Hydrological and biogeochemical processes in small catchment areas significantly influence the water quality in the headwaters of rivers and streams. Long-term trends in time series, such as the globally increasing concentrations of dissolved carbon in the runoff, can be understood as a system response to changing hydrological/climatic boundary conditions. Often, however, a clearly identifiable relationship between climatic drivers and changing processes in the catchment area and the resulting system response is difficult or impossible to detect. For the statistical description of these relationships as well as the systematic description and characterization of datasets, time series analysis offers a wide range of methods.
The aim of this project is to systematically analyze the long-term data from the monitoring program of the Bavarian Environment Agency (LfU) for correlations and patterns that can be attributed to the effects of local climate change in the last 20-30 years. In addition, future developments will be estimated using a statistical forecasting model (artificial neural networks) based on local climate forecasts and scenarios. The results from the statistical analysis will, then, be compared to results obtained by process-based models of selected catchments. Findings from this project will be used in practice-oriented recommendations for decision-makers at local and regional level.