Heavy Precipitation Events of Various Durations Across Germany: A Station-Based Assessment of Spatial and Temporal Variability Using the Block Maxima Method

Abstract. As heavy precipitation events (HPEs) pose substantial risks to natural and human systems, a growing body of research has focused on their behaviour under ongoing climate change, hypothesising that rising air temperature can have a pronounced influence on precipitation patterns, including HPEs of various durations. Recent observational and modelling studies suggest that this influence tends to be particularly strong for short‑duration HPEs, albeit their assessment appears challenging due to limited availability of precipitation data with both high temporal resolution and long-term observational records. Therefore, in this study, we made use of recently collected, digitised, and quality‑controlled 5‑minute precipitation data from rain‑gauge stations across Germany in order to assess the spatial and temporal variability of HPEs of nine durations ranging from 5 minutes to 7 days. Using the block maxima method, we confirmed that the spatial and temporal variability of annual maximum precipitation totals (AMPTs) is strongly duration‑dependent. While the medians of short‑duration AMPTs calculated for individual stations are relatively evenly distributed across Germany, those of medium‑ and long‑duration AMPTs increasingly reflect the influence of topography. Moreover, short‑duration AMPTs exhibit higher and spatially scattered event to event variability, contrasting with the relatively consistent and regionally organised event to event variability observed for long‑duration AMPTs. The duration‑dependent nature of AMPTs is further reflected in their long‑term variability. While positive trends prevail for AMPTs with durations of 1 to 7 hours, trends for AMPTs with durations of 3 to 7 days are more balanced or even slightly negative, depending on the investigation period. Yet, statistically significant – both positive and negative – trends are relatively rare and sensitive to changes in the measurement system and investigation period.