Texas develops reliable system to predict when wildfires occur
Texas develops reliable system to predict when wildfires occur
03 March 2011
published by www.bellinghamherald.com
USA — Electronic highway signs across Texas have been warning of “extreme wildfire danger.”
Behind those words is an increasingly sophisticated understanding of why some days are likely to bring blazes covering tens of thousands of acres while others pass in peace.
Sunday’s outbreak of 21 West Texas wildfires, which burned nearly 121,000 acres, occurred almost exactly as expected. An advance notice from the Texas Forest Service warned that a “powerful storm” with high sustained winds and gusts of 60 mph posed danger for much of the flat, dry region.
The first warning that Sunday would be a day of disaster had come five days earlier, based largely on the work of Greg Murdoch, a National Weather Service forecaster on temporary assignment to a state wildfire command center in tiny Merkel, just west of Abilene.
In tandem with the weather forecast was a projection of likely fire outbreaks, speed of fire spread and even flame lengths from Byron Kimball, a U.S. Forest Service fire behavior analyst also on short-term assignment in Merkel.
Lewis Kearney, a public information officer with the Texas Forest Service, was there and watched Murdoch’s forecast take shape.
“He saw this system coming from the Pacific, and then he saw this cold front coming down from the north,” Kearney said days later. “He’d say, ‘It’s going to happen Sunday.’
“What he said came to fruition.”
The advance notice gave Texas time to send in more people and equipment, deployed from operations elsewhere or borrowed from fire departments across the state.
Although 80 homes were destroyed, 80 more were saved from immediate danger and no one was killed – a testament to a forecasting system that combines meteorology, botany, physics and advanced computing, sharpened to a remarkable precision during the last six years.
Murdoch credited improved technology and a network of experts in weather, grasses and fire who work each day during Texas’ winter and spring fire season to mark which areas face the biggest risk.
But accurate forecasts also come down to individuals such as Murdoch, whose skill as an incident meteorologist has taken him, year after year, from his duties as the weather service’s lead forecaster in Midland to extended assignments on fire watch.
His longest time away, he recalled, was 45 days. For now, he’s in Merkel, population 2,600, which offers few distractions from the task at hand.
“Go eat at Skeet’s and Subway,” Murdoch said. “That’s about it.”
In an adjacent chair at the Merkel command center is the U.S. Forest Service’s Kimball. His home base is the flame-prone hills east of Los Angeles, but like Murdoch, he’s pulling temporary duty on Texas wildfires – and not just those on the West Texas plains.
South and East Texas also are at heightened risk of fire.
“Generally, I go to a single fire and I predict what that fire is going to do,” Kimball said. For his current assignment, he said, “they came out and said, ‘OK, you’re responsible for predicting fire over 80 percent of the state of Texas.’
“Oh. OK.”
The meteorologist and the fire behavior analyst are part of a staff of 50 to 60 people working out of the state’s command center. They include supervisors, fire and plant experts, firefighters and support staff – bookkeepers, mechanics, even a carpenter.
The goal is simple enough – to protect people and property – but achieving it requires complex science, done at a full sprint. The making of Sunday’s accurate forecast was a lesson in how that happens.
Murdoch uses a computer-generated composite map that includes the history of thousands of wildfires on particularly explosive days. The spasms of fires have come to be called Southern Plains outbreaks.
Sunday was the 11th Southern Plains outbreak since 2005, when experts intensified their research to find the patterns in seemingly random bursts of flame.
They found that specific combinations of wind speed, temperature, relative humidity and plant dryness dictated the outbreaks. If the conditions weren’t right, few fires would start. If they were right, disaster might ensue.
Experts have known for months that the winter and spring of 2011 would be a high-risk fire season in Texas, based on long-term dry conditions. The trick in forecasting bad fire days is to know when short-term weather conditions will overlay the drought.
“When we see this come about, we start to follow this especially closely,” Murdoch said. “That’s what happened about five days before Sunday.”
That was on a Tuesday. Strengthening evidence said a bad windstorm would arrive on Sunday, plunging relative humidity and stripping away what little moisture remained in the air.
At 2 p.m. Thursday, Murdoch led a conference call with National Weather Service forecasters from Amarillo, Lubbock, Midland, San Angelo, Fort Worth, El Paso, the service’s Storm Prediction Center in Norman, Okla., and Texas Forest Service experts. The 15 people on the call compared their observations and expectations.
The group, Murdoch said, reached a strong consensus: “The wind was going to howl that day.”
Questions remained, however: Would the storm pass through at night, with cooler air, lighter winds and less danger; or at the height of daytime heat, supercharging the risk?
And would the mass of dry, dangerous air stay over ranch, farm and small-town territory, or stretch farther east into the growing suburbs west of Fort Worth?
Forecasters could only wait, watch and calculate.
Weather is only part of the risk. The dryness of the grasses and scrub – “fuel,” in firefighting terms – is the other. That’s where Byron Kimball and the Rothermel fire-spread model came in.
The model is a computer simulation of how weather and fuel dryness and type combine to make fires. It can predict how fast a fire will spread and how long the flames will reach – critical, possibly lifesaving information for firefighters.
Kimball plugged drought and plant-dryness data from the Texas Interagency Coordinating Center, a state-federal venture in Lufkin, into the model. The results were sobering.
West Texas plants “haven’t had any moisture all winter long,” Kimball said. “They haven’t gotten into their growing season yet. So it’s just a whole big layer of dead fuel out there that’s available to burn.
“With the winds that we’ve had, those fuels react very quickly. That’s why the winds are such a concern. They just push fire through those fuels, and those fuels will react and burn very, very quickly.”
The model produced a chart – moisture in the grass on one side, wind speed on the other – that fire crews could carry into the field.
“Here’s what your fuel moistures are going to be for the day,” Kimball said. “Here’s what your winds are in your area. This is the type of behavior that you may see on your fire.”
The combined weather and fuel forecasts turned out to be “very much right on,” Kimball said. It was a daytime storm, the worst. Fire hit some small towns but spared the suburbs west of Fort Worth.
Still, the severity of near-hurricane-force winds and desert-dry air shocked even the most experienced experts.
“The wind started gusting around 50 by late morning,” Murdoch said. By afternoon, Amarillo recorded a gust of 69 mph; a spot north of Amarillo, 70; Lubbock, 65; Midland, 53; Abilene, 49.
Midland reached 3 percent relative humidity, basically turning the air into tinder.
“I’ve been out at Midland for 19 years forecasting the weather,” Murdoch said. “I was amazed by that.”
Kimball said he’ll take a Texas lesson back to Southern California.
“Your weather forecasters here and your local knowledge have done an excellent job of learning what the trigger points are, learning the warning signs of these conditions,” he said.
“And these conditions are not catching people by surprise.”