New cat model brings ‘step change’ to flood risk management: Fathom’s Jones

Covering all major flood perils globally with a comprehensive representation of uncertainty and conditioned to the current climate, Fathom’s new Global Flood Cat model gives the industry a “step change” in its ability to manage flood risk, according to the firm’s Chief Product Officer, Dr Matthew Jones.

Built upon Fathom’s Global Terrain Data, FABDEM+ and its Global Flood Map, Global Flood Cat reportedly offers comprehensive multi-peril coverage, a customisable toolkit, advanced uncertainty representation, best-in-class components, global exposure data, and more.

Speaking with Reinsurance News following the launch, Fathom’s Jones explained that there are other flood models out there, but none that are based on so many components that are extensively and openly validated in peer-reviewed literature.

“This gives the industry a step change in its ability to manage flood risk. We’ve also ensured our Global Flood Cat model is as transparent, customisable and user-friendly as possible,” Jones said.

He continued, “The model is built on the Oasis Loss Modelling Framework and is available on multi-model platforms, such as Nasdaq’s Risk Modelling for Catastrophes platform and will soon be available on Moody’s Intelligent Risk Platform.

“Not being tied to one platform helps reduce barriers to adoption and increases competition and model availability – which is a good thing for a peril as difficult to model as flood, where often a multi-model view is the best approach.”

Discussing the firm’s decision to make a cat model, Jones said that Fathom was formed to provide “comprehensive flood risk intelligence for the entire planet”.

He added, “Our Global Terrain Data and Global Flood Maps are part of this journey and now our Global Flood Cat model marks the final piece in the Fathom foundational jigsaw. Cat models are the ultimate tool for assessing flood risk because they are essentially huge catalogues of synthetic events and so contain information on which assets can flood at the same time.

“They allow us to move from hazard assessment (e.g. flood depths at different frequencies of flooding) to risk assessment (i.e. financial loss metrics) as they include vulnerability curves that contain the relationship between depth and damage, and enable uncertainty to be represented in a structured way.

“Cat models allow us to understand portfolio risk, taking into account spatial diversification and correlation. This means that, anywhere on the planet, the risk for all major flood perils can now be quantified for all major asset types and coverages, which is something we’ve been striving towards since the foundation of the company.”

Commenting on the cat model’s implications and use cases for the re/insurance industry, Jones observed that these are similar to any other and include accumulation and capital management, reinsurance structuring, pricing, planning and regulatory and rating agency returns.

“What’s particularly valuable with our model is how customisable it is – instead of spending valuable resources creating a proprietary model to reflect their view of risk, organisations can incorporate their custom information (such as flood defences, or even client-specific vulnerability curves) directly into our model,” he explained.

Jones went on, “Flood is a risk that is present everywhere in the world, and there has been a wider recognition over the last few years that flood risk shouldn’t be lumped into a ‘secondary peril’ or ’non-modelled’ category but rather should be given more attention.

“Flood losses can significantly impact earnings and, in some cases, capital. In many of my previous roles for risk carriers, I would have loved to have a tool like this available. It allows companies to avoid surprises (always good if you are in a risk/exposure management role!) and grow into new areas with confidence that ‘eyes are wide open’ with respect to flood risk. This means that there is much more chance of growing profitably and in a diversified way.”

Concluding the interview, Jones commented on some of the future challenges that still need to be tackled to make cat models more accurate, stating that from a flood risk perspective, the firm envisages further improvements to terrain data, which will have a direct impact on the ability to assess flood risk.

“Another area where there will likely be a step change in the next few years is better knowledge of the location and dimensions of river channels. This will largely be driven by NASA’s SWOT mission; an exciting project we are closely involved in which involves a new satellite that surveys all of the Earth’s oceans, major lakes, rivers and wetlands in a 21-day cycle. SWOT will also help us to more accurately capture flood event water levels which will improve our flood map validation,” Jones said.

He finalised, “A very significant uncertainty in any cat model – and particular flood cat models – is vulnerability (the relationship between flood depth and damage). The challenge here is the availability of a large quantity of detailed, location-specific flood claims with associated depths of flooding.

“We have published academic papers on this topic in the past but unless initiatives like OpenFEMA and the NFIP claims database are replicated elsewhere, a lack of quality claims data will be a limiting factor.

“This means that it’s very important to ensure that uncertainty is well represented in the model, otherwise the metrics from the model, and the consequent risk management decisions, will be incorrect.”

This website states: The content on this site is sourced from the internet. If there is any infringement, please contact us and we will handle it promptly.