Fathom, a research-led flood risk modelling specialist, recently launched its U.K. flood model on Nasdaq’s Risk Modelling for Catastrophes service. We sat down with Oliver Wing, Chief Research Officer at Fathom, to learn more about the new model and how it helps insurers, reinsurers, brokers and property owners understand flood risk and related losses across the United Kingdom.
1. Can you start by telling us about Fathom – why was the company formed, and what’s the vision behind Fathom?
Fathom consists of a team of world-leading experts in hydrology and flood modelling. In 2012, a group of researchers from the University of Bristol Hydrology Group had a discussion where they identified the obstacles faced within the flood risk modelling industry and, subsequently, set out to solve them.
The result was the formation of Fathom, with a mission to create the most detailed, accurate and comprehensive global flood risk data in the world. Our models are developed from pioneering methodologies that have been built in-house and undergone some of the most comprehensive validation in the industry. Only this month, Fathom Chairman Professor Paul Bates produced a paper exploring the methodology underpinning our U.S. model, which has been extensively validated against independent datasets and has found to demonstrate a good correlation. We will be producing a similar paper for the U.K. later this year.
Our philosophy is simple: everything that we do is driven by science; and with this, research has remained an essential part of the Fathom DNA. We have established ourselves as leaders within the academic field and regularly publish papers in leading journals.
2. What type of organizations use your models and hazard data, and for what purposes?
Our ambition with our products and services is to meet the needs of a wide variety of industries and sectors, but essentially, anyone with an interest in flood risk can find value in our model.
Principally, the insurance industry can benefit from our data as it enables them to understand what their flood risk is for the buildings that they insure. Particularly, those working in underwriting may use our flood hazard data and catastrophe models to gain insight into flood exposure for specific locations and can understand climate risk under a number of scenarios for multiple time horizons. Reinsurers also benefit from using our models for pricing their risks at portfolio level and managing their capital adequacy. Beyond the insurance sector, Governments and NGOs can leverage our data to mitigate risk at a regional level and build resilience both financially (i.e., insurance) and physically (i.e., flood defences).
Although these are just a handful of ways that organisations can apply our data, we have worked extensively with international development agencies, conservation agencies, emergency responders and multi-national corporations on a global basis.
3. Which are the main aspects to take into consideration when modelling flood hazards, and what are the unique strengths of Fathom’s new U.K. model?
From our perspective, the most important requirement of a large-scale flood model is to have seamless, high-quality terrain data. For Fathom-UK, we have incorporated over 160,000km² of the latest LiDAR data and provide simulations at 10-meter resolution.
Flood science in the U.K. is very mature, thanks to decades of study from our nation’s talented hydrologists. This means we benefit from the dense network of river gauges with long historical records that have been maintained in this country, allowing us to quantify what extreme flood events look like. When we couple this rich dataset with our peer-reviewed statistical modelling methods (what hydrologists call “regional flood frequency analysis”), we can estimate extreme river flows and rainfall anywhere, filling in data gaps in time (some plausible extreme events are yet to be observed in history) and space (many catchments have no observational data available).
All of our models are based upon pioneering methods that we have developed in-house. The size of river channels is not picked up in remote-sensing datasets, so we have to find smart ways to record the location of river channels and estimate their conveyance in the absence of in situ measurements. At Fathom, we have created a dynamic channel solver, which allows us to model river bathymetry in a much more accurate way. In doing so, we can provide predictions of high-frequency flood events with much more fidelity, which, by definition, have an outsized impact on average annual losses.
A unique strength of our model is its vulnerability functions. With over 10,000, the model explicitly represents damage to a variety of different residential, commercial and industrial buildings – for direct (e.g., structural and contents) and indirect (e.g., ALE and BI) damages. The relationships vary based on a number of different variables, including building age, the number of stories, and even which floors are occupied in the building.
4. Climate change is an important aspect in this industry. How is your model taking future impact from climate change into account?
Our flood hazard data provides predictions for multiple climate scenarios from the present-day through to 2030, 2050 and 2070. We use climate model output from the Met Office UKCP18 dataset, offering a number of concentration pathways (RCP2.6, RCP4.5 and RCP8.5) and temperature rise scenarios (2C and 4C) to provide as rich a view as possible on the impact of changing weather patterns on flood risk.
When developing Fathom-UK, we felt that it was important to incorporate climate change into our model to ensure that end-users understand that their flood risk today may very well be different from their flood risk tomorrow.
From our research and modelling, we have found that even where changes in mean catchment conditions are somewhat negligible, virtually uniformly, we see extreme precipitation intensifying. Our model can be used to identify the areas that are most sensitive to climate risk. This is particularly valuable for event response and also those within the financial sector who are facing new regulatory requirements with respect to better understanding future flood risk.
We already have a number of exciting updates that we will be bringing to our models for the U.K. as they mature. For the catastrophe model, this will be the inclusion of the future climate scenarios that already exist within our flood hazard data.
5. This is the second model that Fathom launches with Nasdaq, the first being your U.S. flood model. What made you partner with Nasdaq Risk Modelling, and how has the partnership evolved so far?
We’re keen to work with service providers such as Nasdaq as they make complex models accessible to the end-user and the wider ecosystem. Not only this, Nasdaq is able to provide a holistic view of climate risk across multiple catastrophes. Flooding is only one component of risk, and although we offer a rich array of data, it is made stronger when it is viewed in conjunction with other risk modeller’s understanding of other perils, such as earthquakes, wildfire, or windstorm.
Our partnership began when we introduced our catastrophe model for the U.S. in 2019. Our most recent inclusion of Fathom-U.K. CAT signals an exciting development in the volume of data that Fathom is able to provide to clients within the Nasdaq platform. Looking forward, we are excited to continue incorporating models of further regions as they become available.