Cape Analytics' technology considers "defensible space" parameters to better inform insurers and local wildfire response groups of wildfire insurance risk
Insurtech start-up Cape Analytics has today released new technology it claims will help both insurance companies and emergency responders better identify and mitigate the risk of wildfire damage.
The software uses computer vision – a type of image recognition tech – to assess the wildfire risk to a property.
It also uses machine learning, a branch of AI, to enable this risk assessment process to take place across millions of US properties.
Kevin Van Leer, client solutions manager at the firm, said: “Cape Analytics is building on the industry’s recent technological strides in response to major events, such as explicit ember transport in catastrophe models, as well as updated hazard scores.
“The suite of wildfire tools now available better address traditional underwriting, rating, and portfolio management.
“Delivering a unique, location-specific understanding of fuel loads around a property precipitates the next leap in understanding wildfire risk and defensible space at the most granular level.”
Wildfire insurance risk models not fit for purpose
Fuel loads – a term used to refer to any flammable materials in a wildfire risk zone – have traditionally been ignored by wildfire risk models, according to research from California’s Department of Insurance (CDI).
A 2018 CDI report states: “Legislators, other public officials, and their constituents have expressed concern that wildfire risk models are not accurate, do not provide satellite imagery that is granular enough to objectively identify fuel sources and other physical characteristics, and do not take into account mitigation done by the homeowner or the community.”
California is the worst-hit area globally in terms of wildfire damage, and suffered 20 of its most extreme wildfires on record during 2018.
By December 21, 8,527 fires had burned an area of 1.9 million acres – the largest amount of burned acreage recorded in a fire season, according to the California Department of Forestry and Fire Protection (Cal Fire) and the National Interagency Fire Center.
Figures from reinsurance giant Munich Re showed wildfires in the state caused $24bn (£18.3bn) in losses – $18bn (£13.7bn) of which were uninsured.
Ernst Rauch, the reinsurer’s chief climatologist, recently told The Guardian that property insurance could become unaffordable in California’s most at-risk locations.
Cape Analytics’ new product is the latest in a series of image recognition technologies developed by insurtech start-ups to better understand property insurance risk – but it is unique in that it focuses solely on assessing and mitigating wildfire risk.
Mitigating wildfire insurance risk
Wildfire risk is increased when a property has risk factors such as being situated near vegetation and tree limbs, as well as the position and perimeter of its own roof in case a fire spreads to nearby trees.
It’s for this reason that Cal Fire advises property owners to maintain a safe distance between their buildings and the surrounding vegetation, creating what it calls “defensible space”.
Cape Analytics claims its technology is complementary to leading wildfire risk models, while also providing the more granular data needed by at-risk homeowners and local response groups to manage the risk of wildfire by taking into account this defensible space.
This information is also critical to insurers and reinsurers, as 100ft of defensible space is required by law in California, and other US states, but insurers don’t have the resources to inspect every home.