Groundwater level time series, meteorological forcings and static feature dataset for 667 wells in Germany
Zenodo dataset for Groundwater level time series, meteorological forcings and static feature dataset for 667 wells in
- Source
- Zenodo
- Published
- 2025
- Region
- Global
- Format
- multiple formats
- Access
- Open access
- License
- cc-by-4.0
- Themes
- Geospatial & GIS Data, Fresh Water
About this dataset
July 2025, compiled by Tanja Liesch Contact: ORCID: https://https://orcid.org/0000-0001-8648-5333 Dataset accompanying the publication "Strategies for Incorporating Static Features into Global Deep Learning Models" The dataset is an anonymized subset of the dataset "GEMS-GER: A Machine Learning Benchmark Dataset of Long-Term Groundwater Levels in Germany with Meteorological Forcings and Site-Specific Environmental Features" published by Ohmer et al. (2025, https://doi.org/10.5281/zenodo.16736908), that was used in the manuscript "Strategies of static features incorporation into global deep learning models for groundwater level prediction" (submitted to HESS) by Liesch, T. and Ohmer M. (2025). It contains weekly groundwater level data from 1991-2022 for 667 wells across Germany, along with meteorological forcings and static environmental data from the original dataset. Additionally, nine time series features per well were computed and added, as well as time series plots for all wells. The corresponding code can be found on GitHub: https://github.com/KITHydrogeology/dynamic_static/.
How the data was collected
July 2025, compiled by Tanja Liesch Contact: ORCID: https://https://orcid.org/0000-0001-8648-5333 Dataset accompanying the publication "Strategies for Incorporating Static Features into Global Deep Learning Models" The dataset is an anonymized subset of the dataset "GEMS-GER: A Machine Learning Benchmark Dataset of Long-Term Groundwater Levels in Germany with Meteorological Forcings and Site-Specific Environmental Features" published by Ohmer et al. (2025, https://doi.org/10.5281/zenodo.16736908), that was used in the manuscript "Strategies of static features incorporation into global deep learning models for groundwater level prediction" (submitted to HESS) by Liesch, T. and Ohmer M. (2025). It contains weekly groundwater level data from 1991-2022 for 667 wells across Germany, along with meteorological forcings and static environmental data from the original dataset. Additionally, nine time series features per well were computed and added, as well as time series plots for all wells. The corresponding code can be found on GitHub: https://github.com/KITHydrogeology/dynamic_static/.