Fire scientists from Maryland, Colorado and Riverside have collaborated to combine fire modeling with satellite data to develop a new method for predicting wildfire behavior in near real time. The resulting predictions will be more accurate and revised based on the fire’s actual behavior.
For years, fire experts have researched and designed models for predicting a fire’s behavior. One problem, according to Dr. Philip Riggan with the U.S. Forest Service’s Fire and Fuels program out of its Riverside facility for the Pacific Southwest Research Station, is the difficulty of scaling lab experiments to the size and proportion of actual wildfires.
Nevertheless, there are models predicting fire behavior, such as rate of spread, based on terrain slope, fuels, moisture and other physical characteristics. “This is the basis for many models available in operation, but [real] fires often get to a certain size and don’t relate well to small lab fires,” Riggan said.
Some models do incorporate weather data, but similar to weather forecasts, their reliability deteriorates quickly after a few days. “Satellite data have, for more than two decades, provided routine active fire information but have not been able to distinguish between individual fire lines or to validate fire behavior on all but the largest fires,” wrote the principal researchers in the landmark article published in the the Geophysical Research Letters in October.
Essentially real-time data describing the location and direction of fires is captured from aircraft. While not predictive, the aerial images (usually infrared from night-time flights) have been helpful to firefighters and demark the conflagration’s size.
The new technique, which Dr. Janice Coen of the National Center for Atmospheric Research and Wilfred Shroeder created, is the melding and combination of the model with the real-time physical data, explained Riggan.
“With this technique, we believe it’s possible to continually issue good forecasts throughout a fire’s lifetime, even if it burns for weeks or months,” said Coen in an NCAR press release. “This model, which combines interactive weather prediction and wildfire behavior, could greatly improve forecasting — particularly for large, intense wildfire events where the current prediction tools are weakest.”
The model’s accuracy was tested with data from the 2012 Little Bear Fire in New Mexico.
“[Coen] has coupled the complex atmospheric equation with equations for fire movement,” Riggan said. For example, this technique allows for the possibility that a fire’s intensity might generate its own winds. “This is not considered in most models,” Riggan added.
The model also incorporates physical data, which change during the fire’s lifetime, but are not part of the static models. Spotting, which can expand and change the fire’s direction, as well as the effects of the firefighting, are examples of the additional data.
The satellite data, which are much better resolution today (about 1,200 feet compared to half a mile), can be available two to three times each day. Thus, the fire’s position is more accurate, but its behavior is now used to modify the models to enable better allocation of firefighting resources, Riggan explained.
Forecasts using the new technique could be particularly useful in anticipating sudden blowups and shifts in the direction of the flames, according to the researchers’ comments. In addition, they could enable decision makers to look at several newly ignited fires and determine which pose the greatest threat.
“This could give us a better record of how fire travels and a better understanding of rates of spread,” commented Dan Felix, fire officer for the San Jacinto Ranger District.
“Many people have resigned themselves to believing that wildfires are unpredictable. We’re showing that’s not true,” Coen was quoted in the release.
J.P. Crumrine can be reached at [email protected].