This script describes the basic functionality of the dynamic vulnerability estimate applied in Sauer et 2021, embedded in an Flood Damage Attribution framework. It was developed for a collection of vullnerability models for multiple hazards for the Horizon Europe project COMPASS. The script is structured providing an overview over all input data required, then it lists all functions used in the vulnerability assessment and the attribution assessments, both applications are the described at the bottom of the script. The exampe application shows damage aggregates for a region in Latin America.
To install all required dependencies install the conda environment stored in environment/regional_vulnerability.yml
All steps required to apply the model are described and visualized in the jupyter notebook https://github.com/ingajsa/compass_3_3_regional_vulnerability/blob/main/notebooks/regional_vulnerability_model.ipynb. The dynamic vulnerability function vul_func takes the input data described in the input data section of the notebook. The attribution assessment is devided into two steps: I) The trend assessment using the non-parametric Mann-Kendall-Test attr_MK and the relative_contributions. Note that impact records and simulated damages must be gathered or modeled prior to the application of the script.