Meteorologische Zeitschrift (Jun 2024)
Towards a spatial snow climatology for Switzerland: Comparison and validation of existing datasets
Abstract
In the European Alps, surface snow cover is of high relevance and a major factor for environmental, ecological and economical systems. The provision of accurate and timely climatological information on the spatio-temporal distribution of Alpine snow cover is hence an important yet challenging task of modern climate services. To assess the quality and the reliability of existing snow cover products in this region, we here compare daily snow cover datasets over Switzerland from the low-resolution ERA5 reanalysis, two intermediate resolution reanalyses (ERA5-Land and COSMO-REA6) and a high-resolution AVHRR-derived remote sensing product with three high-resolution 1 km offline snow models for the past 30 to 60 years. We focus on the parameters snow water equivalent (SWE) and daily snow coverage (snow/no snow). In the challenging Alpine terrain, all datasets are able to broadly represent the mean and the seasonal cycle of SWE and the number of snow days. They also agree on the direction of the trends. Decreasing mean SWE and snow day trends are found for the whole of Switzerland and all subregions in the period 1982–2019. Regarding interannual variability and trends, ERA5, ERA5-Land and COSMO-REA6 seem to perform reasonably well in most regions although the absolute and relative biases can be considerable. Especially ERA5-Land strongly overestimates SWE at high elevations. The biases are often larger in the southern Alps than in and north of the Alps. Most datasets, including the high-resolution ones, have problems correctly representing small SWE values at low elevations. The results indicate that in order to estimate absolute SWE or snow days with a reasonable accuracy, or to conduct a detailed study of the elevation dependence of snow, a km‑scale model that assimilates snow measurements is highly preferable. MeteoSwiss and the WSL Institute for Snow and Avalanche Research SLF are currently setting up a 1 km daily operational climatological product covering the period from 1961 to today.
Keywords