A high-resolution streamflow and hydrological metrics dataset for ecological modeling using a regression model

Hydrological variables are among the most influential when analyzing or modeling stream ecosystems. However, available hydrological data are often limited in their spatiotemporal scale and resolution for use in ecological applications such as predictive modeling of species distributions. To overcome this limitation, a regression model was applied to a 1 km gridded stream network of Germany to obtain estimated daily stream flow data (m 3 s −1 ) spanning 64 years (1950–2013). The data are used as input to calculate hydrological indices characterizing stream flow regimes. Both temporal and spatial validations were performed. In addition, GLMs using both the calculated and observed hydrological indices were compared, suggesting that the predicted flow data are adequate for use in predictive ecological models. Accordingly, we provide estimated stream flow as well as a set of 53 hydrological metrics at 1 km grid for the stream network of Germany. In addition, we provide an R script where the presented methodology is implemented, that uses globally available data and can be directly applied to any other geographical region. Design Type(s) modeling and simulation objective • data integration objective Measurement Type(s) hydrological process Technology Type(s) Linear modelling • data transformation Factor Type(s) precipitation Sample Characteristic(s) Germany • stream • river Machine-accessible metadata file describing the reported data (ISA-Tab format)