In our new Wildfire Risks Tool, we provide you with the Wildfire Hazard Potential (WHP) for any location in Washington state. Today, we talk with one of the co-developers of the WHP about what the data means and how you can use it accurately.
Greg Dillon is a Spatial Fire Analyst with the Fire Modeling Institute, a program of the US Forest Service’s Rocky Mountain Research Station. The institute is housed at the Missoula Fire Sciences Lab in Missoula, Montana, the country’s only federal research lab focused exclusively on wildfire.
Dillon, along with other fire scientists, developed the WHP in response to the growing need for information about wildfire risk. Here, he shares a bit of the history of the WHP and his insights into its use.
WSRB: What spurred the creation of the WHP?
Dillon: There have been versions of this product going back to about 2007, and there was another data product before that, too. Over time, it became apparent that people — federal land managers, state land managers, policymakers and others — wanted spatial information about wildfire hazard potential. That’s why the WHP was born: to meet those needs.
We have a lot of useful data about wildfire occurrence and modeled fire probabilities, and the WHP is a way to integrate a lot of that information into a five-class map.
The Wildfire Hazard Potential helps forest managers and policymakers deploy
resources. It can also help insurance companies understand wildfire risk.
WSRB: What does the WHP mean?
Dillon: It’s an index of wildfire hazard, but what does that mean, right? It’s intended to quantify the relative potential for a wildfire in any place that might be difficult to control.
By relative, I mean one area’s WHP is relative to other areas’ WHPs. The absolute value of a WHP doesn’t mean a lot. When you see “very high” and “high” WHPs, you’re looking at an area with a value in the upper percentile ranges of the index.
By difficult to control, I mean the probability of a fire happening and it being high intensity. We talk about intensity in terms of flame length. Places with the highest WHPs are those with a high likelihood of wildfire occurring and higher likelihood of longer flame lengths.
If you have an area with a high probability of fire but where flame lengths are on the lower end, we would consider that to be a place where suppression can be more effective, and so the WHP would be lower.
WSRB: What factors go into the WHP?
Dillon: We use a collection of input datasets. One set depicts fuel characteristics based on vegetation from the national LANDFIRE project. Another summarizes about 20 years’ worth of weather observations including temperature, humidity, wind, etc. We also look at where fires have ignited since the early 1990s, when we have good records.
The next step is to run thousands of simulations on the computer. Each simulation is an iteration of a hypothetical fire season with different weather conditions and ignition locations.
The pattern of past ignitions is used to inform where ignitions are placed in the simulations during the model runs. Ignitions can occur anywhere in the landscape, but there is a higher probability of them happening where there were historic fires.
The simulator does its work in small regional polygons, about 130 across the U.S., that are tuned to statistical patterns in each region’s weather and fire occurrence in the last 20 years or so.
The results of those simulations are used to construct the WHP.
Some Washington state communities face high risk of wildfire. The Wildfire Hazard
Potential, in combination with other data points, can help you understand that risk.
WSRB: What would you like property insurance professionals to know about the WHP?
Dillon: The biggest thing I would say is to keep in mind that this is a product developed at the national scale. What that means is it doesn’t account for a lot of localized issues.
If you use this data at a specific property point, I would interpret it with a little skepticism. We often recommend that people who want to do that also think of the average value of the WHP within some region. I’ve looked at it at the ZIP-code level, which I think is useful for a community scale.
A single pixel of the map is 270 meters on a side, so one pixel represents an area of 18 acres, roughly. You really want to go out at least several pixels to consider the context of what's around you, so a couple kilometers or a mile out, from any individual property.
WSRB: Some areas show up as “non-burnable lands.” What does that mean?
Dillon: That’s another caution. Most areas with vegetation are classified by LANDFIRE as having burnable fuel, but if an area isn't covered by natural vegetation, it gets classified as "non-burnable."
This includes most developed urban and suburban areas. Our fire behavior models simply aren't built to account for all the ways that fire might spread in a built environment. Many homes will be located in areas mapped as "non-burnable," and this is why it's so important to zoom out a little and consider the surrounding area.
We know there is some degree of wildfire hazard in communities, even in some places mapped as non-burnable. We've tried to account for that in a new website that recently launched called Wildfire Risk to Communities.
There is an interactive map where you can search for a community and explore information about relative wildfire risk. The "Exposure Type" map there shows areas we've mapped as indirectly exposed to wildfire, meaning that they are mapped by LANDFIRE as non-burnable, but because of their proximity to burnable fuels, they may still be exposed to wildfire.
There is a download page on the site where people can access GIS data, including a new version of the WHP that includes values for indirectly exposed areas. There are also a lot of useful educational resources to help homeowners understand wildfire risk and reduce their risk.