CLIMADA features

Main characteristics

CLIMADA is a framework, not a data model. It focuses on broad applications and flexibility rather than data fine-tuning.
Code Architecture: The code is divided into core modules (very high stability) and petals modules (advanced optional features and data models).
Supports multiple geographical scales: Worldwide, Country, Region, City, ...
Handles various event types: probabilistic sets, single events, time series, ...
Adapts to different time scales: annual extremes, isolated events, seasonal or monthly extremes, hourly forecasts, ...
Supports multiple hazards: Tropical cyclones, Winter storms, Wildfires, Floods, Drought, Heatwaves, ...
Applicable to diverse exposures: People, Ecosystems, Assets, Economic supply chains, Critical infrastructures, Heritage sites, Coral reefs, Wind turbines, ...
Risk and impact: past, present, and future, including adaptation options.
Offers various output metrics: impact at location, return period curves, risk transfer metrics, average impact, risk maps, forecast probabilities, warning levels, ...
Applications : risk assessment, adaptation option appraisal, storylines, forecasting, insurance pricing, impact attribution, ...

Major modules

UNSEQUA (Uncertainty and Sensitivity Quantification)

The UNSEQUA module allows running global uncertainty and sensitivity analysis on CLIMADA's impact and adaptation option appraisal outputs. It utilizes state-of-the-art algorithms for quasi-Monte Carlo sampling and sensitivity index calculation using the SALib package.

Kropf et al. (2022) [DOI: 10.5194/gmd-2021-437].

Impact Function Calibration

Calibrate impact functions for each impact assessment to ensure the model is truly aligned with your data. Seamlessly connect the obtained ensembles of impact estimates or average ensembles to the UNSEQUA module for uncertainty quantification.

Riedel et al. (2024) [DOI: 10.21105/joss.06755].

Lines and Polygon Handler

The lines and polygons handler enables seamless disaggregation and aggregation of exposure and impact data into lines and polygons.

Mühlhofer et al. (2024) [DOI: 10.1088/2515-7620/ad15ab].

OpenStreetMap Integration

Seamlessly integrate any element from OpenStreetMap as an exposure layer into CLIMADA using the OSMflex package.

Mühlhofer et al. (2024) [DOI: 10.1088/2515-7620/ad15ab].

LitPop - Asset Exposure Estimation

The LitPop module generates globally consistent physical asset exposure layer estimates based on population data, nightlight satellite imagery, and GDP estimates.

Eberenz et al. (2020) [DOI: 10.5194/essd-12-817-2020].

Various Modules to Support the Creation of Hazard Datasets

Hazard data is supported in various formats, including raster files (e.g., NetCDF, GeoTIFF) and vector data (Shapefiles, CSV, Excel). Data from well-known sources such as ISIMIP, ERA5/6, and CMIP5/6 can be used. Certain datasets are also directly available through the data API.
Tropical cyclone wind intensities can be computed using several parametric models based on tracks generated from various sources, with direct support for tracks from IBTrACS, STORM, CHAZ, and Kerry Emanuel.
River flood maps can be derived from GloFAS discharge data using the dedicated module: Riedel, L. et al., Geosci. Model Dev., 17, 5291, [DOI: 10.5194/gmd-2021-437].

Major features

Cascading Critical Infrastructure Risks

The cascading infrastructure module enables the computation of indirect impacts of extreme events on critical infrastructure service provision by modeling cascading failures.

Mühlhofer et al.(2023) [DOI: 10.1016/j.ress.2023.109194].

Multi-Hazard Risk and Recovery

A recent study by Stalhandske et al. (2024) introduced code for integrating multi-hazard impact and recovery analysis, which is now publicly available.

Stalhandske et al.(2024) [DOI: 10.1038/s41598-024-55775-2].