From Deseret News archives:
U. researcher homes in on wildfire prediction
The research is to be published this fall in the International Journal of Wildland Fire. The prediction is based on discovering when vegetation will get so dry that it can become subject to huge wildfires.
"I looked at the Santa Monica mountains in Southern California because of the type of vegetation that grows there, chaparral," said Philip Dennison, assistant professor of geography at the University of Utah. The chaparral, a brush somewhat taller than sagebrush, is uniform in places, and its water content could be calculated.
Dennison's co-authors are Max Moritz of the University of California at Berkeley, and Robert S. Taylor, a fire specialist at Santa Monica Mountains National Recreation Area.
Dennison said the same sort of research could be carried out in Utah, too, but the project would be more complicated. For one thing, the abundant wild grasses add to the fuel load.
"Right now it's only been tested in chaparral," he said of the theory. "It may work in sagebrush, but it hasn't been tested."
The scientists studied records of the largest fires that occurred in the Santa Monica Mountains, going back 22 years. By large fires, they don't mean the blazes that break out nearly monthly but huge infernos that scorch more than about four square miles.
Three ingredients were needed for the big fires, they found: high winds, an ignition source and critically dry vegetation.
High Santa Anna winds often bluster through the area, and lightning strikes, careless campers and other causes frequently get fires going. But those factors alone don't guarantee a monster fire.
"We determined that the moisture in the vegetation needs to be below a certain level for big fires to occur," he said.
"We looked at data from satellites, and we looked at data from weather stations," Dennison said. They discovered that the amount of precipitation in March, April and May was a good indicator of what the fire season would be like a little later in the year.
They tested their theory against records of precipitation and fires.
"We found that in half the years we could predict within one week" when a bad fire was likely to start. Usually, they could predict within two weeks.
Dennison said he hopes that soon, fire forecasts can give warnings one to three months in advance. Someday, a solid prediction could be available by June 1, which is early in the fire season.
Someday the Beehive State may benefit from specific wildfire forecasts. "I think eventually that's where we're going," Dennison said.
"I definitely want to see this expanded into Utah and used locally."
E-mail: bau@desnews.com
