This repository contains the code used to derive the results of the paper: Gebhart V., Schmid T., Mühlhofer E., Villiger L., Bresch D. N. (2025) User-tailored impact-based hail forecasts in Switzerland, (not yet submitted).
Publication status: not yet submitted
Contact: Valentin Gebhart ([email protected])
Jupyter notebooks to reproduce figures and statistics that appear in the paper. Before running impact_based_forecasts.ipynb for the first time, you need to first run impact_function_calibration.ipynb to calibrate the impact function.
Contains several utility functions for data processing, calibration, visualization. These functions represent a subset of all utility functions used in the damage modelling part of the scClim project, see also https://github.com/timschmi95/crowdsourced_paper.
Requires:
- Python 3.11+ environment
- CLIMADA repository version 6.*: https://wcr.ethz.ch/research/climada.html https://github.com/CLIMADA-project/climada_python
- Hail forecast data from MeteoSwiss (COSMO or ICON models).
- Shape data for Swiss cantons and warning regions.
- Exposure and damage data for the calibration. The hail damage data used in the paper are not public and only available within the scClim project.