Fighting Wildfires with Computer Models

Los Alamos National Lab
5 min readSep 18, 2019


By Rodman Linn, J. Kevin Hiers

Many ecosystems in the United States developed in the presence of fire. Over millennia, flames have regularly swept across landscapes, often at smaller scales and with lower intensity than today’s forest fires. They cleared out excess fuels such as smaller trees, fallen wood and leaf litter in a self-regulating process that fragmented the landscape, limiting the size of fires and keeping the forest healthy. In many western forests, this resulted in widely spaced trees, thinner understory and less combustible litter on the ground than we commonly see now.

But human intervention has interrupted that natural process, allowing excess fuels to accumulate — and those fuels have got to go one way or another. For many years in the southeastern United States, land managers have engineered controllable, prescribed burns to remove a majority of the accumulated fuels on the ground; a significant amount in the forest midstory and some in the canopy.

Sometimes, however, prescribed burns go badly wrong.

It happened in Bandelier National Monument in New Mexico in 2000, when a prescribed burn blew up into a megafire that torched more than 47,000 acres in the Jemez Mountains and more than 230 homes in the nearby town of Los Alamos, displacing more than 400 people.*

More recently, huge fires in California and Tennessee have focused long-overdue attention on how best to use prescribed fire to prevent such conflagrations. And while intentionally setting a controlled fire clears out excess fuels and rebalances ecosystems, planning and successfully carrying out this strategy can be a tricky business. That’s especially true in the complex mountain-and-canyon terrain of the West, where more and more people live near wild lands. And yet, if fuel loads are not kept in check, we can expect more catastrophic fires — huge, immensely destructive and deadly.

These fires are extremely sensitive to even small changes in conditions — a sudden wind gust, for example, or a patch of fuel exposed for longer than usual the drying effects of the sun. To overcome the challenges of these more marginal burning conditions, fire crews adjust the patterns and rates of ignition based on wind and fuels. For instance, a crew might ignite five parallel lines perpendicular to the wind. Things can get complicated fast: Multiple parallel fire lines burn differently than a single line, as the swirling gases converge. Another consideration is whether the smoke will smother a nearby community.

Experience with previous free-burning wildfires can help fire managers decide how to proceed, but since every fire is different, this is an imperfect solution This is where physics-based fire modeling can make a big difference. Modeling allows fire managers to simulate prescribed fires in advance, so that when crews start laying down flames from drip torches across the landscape, they can confidently set the right fire at the right time. They want just enough fire to sustain itself but not so much that it gets out of control. The FIRETEC modeling tool, developed at Los Alamos National Laboratory, leverages fluid-dynamics research originally developed for national security science and uses physics to represent the critical interactions among the multiple ignitions of a prescribed burn in very complex terrain.

Based on specific details about ignition patterns, terrain, weather, atmospheric conditions, fuel loads, vegetation and more, FIRETEC simulates the buoyant rise of air, the way the fire draws in competing drafts of air, mutual influence among fire lines, and interactions among the gases of the fire and the fuel loads, terrain, atmosphere and so on. FIRETEC also clarifies counterintuitive interactions, such as how fires spread faster at the surface under less canopy coverage because less aerodynamic drag from the canopy allows more wind to fan the flames.

For the first time, prescribed fire managers can use the equivalent of a flight simulator to understand and plan for prescribed burns and the real complexities that drive fire outcomes. Under the direction of the U.S. Forest Service, Eglin Air Force Base and Tall Timbers Research Station, FIRETEC is exploring how prescribed fires respond to ignition patterns and rates in fire-dependent southeastern forests. Practitioners are using these simulations to develop training materials and explain what they know through years of experience.

One drawback is that FIRETEC runs on supercomputers — not very handy for fieldwork. To bring similar modeling to laptops accessible to a wider range of people, Los Alamos and the U.S. Forest Service are developing QUIC-Fire, which uses simpler and faster-running algorithms. The hope is that QUIC-Fire will increase the two-way flow of information among scientists and practitioners, including experienced fire managers and those new to the field. Through a partnership with Tall Timbers, the Department of Defense, and the Department of the Interior, initial users are beginning to test drive QUIC-Fire to explore prescribed fire planning.

Prescribed burn practitioners and fire scientists alike still have much to learn. Perfecting prescribed fires to accomplish goals in land and forest management while keeping the fires under control and protecting people, towns and forests is no easy order. But physics-based tools explicitly representing the interaction between the fire and surrounding atmosphere provide new ways to investigate trade-offs in prescribed fire practices as the country begins implementing more of this kind of burning in increasingly complex scenarios.

Note: Los Alamos is developing QUIC-Fire in partnership with Scott Goodrick of the U.S. Forest Service.

Rodman Linn studies and models a wide range of atmospheric phenomena using computational fluid dynamics and other techniques at Los Alamos National Laboratory. Linn led the initial development of the FIRETEC computer program for predicting wildfire behavior and is currently co-leading the team developing QUIC-fire.

J. Kevin Hiers is both a fire scientist and a prescribed fire practitioner at Tall Timbers Research Station. Hiers leads experimental and modeling research efforts and is leading the charge to bring next-generation prescribed fire modeling tools, such as QUIC-Fire, to the prescribed fire community.

This Story first appeared in Scientific American



Los Alamos National Lab

Breaking stories from Los Alamos National Lab, a multidisciplinary research institution engaged in strategic science on behalf of national security.