Parameterization-based uncertainties in RegCM simulations over the Carpathian region in a wet year

Precipitation is one of the most important climate variables in many aspects due to its key impact on agriculture, water management, etc. Furthermore, extreme precipitation events can lead to excess surface water and floods and are becoming an amplifying societal cost as a result of urbanization and our warming climate. However, it remains a challenge for climate models to realistically simulate the regional patterns, temporal variations, and intensity of precipitation. Detailed knowledge about extreme precipitation events is important for advanced predictions on weather-to-climate time scales. The difficulty arises from the complexity of precipitation processes within the atmosphere stemming from cloud microphysics, cumulus convection, large-scale circulations, planetary boundary layer processes, and many others. This is especially true for heterogeneous surfaces with complex orography such as the Carpathian region.

In order to quantify the impact of the use of different parameterization schemes on regional climate model outputs, hindcast experiments were completed applying RegCM4.7 to the Carpathian region and its surroundings at 10-km horizontal resolution using ERA-Interim reanalysis data as initial and boundary conditions. In this study, 24 simulations were carried out by using various combinations of the physics schemes (2 land surface, 2 microphysics, 3 cumulus and 2 boundary layer schemes) for the year 2010, which was the wettest year in the region since the beginning of the regular measurements. Each parameterization combination leads to different simulated climates, so their spread is an estimate of the model uncertainty arising from the representation of the unresolved phenomena. The analysis of the RegCM ensemble indicates systematic precipitation biases, which are linked to different physical mechanisms in the summer and winter seasons.

Based on the results, RegCM is sensitive to the applied convection scheme, but the interactions with the other schemes (e.g., land surface or microphysics) affect not only the total precipitation, but also the convective and stratiform precipitation in some cases. Due to the different treatment of moisture in the schemes, there are differences not only between the representation of the precipitation cycle, but also in other climatological variables such as soil moisture, temperature and cloud cover.