Delivery alert

There may be an issue with the delivery of your newspaper. This alert will expire at NaN. Click here for more info.

Recover password

Computer modeling helps us learn to live with wildland fires

SANTA FE, N.M. — Wildland fires play an important role in many ecosystems, yet in the western United States, land managers have spent a century excluding it from the landscape. The resulting overgrown forests, along with hot and dry conditions, have changed the nature of fires when they do happen, making them more intense and more destructive. Figuring how best to respond is important for the health of our forests, the safety of nearby communities and the well-being of firefighters on the job – and it’s a task that can now draw on some of the most powerful computers in the world.

Safely reintroducing fire and enabling decision-makers to respond more effectively to wild blazes requires anticipating how fire will interact with the mountains and canyons, dynamic winds and mixed vegetation of the West. In the Southeast, where terrain is less of a factor, very active prescribed-fire programs let fire play its role in ecosystem management without endangering people or infrastructure. However, the dense vegetation in this region requires burning every few years, forcing prescribed fire practitioners to increase their understanding of “adequate but safe” conditions for burning and be more efficient when they do burn.

In all these cases, land managers and firefighting agencies need much more information about how fires behave, information that has been lacking.

Since wildfires are not new, why don’t we completely understand them? First, the questions being asked four decades ago were focused on fire prevention, whereas now they are about dealing with inevitable wildfires and ecosystem health, which have become more critical issues after a century of fire exclusion. Furthermore, wildfire behavior results from a baffling, complex interaction between fire, surrounding winds, vegetation and terrain. The vast number of fire scenarios prevents sufficiently measuring fire behavior without the help of sophisticated computer models that fill in the gaps or extrapolate into conditions outside of observations. Until recently, such models capturing the two-way feedback between the combustion processes and their surroundings have been impossible.

Not any more. Los Alamos National Laboratory has a long history of bringing its unique capabilities in physics, computational modeling and high-performance computing to just this kind of multidisciplinary problem as part of its national security mission work. Recent advances in high-performance computing at Los Alamos are enabling scientists to study these feedbacks in collaboration with various partners.

With the U.S. Forest Service, the lab is using a tool called FIRETEC to simulate the fire/atmosphere interaction that controls fire behavior, from low-intensity fires under marginal conditions to catastrophic wildfires – two extremes where our ability to predict fire behavior is least developed. This includes addressing key questions about both prescribed fire tactics and the fundamentals of fire behavior responses to terrain, fuels and wind conditions.

In the Southeast, in collaboration with the Air Force Wildland Fire Center and the Tall Timbers Research Station, Los Alamos is applying its supercomputing power to prescribed fire events under various wind and vegetation conditions so researchers can determine the optimal ignition strategy for managing ecosystems with fire.

This research can illustrate how prescribed fire can help ecosystems regain their fire adaptedness or explain why larger numbers of ignitions during a prescribed fire can actually decrease fire intensity and reduce tree mortality. It can also help tailor burns to fine-tune fire and smoke behavior, particularly in tricky conditions with light winds and light fuels.

On the other end of the spectrum, Los Alamos is supporting a Forest Service initiative to improve decision-making about fighting fires in complex terrain. A majority of the recent tragic western wildfires involved unexpected fire behavior. It is generally accepted that fires move faster up slopes than on flat ground, but the fire community still struggles to reliably predict sudden accelerations driven by the interaction among fire, weather and terrain. By developing a better understanding of the fundamental drivers of this behavior, this research intends to identify conditions underlying rapid changes in fire behavior and high-risk scenarios.

That kind of information could drastically improve how the wildland fire community predicts fire behavior so they can effectively manage prescribed fires and safely fight the wild ones. To that end, the Lab is working to provide the Forest Service information to develop next-generation computer tools that will run on a laptop instead of a supercomputer. This next-generation tool would enable large ensembles or bundles of simulations to be run for potential conditions and would take much of the guesswork out of anticipating the fire’s next move. Then incident fire commanders could make crucial decisions about managing a controlled burn or fighting a wildfire while it is happening.

That would help preserve the health of our forests, spare homes and even towns from conflagrations in wildland areas, and protect the lives of firefighters who risk everything against their most erratic adversary.

Rod Linn is a senior scientist at Los Alamos National Laboratory who applies computational physics to atmospheric phenomena. This is the first in a series of Lab Science columns provided by LANL.


Albuquerque Journal and its reporters are committed to telling the stories of our community.

• Do you have a story about how coronavirus has affected you, your family or your business? Do you have a question you want someone to try to answer for you? What issues related to the topic would you like to see covered? Or do you have a bright spot you want to share in these troubling times?
   We want to hear from you. Please email or Contact the writer.