How Forest Fires Start: Modeling Wildfire Ignitions

Officials think this wildfire near Bastrop, Texas, started from sparks from power lines. The fire destroyed almost 1,700 homes. Photo by Joe Wolf (JoeInSouthernCA, Flickr).
Officials think this wildfire near Bastrop, Texas, started from sparks from power lines. The fire destroyed almost 1,700 homes. Photo by Kerri West of Bastrop, Texas.

Most wildfires are started – whether accidentally or intentionally – by people. Understanding where wildfires are most likely to occur and how they can be predicted has been a major focus for U.S. Forest Service Southern Research Station (SRS) researchers Jeffrey Prestemon and his colleagues and experts from state, other federal, and tribal land management agencies along with representatives from the law enforcement community.

Together, these organizations published Wildfire Ignitions: A Review of the Science and Recommendations for Empirical Modeling, a general technical report based on original work done for the National Cohesive Wildland Fire Management Strategy. To read more about the report’s recommendations for modeling wildfire ignitions, read “The Science Behind Wildfire Prevention.”

Arson, escaped debris burns, campfires, and engine sparks are the most common types of human-related ignitions. Over the past few decades, several of these ignition sources have become less common. Cigarettes used to frequently provide the spark that would set off a wildfire, but since the early 1970s there has been a 90 percent reduction on federal lands in this ignition source. The drop is probably related to the fact that fewer people today are smoking than 30 years ago (although other factors might also lie behind the decline in reported smoking-caused fires).

Arson has also become a less common source of wildfire on most public and private lands. “This could be due to stricter sentencing and greater efforts to identify, arrest, and prosecute suspected arsonists, but it also follows a long-term nationwide drop in the crime rate overall,” says Prestemon. Declines in the number of fires started by railways and other industrial equipment could be attributed to better spark arrestors and their more widespread and consistent use. There is also evidence that public service announcements and brochures can reduce the number of human-caused wildfire ignitions, especially if outreach is conducted before or during times of the year when wildfire risks are highest.

Nevertheless, wildfires still cluster around places where there are more people, and heavily used transportation corridors and recreation areas are at greater risk of wildfire ignitions. Higher population density is also linked to more frequent wildfires –- up to a point. As natural areas become fragmented by neighborhoods or other developments, fuels become increasingly disconnected, and the risk of wildfire drops.

“In short time frames, human-caused fires cluster in regular, predictable patterns associated with work, leisure, and time of year,” says Prestemon. “Weekends and holidays often lead to spikes in human caused ignitions, while ignitions from escaped debris burning or arson increase in certain seasons.”

“A major focus of this publication is to summarize past attempts at predicting ignitions, and provide recommendations for future modeling of wildfire ignition,” says Prestemon. “Fire managers seeking a clearer understanding of ignition causes and trends in their area will find useful information in our conceptual model of wildfire ignitions and accompanying list of recommendations.”

– Adapted from “Wildfire Ignitions: State of the Science in the Southeast” by Alan Long and Jeffrey Prestemon, a fact sheet published by the Southern Fire Exchange

For more information, contact Jeffrey Prestemon at jprestemon@fs.fed.us

Read “Wildfire Ignitions: A Review of the Science and Recommendations for Empirical Modeling”

Access the latest publications by SRS scientists.

Subscribe to our newsletter!

Receive weekly updates