Why today's hazard maps fail tomorrow's decisions
June 23, 2026
Most hazard maps show where major acute hazards occur today. Owners of long-lived assets need to know how those hazards will change. Two locations with similar exposure now can diverge sharply under different warming pathways, and disclosure frameworks such as TCFD and CSRD ask for physical risk under more than one scenario.
Major acute hazards, mapped to 2080
Emmi Geospatial Physical Risk Data covers major acute hazards at global scale, built on peer-reviewed climate and hazard datasets. Hazards are mapped at between 1 km and 11 km resolution, depending on the hazard. Each is projected for 2030, 2050 and 2080 against a late 20th-century baseline, under scenarios ranging from RCP2.6 to RCP8.5.
Change from baseline is the core measure
Global climate models capture how extremes intensify more reliably than they resolve local infrastructure and defenses. The data therefore measures change from a historical baseline, which isolates the climate signal. Intensity is scaled to a common global reference, so higher-warming scenarios show proportionally higher hazard instead of being flattened by local scaling.
Severity and frequency combine into expected loss
Each hazard has two metrics. A normalized intensity index captures severity, and an annual exceedance probability captures how often damage-relevant thresholds are crossed. These combine into Average Annual Loss (AAL), the expected yearly damage as a fraction of asset value, which can feed expected loss and Value-at-Risk models. The underlying physical variables are included for users who want to apply their own damage functions.
A screening layer for portfolios
The data is built for portfolio and geography-scale screening, scenario comparison, and deciding where detailed analysis is needed. AAL uses standardized damage relationships and doesn’t account for asset-specific vulnerability or local defenses. Locations with high or fast-rising hazard can be flagged for asset-level assessment.
Download the full white paper for the methodology, data sources and coverage tables.