Climate Data Sources
Every climate model is a data problem before it is a modelling problem. This page sets out the datasets used, how they are joined to exposure, and how gaps are proxied and documented.
Hazard and exposure data
Hazard layers come from public reanalysis and national catalogues, supplemented by vendor catastrophe models where resolution or peril coverage requires it. Exposure data is geocoded and quality-scored before any hazard is applied.
- Public reanalysis and national hazard catalogues
- Vendor catastrophe model output for peak perils
- Geocoded exposure with address-quality scoring
- Emissions and energy-intensity data for transition work
Proxies and lineage
Where data is missing, a documented proxy is used rather than a silent assumption. Each proxy records its source, the population it stands in for, and the direction of the bias it introduces.