The project with the aim to develop a regional climate model (RCM) for the territory of central Europe was launched in 2001 in the Czech Republic. The RCM is being developed by modifications of the spectral numerical weather prediction (NWP) model ALADIN, run operationally at the Regional Centre of LACE (Limited-Area Modelling in Central Europe) in Prague. It concentrates on the area of central Europe and is intended in close future to serve as a source of climate change scenarios on a regional and local scales for countries in that region, especially for the Czech Republic. The main feature of this RCM is its very fine horizontal resolution, first tests were done with 12 km, now due to technical reasons in 24 km, still being at the high end of current RCMs.

The ALADIN model proves to be integrable for longer time periods, far beyond its current operational use (up to 48 hours), after only minor modifications of rather a technical nature are implemented as well as the first attempts with modification of physical parameterizations. The presentation will summarize validation of recent experimental runs of the RCM, nested in the operational assimilations by the ARPEGE NWP global model, representing observed conditions, both for monthly experiments in summer and winter seasons and for one year simulations. The influence of a treatment of lower boundary conditions, interpolation at lateral boundaries, the effects of repeated restarts of the RCM as well as some sensitivity tests of parameterization are studied. The validation concerns (i) the upper-air fields, (ii) surface temperature and precipitation at the dense station network in the Czech Republic, as well as against continental-scale gridded climatologies, and (iii) vertical cross-sections.

During the development, supporting tests were completed with the RegCM2 and, finally, RegCM3 concerning the modelling sensitivity to geometry of the model area. Moreover, we use the RegCM to test the methodology for planning and organizing the experiments and for comparison with ALADIN. Finally, to understand the ability of RCM’s to capture the extremes in the distributions of temperature and precipitation 40 years experiment with RegCM3 driven by NCEP reanalysis is compared to observational data.