A multi-dataset trend analysis of precipitation characteristics over Europe

Understanding long-term changes in precipitation is essential for climate research. However, it is a well-established fact that various types of data (station, gridded, reanalysis) can yield different results in extreme and trend statistical analyses. In this contribution, we investigate the long-term changes in precipitation characteristics across different data sources, focusing on the European land area. Our study incorporates station data from the ECA&D project, station data interpolated onto a regular grid (Eobs, Regen) and reanalyses (NCEP/NCAR, 20CR, ERA5). The objective is to explore the seasonal trends of precipitation total amounts, intensity, and probability. The analysis concentrates on the long-term trends of these precipitation characteristics from 1961 to 2010, with emphasis on winter and summer. Trends are estimated using the nonparametric Sen’s slope estimator, and the Mann-Kendall test is applied to assess the statistical significance of these trends. In general, a similar trend pattern is evident within the ECA&D, Eobs and ERA5 datasets. Conversely, the NCEP/NCAR and 20CR reanalyses exhibit the most significant disparities in trends for all precipitation characteristics. Our aim is not only to identify differences between the data sources but also to pinpoint the causes behind these differences.