In the world of aerial firefighting, one MSU professor is looking at how science can transform the multimillion dollar industry of wildfire retardant. And save lives in the process.
A Neptune Aviation BAe-146 jet drops fire retardant on a wildfire near Montana's Homestake Pass in 2021. New computer analysis methods are helping expose how the clouds of retardant disperse when deployed, leading to better chemical formulations and delivery designs. Photo by Robert Chaney
by Robert Chaney
Imagine a jet airplane zooming above a wildfire, releasing a cloud of zeros and ones.
That’s what Mark Owkes sees when he studies what happens to fire retardant when it’s dropped out of a Very Large Air Tanker. In his lab at Montana State University’s Department of Mechanical and Industrial Engineering, he studies the physics of fluid dynamics by computer simulation. The answers he finds could drive the next evolution of aerial firefighting.
In Forest Service Region 1, covering much of Greater Yellowstone, aerial firefighters released 581,613 gallons of retardant in 2023, the most recent year data is available. That was a relatively small amount, compared to California-centered Region 5’s 1.6 million gallons and Region 3’s 2 million gallons in Arizona and New Mexico.
In the first six months of that year, 19,358 wildfires burned 584,000 acres in the United States. Already in 2025, nearly 30,000 fires have scorched 1.2 million acres. Since the 1950s, the U.S. Forest Service has been dropping fire retardant — a mixture of fertilizer and water that sticks to vegetation insulating it from fire — to help tamp down blazes. But between stirring up the chemical recipe in a factory and expecting it to slow a fire in a forest, a lot of strange science takes place.
Currently, when the Forest Service wants to certify a new tanker plane or fire retardant formula, it orders what’s known in the field as a “Dixie cup test.” Someone, often a troop of volunteer Boy Scouts, places hundreds of little paper cups on an airfield. The test subject flies over and drops its load. Then the Boy Scouts collect all the cups, a technician measures how much liquid landed in each, and the results describe how effectively the plane or recipe covered the ground target.
“That’s state-of-the-art analysis of fire retardant and water drops,” Owkes told Mountain Journal. “It doesn’t lead the industry to push engineering boundaries. If you meet the bar for certification, that’s all you need to do.
“But the Dixie cup doesn’t tell what happens when the retardant hits the ground, or how it impacts the combustion process. It doesn’t tell anything about what happens to the fluid in the air.”
And all kinds of physics get involved between when the pilot pushes the tank release button and when the mud (as wildland firefighters call retardant) hits the dirt. First, that liquid passes through the nozzle of the tank holding it, which begins to atomize, or disperse it into droplets. Then it encounters the slipstream of the aircraft dumping it — a 120-mph blast of wind — which spreads the drops in a different way, sort of like shooting a can of spray paint next to a leaf blower.
"The Dixie cup [test] doesn’t tell what happens when the retardant hits the ground, or how it impacts the combustion process. It doesn’t tell anything about what happens to the fluid in the air.” – Mark Owkes, Associate Professor, Mechanical and Industrial Engineering, Montana State University
Next, gravity, crosswinds and radiant heat from the fire push and pull at the cloud. Time gets involved, depending on how high the aircraft was flying and how far the retardant has to fall. Finally, the stuff lands on trees, bushes, grass and dirt, where it’s expected to slow the wildfire’s progress or sap its energy. That’s when things get really weird, depending on what’s getting dropped (more on that shortly).
All these steps follow the laws of fluid dynamics. And if you can gather enough data about them and feed that into a big computer, you can learn what’s really happening after the pilot pushes the release button. Owkes uses MSU’s largest computing cluster, known as Tempest, to simulate how that cloud of retardant behaves.
“The average laptop has four or eight processors,” Owkes said. “The simulations we use need thousands of processors on Tempest. And even that won’t resolve everything, so we need to make smart decisions about how we divide the problem into thousands of smaller pieces that all talk to each other.”
Mark Owkes, associate professor of mechanical and industrial engineering, is using fluid dynamics and supercomputers to maximize efficiency in aerial firefighting. MSU Photo by Adrian Sanchez-Gonzalez
Owkes is a relative newcomer to the firefighting world. His previous work explored how gases and fluids flow by developing high-fidelity computer models. But he was mostly focused on relatively small things like fire sprinklers or paint sprayers.
About two-and-a-half years ago, one of his students made a presentation on the computer simulation algorithms they used. The audience included Dominique Legendre, a professor at the Toulouse Institute of Fluid Mechanics in France.
“He told my student, ‘Your work would be beneficial to the aerial firefighting community,” Owkes recalled. “He’d been working there for 15 years. Once Dominique talked to us, we set up simulations to study his fundamental problem.”
The partnership eventually resulted in Legendre making his own presentation at MSU’s Institute for Liquid Atomization and Spray Systems conference in May. He spoke on “Airtankers: A large-scale spraying system.”
“If you look at the footprint [of a retardant drop], you have a non-uniform deposit of liquid that reaches the ground,” Legendre told MSU News Service in May. “As a firefighter, you want a uniform deposit, with maximum concentration in the middle, and you want to be able to control the length and the width of the deposit.”
Past researchers tried to understand that objective by videoing air tankers as they made their drops, or flying drones over a drop zone to measure coverage. But those outside perspectives couldn’t reveal what was happening within the cloud of retardant or water as it dispersed. That’s where Owkes’ simulations come in.
“All these fluids are described by the same equations,” Owkes said. “Paint sprayers, fire sprinklers, waves breaking in the ocean; they’re all the same. It’s just the size of the problem — the boundary conditions.”
Aerial firefighting companies spend millions of dollars and years of research time perfecting their payload technology.
But the characteristics of a Bozeman-based Bridger Aerospace Super Scooper plane dropping a thousand gallons of water on a wildfire vary greatly from a Missoula-based Neptune Aviation BAe-146 jet dropping a thousand gallons of retardant. The mists they form are very different, as are the way those mists drift to the ground. And after landing, water drains away while retardant sticks around.
Water, when it’s hit by gravity and wind as it shoots out of an aerial tanker, turns to mist. So does fire retardant, which is typically a mixture of water and salts such as ammonium polyphosphate (a common fertilizer) that stick to plant material after the water evaporates. This protective layer slows the plant fiber’s ability to ignite, which in turn reduces wildfire spread. Other chemicals modify how the mixture drifts when sprayed, keeps the mixture evenly distributed in the tank or prevent it from corroding aircraft equipment.
Aerial firefighting companies spend millions of dollars and years of research time perfecting their payload technology. Retardant systems have proved exceptionally hard, with some companies like Neptune using built-in systems for its commercial jets and civilian contractor MAFFS developing a pressurized tank that can be inserted into various military aircraft.
In 2023, California congressmen David Valadao (R) and Jim Costa (D) tried to pass a bill to study “containerized aerial firefighting systems” that would use military bomber planes to drop containers of water or retardant on a wildfire. The bill was approved by the House of Representatives but did not make it through the Senate.
"All these fluids are described by the same equations. Paint sprayers, fire sprinklers, waves breaking in the ocean; they’re all the same. It’s just the size of the problem — the boundary conditions.” – Mark Owkes
Variables change again when a helicopter gets involved. Tanker planes lay strips along the edges of a fire. Helicopters usually drop their loads straight down, where the downward blast from their rotors mixes with the uprising heat from the fire.
“That’s super-important, and makes it much more challenging,” Owkes said. “Retardant is a shear-thinning fluid. As it goes through the atmosphere, the air shears it and breaks it up into droplets. But once it impacts the fuel, it wants to just sit there. It coats the fuel. It’s engineered to do that.”
“Shearing” is an important term here. In physics, it describes how something changes shape when acted upon by opposing forces. Water mist behaves according to certain Newtonian laws of physics, named for Issac Newton.
But there are also non-Newtonian fluids. Kevlar is one example, which bends like fabric in a bulletproof vest but turns solid when struck by a speeding bullet. Fire retardant is also a non-Newtonian fluid, and it behaves very differently from water. Owkes compared it to toothpaste, which flows out of the tube when you squeeze it, but sticks to the toothbrush when you leave it alone.
“We want to know when you make a large drop of water or retardant in the air, and it gets hit by high-speed turbulent flows, how does it break up?” Owkes said. “How do droplets get created? How does it deform when shearing out into the atmosphere? How does it coat fuels when it reaches the ground?”
Legendre had been working with wind tunnels, where he could release fluids into a 2- or 3-meter space and measure the effects.
“Here we drop tons of liquid out of a plane over 100 meters up,” Owkes said. “The scales are so massive, and the droplets are measured in microns. We create models that describe the whole breakup process.”
Running one of those models through Tempest can take two or three weeks of computing time. The resulting insights may lead to better designs for dispersal systems that ensure most of the load lands where it’s needed.
It might also inspire changes to the fluids themselves. Owkes has started exploring different non-Newtonian liquids that have the same dispersal and coating characteristics of traditional fire retardant but avoid some of its environmental toxicity problems. And he’s working on faster-running models that might help on-scene firefighters adjust to wind conditions and ground topography for better tanker drops.
“The history is you take a plane, put a tank in it, cut a hole in the bottom of the aircraft and see if you can put out a fire,” Owkes said. “That’s where the field started. Now I’m kind of hooked in this aerial firefighting world. It’s a super-fascinating problem.”
Mountain Journal is a nonprofit, public-interest journalism organization dedicated to covering the wildlife and wild lands of Greater Yellowstone. We take pride in our work, yet to keep bold, independent journalism free, we need the support of readers like you. Thank you.
About Robert Chaney
Robert Chaney grew up in western Montana and has spent most of his journalism career writing about the Rocky Mountain West, its people, and their environment. His book The Grizzly in the Driveway earned a 2021 Society of Environmental Journalists Rachel Carson Award. In Montana, Chaney has written, photographed, edited and managed for the Hungry Horse News, Bozeman Daily Chronicle, Missoulian and Montana Free Press. He studied political science at Macalester College and has won numerous awards for his writing and photography, including fellowships at the Nieman Foundation for Journalism at Harvard University and the National Evolutionary Science Center at Duke University.
