Read our COVID-19 research and news.

Firefighters confront the boundary fire on the Kaibab National Forest in Arizona earlier this year. Started by a lightning strike, it burned some 7000 hectares.

Brandon Oberhardt/U.S. Forest Service (CC BY SA 2.0)

Who is starting all those wildfires? We are

As parts of the western United States choke on smoke from wildfires scorching more than 660,000 hectares, renewed attention is falling on the role that people have played in starting some of these blazes. An Oregon fire that has consumed 13,000 hectares, for instance, is thought to have been started by teens tossing firecrackers.

Jennifer Balch, a wildfire ecologist at the University of Colorado in Boulder, has examined just how big a role people are playing in starting wildfires in the United States. Nationwide, humans are responsible for starting 84% of wildfires, according to a paper co-authored by Balch, published this past March in the Proceedings of the National Academy of Sciences. In California, the eastern United States, and the coastal Northwest, people are behind more than 90% of wildfires. And, by starting so many fires, humans are essentially lengthening the fire season, into times of the year when natural causes—such as lightning—don’t play a major role.

ScienceInsider spoke with Balch about those numbers, and their implications. This interview has been edited for brevity and clarity. 

Q: How did you get interested in how people start wildfires in the United States?

A: We recently stumbled across a stash of government records that contained information about the cause [of a wildfire]. Exploring that database—over a million and a half records of government-recorded fire events [between 1992 and 2012]—we realized what a substantial role today people are playing in starting wildfires. We found that 84% of all wildfires were started by people over the last couple of decades in the U.S.

I was shocked. We've used fire as a tool for a long time. But the fact that people have such a large print on our wildfires was surprising.

Q: Where are most wildfires started by people?

A: If you look at a map, you see a majority of human-started fires happening in the Southeast, in the eastern U.S., and on the western coastal areas—California, Oregon, Washington. It's only in the Intermountain West, where it is fairly devoid of people, where you see a strong signal of lightning-started fires.

[In some mountain areas, such as Colorado,] we see predominantly human-started fires along road networks and smack against the major metropolitan areas—Boulder and Fort Collins, Denver and Colorado Springs. So we see a really strong signal of humans essentially swamping out natural, lightning-started fires.

Lightning storms [are] mainly concentrated during summer months. With people, what we're doing is effectively tripling the length of the natural fire season because we're distributing ignitions throughout the year.

Q: Were you able to track the ways people start fires?

A: So the breakdown: Of the approximately 1.5 million wildfires in the government record, 25% were burning of trash and debris; about a quarter (22%) were unknown human causes. The next biggest category is arson, [then] heavy equipment, campfires, children, and smokers. Those are the seven biggest categories.

Fireworks didn't rank in the very top for the whole year, but it does pop on July 4th. It’s the day with the most fires. Over 7000 events started on July 4th alone. They were predominantly started by fireworks. It's unfortunate that our Independence Day didn't fall in January or December when it's cooler and wetter.

Q: It looks like the biggest increase in human-caused wildfires is in the Great Plains. Why?

A: There's a lot of agricultural industry and there's a lot of cattle ranching going on in the Great Plains. So it's important to recognize that there are a lot of intentionally set fires to manage landscapes. Fire can get rid of stubble and old crop residues. Fire is also used in cattle ranching to rejuvenate pasture grasses. The data we're looking at is fires that needed some sort of suppression response. Se we are not capturing the intentionally set fires. But we are capturing the ones that escaped.

Q: There’s a lot of talk about the problem of people building houses at the edge of fire-prone forests. Have you looked at how that connects to people starting wildfires, and whether it might change in the future?

A: That's actually work that we’re doing right now. I can tell you that 9% of the total land area of the U.S. has this intermix between houses and natural areas. That's projected to double by 2030. So we don't expect this problem to go away. In fact, we expect that—because people are living in these landscapes and carrying ignitions to wherever they go—we're actually going to see more wildfire started by people in the future.

Q: Are there parts of the United States at particular risk?

A: Yeah, it's projected to starkly go up in the Intermountain West.

Q: It appears that the number of ignitions is going up in every part of the country except California. Why is it that?

A: It's something that has been noted not only by our research but by some other really prominent papers that have come out recently. [They show] in the western U.S. an overall increase in the number of fires and burned area, except for [the part of California with a Mediterranean-type ecosystem]. I've been asking a lot of questions about what's going on there, and I haven't been able to get a clear answer. Here are some hypotheses or guesses: increased public recognition of the danger of wildfires, increased suppression, or another possibility is that as more people are fragmenting the landscape, you have a decrease in the overall size of fire events. I think there's an important story, because there might be a lesson there in terms of how we're dealing with fires in the West as a whole.

Q: Any insights into measures that could reduce the number of fires being started by people?

A: Part of the take-home is that, given we do play such a substantial role in starting our own wildfires, we need to accept living in flammable places. There's good fire, there's bad fire, and we need more of the right kind of fire. What I mean by that is we need more prescribed burning and we need more cultural acceptance around prescribed burning, because we cannot remove fire from the landscape. We've already tried that experiment for 100 years and it didn't work. What our results provide justification for is shifting people's contribution to something where we're setting more prescribed fires, rather than setting wildfires and then suffering the consequences. We need to start the right kinds of fires.

It's hard to swallow that, especially right now when a lot of the western U.S. is choking on smoke. But ultimately the question is: “How do we want our smoke? Do we want our smoke in large concentrated pulses like this, or do we want it more dispersed?”


In a follow-up email, Balch provided a scorecard for the 2017 fire season. “This year so far (1 January 1 - September 11), there have been 41,775 wildfires started by people vs. 6,354 wildfires started by lightning across the U.S.,” she wrote. The total area burned by human-started fires, she added, was about 1.459 million hectares, compared with 1.843 million hectares burned by lightning-started fires.