Underwriting Wildfire Risk & Athena’s Yellow Flags

Athena Intelligence
5 min readJun 8, 2024


Alternative Title: Maximizing Profitability with Excess & Surplus Pricing Using Athena Intelligence

Athena is a data vendor whose Voice of the Acre® offers superior wildfire insights. By analyzing the characteristics of the land in past wildfires, Athena can generate probabilities for insured losses, even in areas newly exposed to wildfire risk due to weather volatility. We provide insurers with Green/Yellow/Red underwriting decision flags, making it easy for underwriters to integrate Athena’s data directly into their existing software.

This article is an expansion upon our article Profitably Underwriting Properties with Wildfire Exposure, which describes how property insurers can use Athena to underwrite wildfire risk with the necessary information to appropriately price and reserve for these risks.

Photo Credit Michael Held, Unsplash

In the era of big data, the challenge lies not just in gathering information from complex natural systems, but in discerning what truly matters. Athena’s business goal is to use machine learning to create sophisticated algorithms that are commercially viable. The larger goal is to help humans understand the complexities of terrain-based information, giving voice to the land to aid organizations in decision-making (hence, the software’s name: Voice of the Acre®).

Harnessing the power of machine learning, we can cut through the noise and extract meaningful patterns from the multitude of factors that carry statistical weight. This enables those who need future wildfire risk insights to prioritize their actions effectively.

The marriage of machine learning and data from nature’s complexity is emerging as a dynamic force in various fields. Adding a geospatial component makes Athena’s data even more valuable, offering decision-makers a streamlined and informed perspective.

We continue to grapple with the human tendency to want to understand by looking to individual examples. Our response is to emphasize that Athena is a probabilities-based solution designed to be commercially viable. It is intended to be useful, consistent, and accurate across a portfolio of properties. While achieving 96% accuracy in portfolio results, it is important to note that this is not equivalent to 100% accuracy at any given location. In other words, a small subset of Green/Low Risk properties may still experience wildfires each year.

For insurers looking to maximize profitability by seeking high premiums on properties with some risk, we offer a Yellow or Moderate Risk label. These properties should be insured in the non-admitted market.

For example, consider a property in Novato, California, near the border of Marin and Sonoma Counties.

A yellow flagged property in Marin County, California

Athena’s underwriting engine scored the property Yellow, with a wildfire destruction probability of 1:137. That means that, if a wildfire occurred anywhere in Marin County, the odds of this home being within the perimeter of the fire is 0.73% or about 3/4ths of one percent.

As described in the article above, Athena’s underwriting tool has been optimized to predetermine the ultimate wildfire perimeters. The goal of Athena’s conditional, geospatial, artificial intelligence risk model is to determine if a structure is inside or outside of the fingerprint of the wildfire.

Risk Assessment for 2023, report generated May 24, 2023

Staying with this example, homes in Marin County generally carry a higher risk than most homes in America. Half the land in the county is designated as federal, state, county, or city parks, watershed set-asides, conservation districts, protected marshlands, or land trusts. From a wildfire perspective, Marin County has relatively few ecosystems, which translates into few natural fire breaks.

However, for this specific home, while the overarching risk in the county is high, the location-specific risk is only elevated. Athena’s analysis is based on 30-square-meter pixels and a series of concentric circles that account for how embers are blown during a fire. Each pixel has a unique alphanumeric string that can be tied directly back to the source information. These 9 unique identifiers also include a time series and a projection of vegetation growth.

From a property insurance underwriter’s perspective, while the location’s conditional risk for wildfire damage is high, this specific address has several green (low risk) conditions. The probability of a burn at this location, the risk to the structure, and the hazard potential are all low. If a wildfire occurs in this area, the intensity here is expected to be low.

However, the wildfire may have picked up inertia or be rapidly moving with the wind. Given the similarity of the vegetation and ecosystems, the county is vulnerable to a wildfire spreading from one area to another similar area. A cluster of homes in close proximity, built in similar ways with the same materials, creates an increased risk because the fuel is consistent. This phenomenon is termed “Building Exposure.”

If a wildfire did burn one home in this neighborhood, it could more easily spread from the first home to a second, and then through much of the neighborhood, until it encountered a change in fuel type or a fire break.

In other words, a single homeowner’s home hardening efforts help their neighbors by reducing the overall neighborhood risk. And one neighbor’s unwillingness to prune trees, or remove highly flammable species, can impact the risk to a dozen or more homes in the surrounding neighborhood.

This is not trivial as currently there over 80,000 communities at risk for wildfire damage. Many of these communities are only now becoming aware of their new peril, which has increased with climate change.

Every year more communities are at risk of wildfires

Communities must balance wildfire risk with the positive aspects of having trees and other vegetation. Living near trees reduces stress and improves health, contributing to overall well-being. As we adapt to changing weather patterns, shaded neighborhoods will become even more desirable. Furthermore, trees reduce particulates, sequester carbon, mitigate the impact of rainfall, and improve water quality.

Accurate and mathematically precise information about future wildfire risk will make informed decisions by underwriters and homeowners easier. For more examples of how the information can be used, please visit the Case Studies at AthenaIntel.io.

Athena Intelligence is a data vendor with a geospatial, conditional, profiling tool that pulls together vast amounts of disaggregated wildfire and environmental data to generate spatial intelligence, resulting in a digital fingerprint of wildfire risk. (website: AthenaIntel.io)

Clients include electric utilities, communities, insurance and financial services companies, where Athena’s geospatial intelligence incorporated into wildfire mitigation plans (WMP) and public safety power shutoffs (PSPS), Community Wildfire Protection Plans (CWPP), property insurance underwriting and portfolio risk optimization.



Athena Intelligence

Athena Intelligence weaves vast amounts of disaggregated environmental data. Drop us a line (Info@Project-Athena.com), or visit www.athenaintel.io