Rainfall erosivity (defined as R-factor of the Revised Universal Soil Loss equation (RUSLE)), a key driver of soil erosion by water, varies greatly across Europe due to differences in climate regimes, precipitation patterns, storm intensities, and their frequencies. Updated and harmonized rainfall erosivity datasets are therefore important to provide robust input data for soil erosion models and to account for changes in climate that occurred in recent decades. This study presents REDES 2.0, the new pan-European Rainfall Erosivity Database, based on high-frequency sub-hourly and hourly rainfall data from 9138 locations across Europe, which corresponds to an average station density of 1.9 stations per 1000 km2. REDES 2.0 represents an order-of-magnitude increase in data availability compared to the 1675 stations included in REDES (henceforth referred to as REDES 1.0). All the collected high-frequency rainfall data was converted to 30-min temporal resolution, either using data aggregation or rainfall disaggregation based on a multiplicative cascade model. Rainfall erosivity was subsequently computed following the RUSLE methodology. As a result, over 2.4 million erosive rainfall events were identified, predominantly covering the 2010–2025 period, with average station data length of 15 years and with an average of around 265 erosive events per station. Spatial patterns of annual R-factor values show pronounced differences across Europe, with the highest R-factor values concentrated in the Mediterranean and Alpine regions, while northern, eastern and central Europe exhibit comparatively lower values. REDES 1.0 and 2.0 overlap at 1062 locations (approximately 17% of REDES 2.0 and 69% of REDES 1.0) across Europe, where mean annual R-factor values increased from 792 MJ·mm·ha−1·h−1·yr−1 in REDES 1.0 to 861 MJ·mm·ha−1·h−1·yr−1 in REDES 2.0. This corresponds to around 9% increase in the rainfall erosivity in the last decades. Moreover, around 65% of stations included in REDES 2.0 exhibit a positive trend in annual R-factor for the 2010–2024 period. The standard error of the annual R-factor for REDES 2.0 was on average around 18% for stations included in REDES 2.0. REDES 2.0 reduces uncertainties and errors in rainfall erosivity in Europe, covers previously unsampled areas, increases temporal coverage, provides trends, and includes many extreme events.

BEZAK Nejc;  BORRELLI Pasquale;  CESARINI Luigi;  MÜLLER-THOMY Hannes;  MATTHEWS Francis;  SAGGAU Philipp;  KAFFAS Konstantinos;  LIAKOS Leonidas;  BRETTIN Jana;  BALLABIO Cristiano;  GALDIES Charles;  PETAN Sašo;  VALIUKAS Donatas;  GIANNAKLIS Christos-Panagiotis;  LAGOUVARDOS Kostas;  GOMES Goncalo;  SALAMON Peter;  BEGUERÍA Santiago;  LAKATOS Monica;  PERČEC TADIĆ Melita;  LONČAR-PETRINJAK Ivan;  CHERVENKOV Hristo;  MICHAELIDES Silas;  PANAGOS Panos; 

2026-07-10

ELSEVIER

JRC146004

1872-6887 (online),   

https://www.sciencedirect.com/science/article/pii/S034181622600603X,    https://publications.jrc.ec.europa.eu/repository/handle/JRC146004,   

10.1016/j.catena.2026.110393 (online),