Ordinal pattern dependence between hydrological time series

摘要: Abstract Ordinal patterns provide a method to measure dependencies between time series. In contrast classical correlation measures like the Pearson coefficient they are able not only linear but also non-linear even in presence of non-stationarity. Hence, noteworthy alternative approaches when considering discharge Discharge series naturally show high variation as well single extraordinary extreme events and, caused by anthropogenic and climatic impacts, non-stationary behaviour. Here, ordinal is used compare pairwise derived from macro- mesoscale catchments Germany. Differences coincident groups were detected for winter summer annual maxima. Hydrological series, which mainly driven conditions (yearly discharges low water discharges) showed other some cases surprising interdependencies macroscale catchments. Anthropogenic impacts construction reservoir or different flood urbanization could be detected.