Oregon State helps update database to predict future wildfire behavior

PORTLAND, Ore. ( KOIN ) – A database that tracks wildfires across the United States was recently enhanced, with help from an Oregon State University professor, to project the possibility of when and where a wildfire could spark.

“There is a tremendous amount of interest in what enables wildfire ignitions and what can be done to prevent them,” said OSU professor Erica Fleishman. “This database increases the ability to access relevant information and contribute to wildfire preparedness and prevention.”

The Fire Program Analysis Fire-Occurrence Database was originally developed in 2013 by the United States Forest Service and has since undergone several updates.

Researchers added nearly 270 attributes to the database, which now includes information on 2.3 million fires in the U.S. from 1992 to 2020.

Now, the database includes new factors that can influence a fire, such as topography and vegetation, social vulnerability and economic justice metrics, and the distance from the ignition to the nearest road.

In addition to helping on-the-ground firefighters and managers, OSU said the database could also help power companies evaluate whether to implement a public safety power shutoff and could help land management agencies determine whether to restrict access to public lands or restrict campfires throughout the year.

“There seem to be a lot of policies that are guided to some extent by intuition or emotions rather than by a large body of evidence,” Fleishman said. “These data present one way to increase the objective evidence to consider when making those decisions.”

“This provides a considerably deeper understanding of the individual and compounded impact of these attributes on wildfire ignitions and size,” said Yavar Pourmohamad, a doctoral student at Boise State University who helped add the new attributes. “It also identifies the unequal effects of wildfires on distinct human populations and ecosystems, which can, in turn, inform efforts to reduce inequities.”

OSU furthered that information from the database can also be used in artificial intelligence and machine learning models to explain what drove fires from the past or can forecast the likelihood or effects of future fires.

“It’s amazing what you can infer when you have the computational capacity and this much information,” Fleishman said. “You can ask a lot of questions that inform different actions in different places and to understand what is associated with wildfire ignitions and fire effects.”

Fleishman told KOIN 6 News she does not believe the database has been used to predict wildfires before, noting the tool is primarily used to understand trends, such as the number of wildfires, locations, sizes and causes. She added that it is “virtually impossible” to predict a wildfire.

Work is in progress to update the system to identify potential causes of wildfires, the OSU professor said, adding this model appears to have a high overall accuracy of more than 70%.

The database was recently featured in the journal Earth System Science Data and is supported by the Joint Fire Science Program – a program from the U.S. Forest Service and the U.S. Department of the Interior.

For the latest news, weather, sports, and streaming video, head to KOIN.com.