New burn-detection-modeling system in development to help identify potential wildfire threats in Texas

Writer: Blair Fannin, 979-845-2259, b-fannin@tamu.edu

Contacts: Dr. Richard Conner, 979-845-7456, jrc@tamu.edu

Wayne Hamilton, 979-845-5589, wt-hamilton@tamu.edu

Jay Angerer, 254-774-6053, jangerer@tamu.edu

Tom Spencer, 979-458-7331, tspencer@tfs.tamu.edu

 

COLLEGE STATION – A burn-risk-detection modeling system in development by a consortium of Texas A&M University System researchers will help predict potential wildfire threats throughout the Lone Star State, according to its developers.

The modeling system will be used by the Texas A&M Forest Service. Researchers with the Center for Natural Resource Information Technology, part of Texas A&M AgriLife Research, are developing the modeling program. One of those researchers, Dr. Richard Conner, also an AgriLife Research economist and professor in the department of agricultural economics at Texas A&M, said the system is a modification of several modeling projects developed a decade ago.

“This current modeling system measures the amount of forage fuel load in a county and is used to predict potential fire danger,” he said.

The modeling system will provide real-time information on fuel loads using vegetation and National Oceanic and Atmospheric Administration weather information for regions susceptible to wildfire.

“In the wildfire predictive services arena, one of the hardest things to get a handle on is herbaceous fuel load across the state,” said Tom Spencer, head of fire predictive services, Texas A&M Forest Service. “It’s challenging to determine the condition and amount of it. There’s no good way to do that through remote sensing. It’s always been the case where someone has to physically go out and look, then make a judgment call.

“This project will help determine if it is possible or not. We think it is, but we still need to determine if the science supports it. Overall, we are looking forward to seeing how this helps us understand potential fire season severity, which will help us assist local governments to better plan ahead. It’s a huge deal if this works out.”

Wayne Hamilton, Texas A&M AgriLife Research range scientist, examines vegetation on rangeland as part of a burn detection modeling system project. (Texas A&M AgriLife Research photo)

Wayne Hamilton, Texas A&M AgriLife Research range scientist, examines vegetation on rangeland as part of a burn detection modeling system project. (Texas A&M AgriLife Research photo)

The project has received $125,000 in funding by a federal fire plan grant. So far, the project has been implemented in three counties – Stephens, Palo Pinto and Jack. In these counties, researchers have identified major plant communities and developed field monitoring sites to be included as part of an overall web interface monitoring system.

The team members working on the project include Wayne Hamilton, AgriLife Research range scientist, Jay Angerer, assistant professor at the Blacklands Research and Extension Center in Temple, Ed Rhodes, a research associate at Blacklands, and Jason Jones, an assistant research scientist with the department of ecosystem science and management at Texas A&M.

“We will delineate land areas into ecological sites for an inventory to be used in the modeling system,” Hamilton said. “Ecological sites are areas of the landscape that produce similar kinds and proportions of plant species and total annual yield. This will allow us to expand our field sampling information across similar areas (ecological sites) and save time and costs in the inventory process.”

Ecological sites also provide “state and transition models” that help identify plant communities and changes likely to occur from management inputs, he said.

“What we are hoping to do is to provide vegetation information that the Forest Service  can use to monitor potential wildlife threats across regions of the state.”

Angerer and the center’s team have developed the simulation model framework to run the monitoring system. It includes data such as types of vegetative plants growing in a county or region, soil data and historical weather data.

“We are also using NOAA data with the other information on soils and plant communities we have collected to help predict how much vegetation is growing in a particular area,” he said.

Angerer said vegetation samples are clipped to help calibrate the model. A similar system was developed during a research project at Fort Hood. The collected data, along with remote-sensing information, will be used to identify similar sites in the county so that the entire county can be monitored.

“Once we get the model working, the model will provide a base view of vegetative characteristics of a particular site, and can use this to see how many similar sites we have in an area,” he said. “We then can model these points and build landscape maps of potential fire threats.”

The plant community database is built by taking measurements of plant cover and clippings of vegetation in the field.

“We take the basal area of grasses and measure the canopy cover of shrubs,” Angerer said.

The project uses the PHYGROW modeling system, which computes grass, herb and shrub growth, forage consumption by livestock, and hydrologic processes. The project also uses the Burning Risk Assessment Support System, or BRASS, which is a decision support tool that provides a continuous means for land managers to assess vegetation and weather to support decisions related to prescribed burning and/or the risk of wildfire by utilizing near real-time weather conditions and fuel loads.

  The PHYGROW model was first coded in 1990 and has undergone many enhancements since that time. The model’s original computation algorithms are a mixture of formulas adapted from other plant growth models, as well as biological relationships from grass-tiller-level research and livestock dietary selection conducted at Texas A&M University.

“The PHYGROW model is capable of simulating growth of multiple species of plants subject to selective grazing by multiple animals on a soil with multiple layers for indefinite periods of time,” Conner said. “The model is designed to be integrated with a wide variety of weather databases, vegetation databases and stocking rule databases, and provides output for a wide variety of data sources and formats including all relational databases.”

BRASS, they explained, is a web-based interface that allows users to examine fire risk for points using an interactive map interface. Simulation modeling of rangeland forage and grazing dates back to the late Dr. Jerry Stuth, an AgriLife Research scientist, who, working with Conner and Hamilton, started the initial Ranching Systems Group seeking research funds for information technology research.

The new modeling tool will aid Forest Service personnel in tackling the challenge of changing range conditions throughout Texas, especially since land has changed ownership through the years and traditional management practices vary.

“Some land is not grazed as heavily as a tract say next door where there may not be any cattle on the property,” Conner said. “There’s fuel build up there and an increasing risk of wildfires. There’s more risk than there used to be, due to people having more land and using it for purposes (other)  than cattle production to make money.”

For more information about the Center for Natural Resource Information Technology, visit http://cnrit.tamu.edu/.

-30-

Print Friendly
Share on FacebookTweet about this on TwitterShare on Google+Pin on PinterestShare on Tumblr