High-resolution climate reanalyses datasets for hydro-climatic impact studies over Switzerland

To study and model hydro-climatic extremes (e.g., droughts and floods), high quality, multivariate and spatially consistent meteorological datasets are necessary. In-situ measurements, however, often don`t cover these multivariate data needs, and are variable in space and time. New generations of high-resolution reanalysis products, with continental to global scales, are available that offer a wide range of internally consistent land surface variables. However, it is yet unclear which of these datasets are most suitable for hydro-climatic impact studies, e.g., to assess the spatiotemporal connectedness of floods and droughts, or to quantify the multivariate meteorological drivers of these extremes. Here, we present a comparison of multiple high-resolution reanalysis datasets (i.e., ERA5(-land), CERRA(-land) and CHELSA-v2) with a gridded observational product from MeteoSwiss. We compare various climatological statistics of precipitation and temperature, such as differences in the mean and the tails of the distribution, as well as the consistency of temporal trends and interannual variability. We further analyze differences in selected univariate and multivariate climate indicators, such as the annual maximum 1–5-day precipitation, the number of dry/wet days, or the fraction of solid/liquid precipitation. Lastly, we present the spatiotemporal consistency of several observed hydro-climatological extreme events, including the drought in 2018 and the floods in 2005 over Switzerland.

For most of the reanalysis products the analysis shows a clear elevational dependence in the biases, i.e., increasing with elevation, compared to the gridded observational dataset. The regional reanalysis product CERRA(-land) can overall reduce the biases in the general climatological statistics (e.g., means, tails of the distribution), but shows inconsistencies when moving to the event scale. It for example shows inconsistencies in the temporal evolution and severity of the 2018 drought in Switzerland, whereas the other reanalysis products are more consistent. This presentation gives a comprehensive overview of the differences in the current state-of-the-art reanalysis datasets over the complex terrain of Switzerland.