PERUN/Reanalys – the high resolution reanalysis for central Europe driven by ERA 5

PERUN/Reanalysis (PERRea) is a product based on the ALADIN system, which shares its model code with the global NWP systems ARPEGE/IFS. It is the first step towards the regional climate simulations within the CMIP6 framework.   

PERRea is being calculated for the period 1989–2019 on the domain covering Central Europe and its surroundings, with the horizontal mesh size 2.3km and 87 vertical levels. It provides not only the consistent 3D information about regional weather and climate on hourly footing, but its verification against the measurements and the driving ERA5 reanalysis quantifies accuracy of underlying ALADIN configuration, intended for simulations of future climate.

In order to guarantee reliable operation at convection permitting scales, PERRea employs ALADIN non-hydrostatic dynamical kernel combined with the so-called ALARO-1 physics. Its main highlights are: 1) self-switching-off convection parameterization with microphysical calculations sandwiched between updraft and downdraft computation, seeing joint output from subgrid and resolved condensation, 2) turbulence scheme incorporating the moist effects and using two prognostic energies, and 3) broadband radiation scheme enabling full cloud-radiation interaction. As for data assimilation component, PERRea analyses the soil properties using the optimal interpolation of SYNOP screen level temperature and humidity measurements. The upper air analysis combines the large scales from driving ERA5 reanalysis with the small scales from the previous high-resolution short range forecast. Their optimal combination is ensured via digital filter blending technique.

In this contribution, comparison of the PERRea with different reference datasets is presented for the period 1990–1999 over the full domain. Daily, monthly and seasonal precipitation, 2m temperature, 10m wind speed and surface solar radiation were selected for the comparison. Validation sets include ERA5 (the driving reanalysis of PERRea), station data from ECA&D dataset and E-OBS v23.0 (gridded station data over Europe) in 0.1° resolution.

The preliminary results show generally good agreement between PERRea and the two reference datasets, ERA5 and E-OBS. A weak systematic overestimation of precipitation in PERRea in selected regions with respect to ERA5 may indicate a difficulties in simulation of dry spells. The comparison of 2m temperature in PERRea against E-OBS shows insufficient diurnal amplitude especially in summer; whilst maximum daily temperature is underestimated in Central Europe, the minimum temperature is overestimated by more than 3 °C.

This research is supported by the Technology Agency of the Czech Republic Czech Technology Agency, under the SS02030040 project.