Location, Location, Location: The Role of Elevation in Flood Models

In a recent blog post, we discussed how floods are among the costliest hazards worldwide. To fully understand flooding risks, institutional investors need to think in multiple dimensions. Knowing the location of your asset is important, but so is knowing its elevation. We found that overestimating elevation by just 20 centimeters can overlook flood risk in over a quarter of global locations.

With so much at stake, investors need high-quality data to build confidence in their climate-risk assessments. Assessing asset exposure to flood risk is contingent on multiple datasets, and projecting the potential financial impacts of flood risk can be highly sensitive to the accuracy of the underlying data.

Flood maps rely on high spatial resolution for accuracy, but vertical accuracy of elevation data is also critical to understanding the potential costs from flooding.¹ Lower-resolution elevation models can result in a misrepresentation of flood risk and subsequently the potential financial impacts to institutions.

We looked at the possible impact of inaccurate elevation data² on flood exposure at asset locations across the MSCI GeoSpatial Asset Intelligence dataset.³ As the elevation bias increases (i.e., the higher the estimated elevation of a surface is compared to our benchmark), the number of asset locations exposed to floods decreases.⁴ While overestimating elevation by 20 centimeters reduces the number of global assets exposed to floods from 35% to 9%, a more extreme overestimation of 1.65 meters reduces the number of assets exposed to just 1%.