2.3 Physical Risk

This part details the technical considerations, the methodological approach, and the information needed for the quantification of physical risk metrics. We use a varied range of data inputs and apply geospatial transformation techniques to yield averaged values of physical risk for each asset in our database. The selected assets cover 66 TICCS asset subclasses across 8 TICCS superclasses and are derived from a representative sample of the Unlisted Infrastructure Universe^[1]. Finally, we merge these output values with financial inputs from infraMetrics to give rise to the full spectrum of physical risk metrics for floods, storms, and thermal stress.

2 models for 3 types of hazards: floods, storms, and thermal stress
Using 33 types of damage functions
Covering 23 countries in Europe, the Americas, and Asia-Pacific
Physical risk calculations stemming from
- floods for assets in 66 TICCS subclasses across 8 superclasses
- storms for assets in 52 TICCS subclasses across 7 superclasses
- thermal heat and cold stress for assets in 52 TICCS subclasses
  across 7 superclasses

2.3.1 General modelling approach

2.3.2 Data

2.3.3 Physical risk models

2.3.4 Robustness of the physical risk results

_{[1] Our models are based on the same sample of 760 assets for which}_infra_{Metrics provides financial information and that we have used for our carbon emission models. This sample represents the Unlisted Infrastructure Universe. It needs to be taken into account that for some assets, our models only provide carbon emissions or physical risks, for few assets, our models provide neither due to limitations in data availability.}