Publications from the joint project KliMoBay
Determination of DEM based Predictor Variables for Königsdorfer Weidfilz peatland and analysis of the relationship between the Predictor Variables and Depth to Water level data
Ishita Jalan, 2021, Chair of Hydrology and River Basin Management, Technical University of Munich, Study project
Peatlands have turned from carbon sinks to carbon sources due to human interference in the ecosystem. Ditches dug on the surface have caused peatlands to drain and release carbon dioxide, methane, and nitrous oxide. Large outflux of these greenhouse gases have the potential to further stress the global carbon budget in the atmosphere and therefore, peatland restoration needs to be addressed urgently. Water level depth is a proxy variable to determine the exposed peat to oxidation due to the drawdown of water. In this Study Project, predictor variables were determined for the bog site of Königsdorfer Weidfilz that influences the hydrology of the peatland. They would feed as input to the statistical model to generate regionalized water level maps for peatlands. The four predictor variables were dependent on the Digital Elevation Model of the site as the main source of information. The two predictor variables, Ditch Length and Nearest Distance to the Ditch were determined using vector data of active ditch network. The other two variables, Topographic Wetness Index and Relative Altitude were determined directly from the Digital Elevation Model using GIS tools. Workflows to determine the predictor variables were automated using the ModelBuilder tool in ArcGIS Pro. Twenty-three datasets were determined and then analyzed to find their relationship with depth to water level measured at 13 piezometer points. Correlation analysis was carried out for the datasets with and without outliers to determine the most meaningful relationships. The results were found plausible for at least one dataset for each predictor variable. Significant linear regression models were found between water level and Relative Altitude. The evidence of finding reasonable relationships can form a basis to select the most useful datasets for developing the statistical model in the future. There are limitations to the outcomes as the Digital Elevation Model represents topographical features of the site before restoration efforts were performed. Also, the impact of rewetting measures and land subsidence is not captured by the current raster data. Further, the network of active ditches was developed based on visual evidence and therefore, subjected to limitations posed by image visibility.