Spatio-temporal patterns of rainfall-runoff event characteristics and their potential drivers in Germany

Rainfall-runoff event characteristics are a useful source of information to reveal hydrological processes and how they change in space and time. However, a small number of studies can be found in literature, analyzing how rainfall-runoff event characteristics change between different climate and landscape conditions. In this study we present a new event separation method that allows for a fast separation of hydroclimatic time series in rainfall-runoff events and base flow periods. It provides a formal framework for judging event independence and refine multi-peak events. Based on a data set of 185 German catchments we separated more than 220,000 events and analyzed the spatio-temporal patterns of different event indicators such as event runoff coefficient, the event flashiness and event peak runoff. There are clear regional patterns of event characteristics showing that different landscape units react differently to changing climate conditions. Self-Organizing Maps and Principal Component Analysis are used to unravel the drivers of the spatial patterns. Climatic drivers and soil properties are principal descriptors of most patterns. Land use and geological features do not seem to exert a major control on event characteristics.