Changing Wildfire Perils

With wildfires seeming to blaze in all directions, the summer of 2015 was a season of disaster in the American West and Pacific Northwest.

Large fires sweeping across multiple states burned at least nine million acres between January and early September, ravaging timber, homes, and wildlife at a nearly unprecedented rate. In terms of acreage, this year’s wildfires were 50 percent more destructive than the 10-year average, according to the National Interagency Fire Center’s statistics for the first nine months of 2015.

For many insurers across the country, wildfires have become one of the top five sources of loss, with more than $3 billion paid out in fire claims during the past decade. Unlike other perils, wildfire risk is unique in the way it is connected to factors already present in the environment. And yet there are cutting-edge technologies available to help insurers better understand and monitor risk factors for wildfires. The result can be more effective underwriting and more efficient handling of claims.

California: A State of Disaster

Already under stress from a long-term drought, California has become a flashpoint for wildfire disasters. In September alone, two fires in Northern California—the Valley and Butte Fires—claimed six lives while consuming more than 145,000 acres and setting 1,500 properties ablaze. The aftermath left several communities in smoking ruins, and California’s governor declared a state of emergency. President Obama later pronounced the Valley fire a “major disaster.” It now ranks as the third most damaging wildfire in California’s history in terms of structures destroyed, according to the state’s records.

Similar highly destructive wildfires also were fought in Idaho, Oregon, Washington, and Montana. All told, more than 48,000 fires were reported in the five states in just the first nine months of the year, raising questions about how best to protect lives and property.

So what’s the most sensible way to assess risks for a wildfire season that no longer spans months but may last through the year? Three factors contribute to wildfire risk, as identified by the National Fire Protection Agency (NFPA). The first is the presence of vegetative fuels: dry grass, brush, or undergrowth that might serve as fuel to spread a fire. The second factor is an area’s topography, which can speed or hinder the advance of a fire. And the third factor involves an area’s roads, since access is critical in enabling first responders and suppression units to reach the location of a fire. Together, those factors can help predict potential risks from uncontrolled fires.

Factors for Wildfires

Vegetative fuels are the primary ingredient that feeds wildfires. Fuels can vary in combustibility, depending on the nature and distribution of wildland ecosystems within the broader climatic setting. Fuels are usually linked to the dominant vegetation classes in an area and include grasses, trees or woody vegetation, and shrubs/brush. For instance, a landscape dominated by brush (as found in parts of Southern California) would be considered full of “heavy” fuel and therefore more prone to burn than an area with grass, which is considered a “light” fuel (as found in parts of southern Idaho). In fact, thick brush (a heavy fuel) has been the chief fuel involved in many highly destructive fires in Southern California.

Topography often influences the growth and development of wildfires. When fires burn on a slope, the resulting hot air rises to dry vegetative fuels farther up the slope. This effect advances fires rapidly upward, putting properties along slopes at risk. In addition, properties located downslope are then sometimes exposed to embers released from wildfires. In general, the complexity of a given area’s terrain can enhance risks from wildfires. Not surprisingly, terrain has been found to be a significant influence on property damages in fires across the western United States: Terrain affected more than 85 percent of damaged properties in the Jesusita Fire (California, 2009) and more than 90 percent in the High Park Fire (Colorado, 2012) and Little Bear Fire (New Mexico, 2012).

Road access can influence wildfire risk in rural areas where properties are difficult to reach for fire suppression. Properties located on dead ends and minimal-access roads could be at elevated risk for damage.

In addition to the factors described, embers carried by wind patterns in Southern California and other areas can increase the risk of wildfire-related losses. These patterns include strong and dry downslope winds that tend to be present during autumn and winter and coincide with the peak fire season. Analysis of losses from the Station Fire (Los Angeles County, California, 2009) reveals that more than 80 percent of damaged properties were exposed to elevated risk of embers associated with Santa Ana wind patterns. Similar insights were obtained from analysis of properties damaged by the Jesusita Fire (Santa Barbara County, California, 2009). In that fire, more than 85 percent of damaged properties were exposed to risk of embers from Sundowner wind patterns, according to an analysis by Verisk Analytics.

Can Analytics Reveal Fire Risks?

It already has been established that factors of vegetation, topography, and weather can be assessed through remote-sensing technology. Digital mapping technology can precisely trace road networks. The combination of data can result in a telling integration of factors for a given site. Those factors can then be plugged into a mathematical model to derive an objective numerical score to be used in comparing wildfire risks for different properties.

Recently, a short-term trend pointing toward more destructive fires that burn longer has raised the prospect of a longer-term reality. In a changing global climate, temperatures are projected to rise and create drier conditions, which could adversely influence risk of wildfires across America’s western states. In this scenario, the incendiary summer of 2015 wouldn’t be a rare year of disaster but a potentially more frequent occurrence. Research has shown, for example, a connection between wildfires and early spring snowmelt in mountainous regions. An earlier snowmelt can often lead to a longer and drier summer season with drier vegetation—and help prepare potential fuel for wildfires.

Further complicating the situation has been a surge in residential building. A University of Wisconsin study found that 8.6 million homes have been built within 30 miles of a national forest in western states during the past three decades. Across the western United States, estimates suggest that 4.5 million homes may be at high or extreme potential risk from wildfires. Historically, the majority of losses in wildfires—close to 90 percent—occur on properties in the high- or extreme-risk categories. The majority of the highest-risk properties are in California, Colorado, and Texas, which together contain 70 percent of the riskiest properties in the western states.

California has the largest number of households at high or extreme risk (2,054,400) or about 15 percent of all households in the state. After California, Texas ranks second in number of households at high or extreme risk with 706,200, followed by Colorado with 363,900. Seven western states have more than 13 percent of households at high or extreme risk. By that measurement, Montana ranks first (27 percent), followed by Idaho (25 percent) and Colorado (16 percent), according to the 2015 Verisk Wildfire Risk Analysis (see sidebar, “Spreading Like Wildfire”).

As populations increase and expand into rural areas, the exposure of households and businesses to wildfire-prone land is likely to increase markedly, bringing demands for better assessment of fire risks.

Handling Wildfire Claims

Timing and safety remain two primary challenges. It typically takes some time before an insurance carrier is aware that a wildfire has affected an insured property. Wildfires also take a while to subside, so there’s often a time lag before properties are deemed safe enough for a claims professional to reach the site for an inspection. Yet while a wildfire is still burning, many insurers need to deploy catastrophe response teams to begin the recovery process for policyholders.

In such situations, remote-sensing technology can be a useful tool. The availability of data of varying granularity streaming from satellites, drones, and other aerial platforms can help provide a tremendous opportunity to monitor and assess wildfire-affected areas in real time. Timely and critical analytics in the form of fire perimeters and maps of areas with active burning can support catastrophe teams. Using a variety of geospatial platforms to analyze damage, it’s possible to determine how best to deploy response teams and understand the potential effects on policyholders.

Many insurers can benefit from tools that help score properties for wildfire risks while monitoring the areas affected by wildfires. Underwriting and exposure management can often benefit from property scores in several ways, including risk selection, underwriting eligibility, pricing of policies, and setting guidelines for management within ZIP codes. Property inspections can be targeted toward properties at higher risk, leading to more effective inspections.

On the claims side, timely information on wildfire activity in relation to exposures can help inform catastrophe management plans. Precise information about loss locations can then help expedite the claims settlement process and lead to improved customer satisfaction. In a changing climate, science and technology are leading to a deeper understanding of wildfires—and what it takes to survive them.