Editor’s note: Understanding wildfire risk has never been more important, especially as the U.S. Department of Agriculture’s Forest Service continues its risk reduction work across the country, implementing its strategy, Confronting the Wildfire Crisis. This article explores the emergence of a recent tool, RiskMonitor, and how the big-picture perspective it provides is being used to guide our work.

Understanding wildfire risk can be complicated. It requires the accounting of many variables that contribute to wildfire behavior, namely terrain, fuels and weather. It also requires land managers to understand all that is at stake across a landscape — homes and communities, roads, powerlines, mature and old-growth forest, wildlife habitat and watersheds among many others.

The formula it turns out is already in use in an industry that makes its money understanding risk--the insurance sector.

“Insurance is all about calculating expected losses, the probability of some event happening and the consequences if it does happen,” said Greg Dillon, director at the Fire Modeling Institute at the Missoula Fire Sciences Lab.  He’s been at the fire lab for 20 years, and with numerous partners from inside and outside the Forest Service he’s been researching and developing useful tools that change how the nation understands and mitigates wildfire risk.

That equation for expected losses, along with wildfire behavior modeling and geospatial data, has evolved into an analytical approach suitably named RiskMonitor that provides land managers with a big-picture perspective of wildfire risk like never before.

Since the beginning of wildland fire management, it has been apparent that fire posed risks to life, property and natural resources. But knowing how to be proactive in mitigating those risks has not always been clear.

“One hundred years ago, we didn’t have the ability to think about wildfire risk in the same way. We thought about fire in terms of fire danger. What's the weather? What's the fuel moisture? What's the likelihood of seeing fire based on those kinds of conditions?” said Dillon reflecting on the Forest Service’s formative years. Those questions are still extremely important for understanding day-to-day fire potential throughout the fire year. But the longer-term strategic understanding of what is at risk, and where, is also important for getting out ahead of problem.

The RiskMonitor has detailed wildfire risk data for the 21 Wildfire Crisis Strategy landscapes (with plans to expand to the entire contiguous United States), allowing land managers to identify areas to treat to reduce the most risk.

“It basically allows us to understand where we need to focus, and it’s even more important when looking at larger landscapes. We have to understand something about where that risk is greatest to prioritize and target the highest probability places first,” said Dillon.

The Fire Modeling Institute runs models on the effectiveness of fuels treatments across these large landscapes, conducts the analysis and provides the outputs to land managers to inform their risk reduction objectives and help them know where treatments can be most effective at reducing risk.

“The Forest Service has a lot of science and technology that we can bring to this question of where we can be most effective. [RiskMonitor] it’s enhancing our ability to think about reducing wildfire risk on broader scales of space and time,” said Dillon.

Further than the Eye can See

Reducing wildfire risk requires a landscape-scale approach. One of these landscapes highlighted by the Forest Service in 2023, the Sierra and Elko Fronts, includes large swaths of land across California and Nevada—3.4 million acres total. That’s about 5,300 square miles, larger than the state of Connecticut. The agency has been charged with lowering wildfire risk supported by $3.2 billion through the Bipartisan Infrastructure Law and the Inflation Reduction Act.

“When looking at large landscapes, large projects and lots of money, there are many places we could target. The experience and judgment that serves us well on smaller projects is really hard to manage on these larger ones,” said Tonja Opperman, the assistant director of fire analytics for the Intermountain Region.

Opperman speaks from experience. She got her start on a helicopter firefighting crew out of the Grand Canyon 30 years ago and has been involved in wildland fire, one way or another, ever since. “At the Grand Canyon, it was game on every day of the summer, high adrenaline and really exciting.”

Her on-the-go, action-packed summers eventually gave way to more analytic pursuits. She went back to school earning a master’s in fire ecology and transitioned her career from fighting fire on the ground with a Pulaski to fighting fire through analytics and modeling. She provides critical information to inform the agency’s work to suppress, reduce risk and restore areas affected by wildfire.

Opperman routinely travels to large wildfires across the country and conducts analysis and modeling on how those fires might behave in the coming weeks and where they might spread.

When on an active fire, analysts like Opperman have a lot of variables that influence their models and predictive analysis.

They usually know where a fire is located and its size. They have weather forecasts. They have maps of fuel types and densities of vegetation. They know exact distances between the fire and homes and other values at risk.

“We have a lot of information about a fire to help us understand where a fire is going and our best chance to confine or stop it. But when we talk about a whole region of the country and where we should put our wildfire risk reduction dollars to work, that's a whole different story,” said Opperman.

“We don't know where the next fire is going to start. We don't know if it's going to start on a hot windy day when it will be hard to catch it or on a cooler calmer day. And it’s not possible to treat every acre, so it’s important to focus our work in areas to get the best possible outcome,” she said.

That's where RiskMonitor comes into play.

Informed by thousands of wildfire simulations based on weather, terrain, temperature, fuel and moisture among many other variables, RiskMonitor calculates the probabilities for wildfire across the landscape. Those probabilities are then mapped and overlayed on a map of known values at risk such as homes, infrastructure, wildlife habitat, communities and watersheds. The model even takes into account the susceptibility of a given value once exposed to wildfire, estimating for example how likely a wildfire is to damage a building versus a powerline.

Opperman remembers when wildfire risk wasn’t really quantifiable on a large scale at the beginning of her career.

“Now we have the luxury of looking at a coast-to-coast wildfire risk map of the contiguous U.S. based on real observations and real climate data to start making the plan,” she said. “It’s just so much more powerful than what we could have done manually in the past.”

Common Picture, a Common Goal

In California and Nevada, on the Sierra and Elko Fronts Wildfire Crisis Landscape, there are many shared benefits of wildfire risk reduction: making homes, communities and landscapes more resilient to wildfire, fostering safer working conditions for wildland firefighters and protecting recreation havens.

“Also, in our community healthy rangelands and thriving wildlife habitat are important. They are synonymous with the economy, and they are essential to consider when looking at what’s at risk,” said Duncan Leao, a wildfire crisis strategy coordinator on the Humboldt Toiyabe National Forest.

Leao started out working in wildland fire a little more than 20 years ago. Since then, he’s worked in fuels, silviculture, timber and vegetation management. He’s very familiar with the types of work and treatments that need to be done to lower wildfire risk. Now he’s coordinating it on a scale he had never quite imagined before – 3.4 million acres.

When Leao first started his career, land management was implemented at a much smaller scale. Often it wasn’t necessary to have big-picture conversations with partners. But, he said, it’s different now.

“The fire season is now a fire year, the changes in the ecosystems, the rate, magnitude, and severity of the wildfires,” said Leao. “With the current situation, we have to work a lot closer with partners across much larger areas.”

When the Humboldt-Toiyabe was selected to receive a large investment, some assessments and modeling had already been done in those areas, but the Forest Service didn't have data for the entire landscape.

Then when Leao and his team received an early rollout of RiskMonitor in June 2024, the tool allowed the team to validate their existing plan and to look further out, ensuring the right work was being done in the right places at the right time.

“It’s not only a good tool for us to do our part to reduce risk, but also helps us assist communities and our partners in making good decisions, too. Everybody has a role to play,” said Leao. “RiskMonitor gave us a way to follow up with our partners and say, ‘Hey we'll do work on national forest lands, but those treatments won't be as effective unless there is work on this jurisdiction over here, too.’”

Back at the Fire Modeling Institute, Dillon and the team continue refining the RiskMonitor application. As of now, it provides a solid framework for thinking about strategic planning, prioritization and monitoring of risk-reduction activities. Going forward, they are working to develop it into a useful tool, not just for the agency but also for partner agencies, organizations and landowners interested in reducing wildfire risk. The specifics of what that tool will look like are still under development.

“We're at the stage where it's beginning to be adopted and starting to be used in the planning process on forests across the county,” Dillon said. “We have the science and technology to inform smart decisions, to make us better partners, and to help people do the work that needs to be done at the scale that’s needed.”