Media

Sources/Usage: Public Domain. View Media Details

Lutz and his colleagues studied forest fires across Utah with a goal of creating a fire atlas. He wanted to go back in time as far as possible, so he turned to the U.S. Geological Survey (USGS) for its Landsat satellite data. The USGS Earth Resources Observation and Science (EROS) Center maintains the archive of Landsat data, and scientists at the center study how land changes over time—including the effects fire has on a landscape. 

Specifically, Lutz modeled his study on the Monitoring Trends in Burn Severity (MTBS) project, with a few twists: His “Utah Fire Atlas” tracks all fires 100 acres and larger, includes the type of vegetation in those areas and details the severity differences between large- and medium-sized fires. These details are key to helping fire managers make decisions about prescribed burns or about when and whether to let smaller fires keep burning.

“I’m interested in promoting better post-fire situations,” said the professor of forest ecology in the Department of Wildland Resources at Utah State, in Logan, Utah. “We can’t keep fire out of the landscape, but we might be able to take measures so that after the fire, maybe we still have a forest.”

MTBS for Large Fires

To predict fire behavior, a person needs to know the fire history, the area that burned and how hot the fire burned, information that is provided by MTBS, an interagency project between the USGS EROS Center and the U.S. Department of Agriculture (USDA) Forest Service Geospatial Technology and Applications Center. 

“MTBS is a fantastic program, and we’ve done a lot of work with their data,” Lutz said. MTBS uses Landsat data to map perimeters of fires 1,000 acres or larger in the western United States and 500 acres or larger in eastern States from 1984 to the present. MTBS also uses color-coding to track the severity of the fires in its maps, with dark red indicating areas where a fire burned the hottest, yellow for medium severity and dark green marking relatively unburnt areas.

MTBS is a good baseline for tracking fire history and trends for fires larger than 1,000 acres. Lutz’s study uses MTBS data to analyze large fires. “In Utah, if we’ve got a fire that’s more than 1,000 acres, we know what to do. We put it out,” he said. “No management question at all. Big fires need to be stopped so they don’t jump across our valley and ridge landscape or endanger things we want to protect.”

Managing Medium Fires

Data about medium-sized fires—100 to 1,000 acres—are the most meaningful for Utah fire management decision-making purposes. Utah has more medium-sized fires and those are the ones that are sometimes better left to burn. The goal is to identify “good fire.” For the past century, the focus was on extinguishing all fires no matter the size. Today, the thought process for managing fires includes the option to let some of them burn to clear out old and dry vegetation that has a higher likelihood of catching fire. Fire managers also set planned or prescribed burns when conditions are right. “We have a lot of forests that haven’t burned for a while,” Lutz said. “We have a lot of fuel build-up for various reasons, and the more fire we can put on the ground in a relatively managed way, the healthier the state’s forests are going to be.” 

Media

Sources/Usage: Public Domain. View Media Details

The initial effort to find all fires 100-1,000 acres since 1984 required some creativity, scouring federal and state fire databases and reading news reports. Reviewing Landsat scenes revealed more fire scars than the team even knew about, spurring more investigations to verify the existence of fires. The National Agriculture Imagery Program (NAIP) data comes at a higher resolution than Landsat and was used to help confirm the accuracy of the fire perimeters. 

Once the fire perimeters were established, Lutz used familiar methods from MTBS to gauge severity: differenced normalized burn ratio (dNBR). With dNBR, a pre-fire Landsat image is compared to a post-fire image. The contrast shows either an increase or a decrease in healthy vegetation and soil properties across the fire’s footprint. The less green, the more severe the burn.

Lutz’s study also looked at the severity of each fire. “Burn severity gets tossed around as a term a lot. What we’re really looking at is the overall effect on the ground,” he said. “How many trees died? How many shrubs died? Was the soil affected by the heat of the fire?”

The “ideal” fire would have patches of green, known as “refugia,” scattered among areas of yellow and red on the map. The unburnt vegetation in refugia provides the seeds needed for regrowth after the fire. “You don’t want big, extensive red,” Lutz said. “How can we promote the existence of these refugia that are going to persist through fire? Because that’s going to preserve a lot of forest values and accelerate recovery.”

Adding Vegetation Data from LANDFIRE

Media

Sources/Usage: Public Domain. View Media Details

The last piece of the puzzle to effectively manage fire was studying how fire behaved differently in the wide variety of different vegetation types in Utah. Lutz turned to the already-existing vegetation maps provided by the multi-agency project LANDFIRE, with many staff hosted at EROS.  

“We don’t have Sierra Nevada mixed conifer here in Utah, so we need to know what to expect if a spruce-fir forest catches fire,” he said. “We have a lot of aspen. At lower elevations, pinyon-juniper. In the wilderness-urban interface, or WUI, we have a lot of maples and oaks.”

Armed with data derived from Landsat, Lutz now was able to compare how large fires and medium-sized fires behaved over a nearly 40-year span in different types of vegetation—where they differed and where they stayed the same. 

The Fire Regimes of Utah study showed some surprising results. For example, medium-sized fires were more severe than large fires in sagebrush, shrublands and grasslands. But forests showed the opposite: Large fires had higher severity in forests. Among tree species, Lutz said, “in some forest types, you can just download the MTBS data, and whether it’s a big one or a small one, it’s going to behave the same. In other forest types, small is different. So, we need to consider size.”

Media

Sources/Usage: Public Domain. View Media Details

Helping Set Fire Norms

Combining these datasets together within Landsat’s long history of Earth observation is the ultimate key to fire behavior and fire management. 

“It’s that uniform long period of time that helps us,” Lutz said. “Fire is really variable year to year, just like the weather. The concept we want is fire normals. For example, if a fire starts in Douglas-fir, we could say, well, in this weather, it’s probably not going to be a very large fire if we left it alone. This is what we’ve experienced in the past 30 years. So just more information about what is typical. Then that’s our baseline.”

The history makes it possible to see real changes in fire trends as well. “We can see where it’s getting a little friskier,” he said. “The fires are getting a little hotter there. There’s more of them. They’re getting bigger. They’re getting more severe. But to make that kind of determination, we’ve got to have that uniform timeline.”

Maintaining the Database

Lutz’s study wasn’t a one-time exercise in data gathering. He’s currently collecting 2023 data and will continue to maintain a database to share with interested agencies.

Media

Sources/Usage: Public Domain. View Media Details

To make sure the information is as easy to acquire as possible, Lutz published an open access paper, and all of the summary data was included online for policy makers, fire managers and researchers. “The actual satellite data and analysis of all the fires is available to those who need it as well,” Lutz said, but it’s a large amount of data to download: 2 terabytes.

The target audience for Lutz’s data would include people like Tom Thompson, the geographic information system (GIS) IT specialist for the Utah Division of Forestry, Fire and State Lands. His department recently used MTBS data to recreate a 40-year Utah fire history, so he is interested in learning more about Lutz’s data. 

“We don’t really track severity in our current fire history, but we’re always looking at different attributes,” Thompson said. “MTBS has definitely been tried and true, and I’d be happy to use this new dataset if it is formatted in a usable way for our purposes.”

Lutz’s comprehensive Utah study is one of many efforts in the United States to use Landsat data to understand and map their own fire history and fire behavior. Josh Picotte, a USGS fire science biologist at EROS, works with the satellite data and various fire science projects every day. “Both MTBS and LANDFIRE provide foundational datasets for researchers to examine their own unique questions,” he said.

Lutz’s combination of these datasets now provides Utah fire fighters with fire history in terms of size and severity to help make more informed forest management decisions. “People who do this on the ground are very experienced,” he said. “We’re placing a huge load of responsibility on fire managers who are trying to implement prescribed fires. We just want them to have a little extra information so they can make the best decision they can.”