Transparent inputs. Governed models. One comparable axis.
We are open about every data source, scenario and what each band means, and disciplined about what screening can and cannot tell you. Each result is produced by a single governed model under formal version and change control.
What screening is, and is not
Our analysis is screening-level: it prioritises assets for attention and ranks exposure across a portfolio on a common axis. It is not a hydraulic flood model, an engineering wind-load study or a catastrophe-loss estimate.
A High band means "investigate with site-level detail": the point at which a deeper, location-specific study earns its keep. It is not a prediction of loss.
0.25° global climate backbone
Global climate variables are generally read on a consistent 0.25° grid. Hazard-specific models supplement these inputs with additional datasets at their native resolution, including elevation, hydrodynamic flood modelling, built-up surface data, coastline geometry and storm tracks.
The common grid smooths local relief: flood defences, small watercourses, urban heat islands and complex shorelines are not fully resolved. Where an asset sits near a cell boundary, snapping can read a neighbouring cell, one reason High bands warrant site confirmation.
Every hazard on the same prioritisation scale
Every physical hazard is translated onto the same Low / Medium / High axis, allowing unlike hazards to be compared consistently across the portfolio so it can be ranked, not just listed. The common scale supports consistent prioritisation: it does not mean a coastal result and a heat result are physically equivalent.
- Low — Limited screening exposure under the assessed conditions
- Medium — Elevated screening exposure worth monitoring and planning for
- High — Priority exposure requiring more detailed review using site-specific information
Multi-scenario, multi-horizon
Physical hazards run across the SSP scenarios; transition risk runs across the NGFS narratives. Both are read at multiple horizons.
Physical: SSP scenarios
- SSP1-2.6 Low emissions
- SSP2-4.5 Middle of the road
- SSP3-7.0 Regional rivalry
- SSP5-8.5 Fossil-fuelled
Transition: NGFS narratives
- Orderly Early, smooth policy
- Disorderly Late, abrupt policy
- Too Little, Too Late Weak, fragmented policy
- Hot House World Limited policy, high physical risk
Horizons: a present-day baseline, reported as 2025, then near-term 2030, medium-term 2040 and long-term 2050.
The baseline represents the model's reference climatology and is labelled 2025 for reporting purposes. It does not necessarily consist solely of observations recorded during 2025: some hazard models use a historical reference period, for example 1985 to 2014 for river flooding, to represent present-day conditions.
Scenarios are not predictions. They are internally consistent, plausible futures used to test the resilience of the portfolio to a spectrum of climate outcomes.
Anchored to public climate science
Every model is built on published, citable datasets and frameworks.
- CMIP6 Temperature and rainfall extremes for heat, cold, pluvial, landslide and drought.
- CaMa-Flood (ISIMIP3b) Global hydrodynamic flooded-fraction for river flooding.
- JRC LISCoAsT / MeteoTide Tide and storm-surge components for coastal flooding.
- NASA MSL · World Bank CCKP Mean sea-level rise projections.
- GHSL Urban fraction / built-up proxy for surface-water flooding.
- Best-track records · CHAZ Historical and synthetic tropical-cyclone tracks.
- NGFS Phase 5.1 Carbon-price pathways for transition risk.
- IPCC AR6 WG1 · IIASA SSP Warming-level pattern-scaling and scenario framing.
Governed, versioned and traceable
Models carry a status (governed, candidate or superseded) so a result can always be traced back to the exact method that produced it.
Each hazard is assessed with a single governed model under formal version and change control. The models translate authoritative, peer-reviewed climate datasets into a Low / Medium / High exposure band per asset using thresholds calibrated to the regional climate and reviewed under the governance process.
Percentages reported throughout are the share of assessed locations classified High. They are not a share of asset value or revenue.
From portfolio screening to decision-useful IFRS S2 evidence →
What screening does not cover
Stated plainly, so a result is read for what it is: a screening-level signal that prioritises where to look, not site-engineering or a loss estimate.
Model resolution
Hazard scores are governed-model outputs over coarse-resolution climate data, not site-specific engineering, hydraulic or catastrophe-loss modelling.
Site-specific conditions
Local defences, drainage, micro-topography and building standards are not represented.
Existing controls
Current adaptation and operational mitigations are not reflected in the screening bands.
Asset criticality
Results are by location count; asset criticality and replacement value are not yet incorporated.
Employee data
Workforce exposure is shown at country level where entered; per-site employee data is not yet incorporated.
Financial weighting
Physical-hazard exposure shares are unweighted by asset value or revenue, and asset-level physical financial effects have not been quantified.
Small country samples
Countries with fewer than five assessed assets are asset-specific.
Intended use
A screening-level analysis to support IFRS S2 / TCFD-aligned disclosure and prioritisation, not investment or engineering advice.
Want the detail behind a specific hazard?
Each model page sets out its inputs, scenario assumptions, outputs and worked interpretation.