Project name: Quantification and characterization of runoff volume and chemical composition of precipitation water

Project duration: 01.08.2022 - 30.09.2024

Funding organization: BEAK – Eisenbahnbundesamt

Contact person: Lucia [dot] Pedrosa [at] geo [dot] tu-freiberg [dot] de (Lucia Pedrosa), Traugott [dot] Scheytt [at] geo [dot] tu-freiberg [dot] de (Prof. Dr. Traugott Scheytt)

Project partners:

 

The project aims to identify potential hazards from railroad operations, which could be carried into the subsoil by infiltration of rainwater, and ultimately derive the potential risks based on the results. The project consists of a holistic analysis based on chemical and physical analysis of samples and modeling approaches. For a comprehensive analysis that is not limited to individual locations, it was necessary to set up representative measuring points. These should represent different climatic conditions and natural areas as far as possible and consider all factors and compartments potentially involved in the input parameters. Based on a comprehensive workflow, five sites were selected and installed. For this holistic approach, soil, groundwater, surface water, and weather data were recorded at all sites and sampled according to a developed concept. During the analysis, all substances potentially introduced by railroad operations (e.g. residues from mineral oil hydrocarbons, heavy and semi-metals as well as herbicides), elements, and compounds responsible for a possible degradation and retention in the soil were included and analyzed. In addition, the microbial biomass and basal respiration in the topsoil zone were determined.  All the data available will be used to simulate and validate the models developed for each site. Hydrus 2-D allows to mimic and understand the complexity of the system composed by the railway’s embankment, site-specific geological setting, and the contaminant’s characteristics. With the intent to shed light on soil water content variation, preferential water flow paths, and the different contaminants dispersion behaviors. Furthermore, the results and conclusions of the five modeled sites will be transferred to an improved roadway network prediction model implemented by artificial intelligence methods (e.g. neural network). Ultimately, resulting in a country scale map predicting a multi-hazard risk analysis of railways in Germany.