human health, agriculture, transport, tourism, disaster management or heritage conservation) in the context of national adaptation strategies.