I’ve worked in the fire technology space for 5 years now, leading teams building products and projects helping fire agencies figure out where fires are likely to go and surfacing the information to them at just the right time to help them make well-informed, data-backed decisions all tracked for easy auditing later on in the process. I’ve learned some surprising things about fires and the fire industry. Here’s a collection of those insights.
Zombie fires are real
In the boreal forests of Canada, wildfires have found a way to survive the winter. They don’t just die out when the snow falls. Instead, they burrow underground.
During the summer, these fires burn deep into the layers of peat or travel along root systems. They smolder there all winter, insulated by the snowpack abov…
I’ve worked in the fire technology space for 5 years now, leading teams building products and projects helping fire agencies figure out where fires are likely to go and surfacing the information to them at just the right time to help them make well-informed, data-backed decisions all tracked for easy auditing later on in the process. I’ve learned some surprising things about fires and the fire industry. Here’s a collection of those insights.
Zombie fires are real
In the boreal forests of Canada, wildfires have found a way to survive the winter. They don’t just die out when the snow falls. Instead, they burrow underground.
During the summer, these fires burn deep into the layers of peat or travel along root systems. They smolder there all winter, insulated by the snowpack above. When spring arrives and the snow melts, these "zombie fires" resurface. This can spark new wildfires weeks before the usual fire season even begins.
This is happening more and more often. After the record-breaking fire season of 2023, roughly 150 to 200 fires managed to overwinter across Western Canada. Some of these reignited and threatened communities like Fort Nelson in British Columbia.
Because they burn underground without any visible flames, they are nearly impossible to detect until they pop back up. They can persist at depths of dozens of feet which makes them incredibly difficult to extinguish.
- https://www.cbc.ca/news/climate/wildfires-zombie-fires-canada-bc-alberta-1.7119851
- https://www.cbc.ca/news/climate/zombie-fires-canada-wildfires-1.7207765
- https://en.wikipedia.org/wiki/2024_Canadian_wildfires
Fires can create the conditions for more fires via the carbon feedback loop
These fires represent a massive climate threat. Canada’s peatlands store more carbon than all of the country’s other ecosystems combined. When underground fires burn through this material, they release carbon that has accumulated over centuries.
The land might never fully recover. Instead, it creates a dangerous feedback loop. Warming temperatures dry out the peat, which fuels more fires. Those fires release vast amounts of carbon, which warms the atmosphere further.
For fire managers, this means one catastrophic season can now bleed directly into the next. It leads to repeat burning and potentially permanent damage to the forest.
- https://natural-resources.canada.ca/forest-forestry/wildland-fires/peatland-fires-carbon-emissions
- https://www.sciencedirect.com/science/article/pii/S0048969721002783
- https://eos.org/articles/feedback-loops-of-fire-activity-and-climate-change-in-canada
Lightning doesn’t always start fires straight away
When a bolt strikes a tree, the extreme heat (up to 27,000°C) can boil the water and sap inside. You might expect an immediate explosion of flames, but that isn’t always the case.
Sometimes the strike just smoulders in the organic material inside the tree. It sits there quietly for days or even weeks, waiting for the right conditions. This delayed ignition is known as a "holdover fire."
When the humidity eventually drops and the wind picks up, that dormant ember can finally flare into an active wildfire.
This makes things difficult for fire managers. They try to correlate historical lightning strike data with satellite hotspot detections to flag likely ignition points. The problem is that most lightning strikes never produce a sustained blaze. While the data helps narrow down the search, you still need to do some ground-truthing to confirm if there is actually a fire.
- https://essd.copernicus.org/articles/15/1151/2023/
- https://link.springer.com/article/10.1007/s11069-024-06741-8
What or Who causes wildfires?
Most wildfire ignitions trace back to people. If you look at the data for southeastern Australia between 1997 and 2009, 47 percent of fires with a known cause started accidentally. This includes things like discarded cigarettes, escaped burn-offs, campfires, and sparks from machinery. Another 40 percent were lit deliberately.
Lightning accounted for just 13 percent of ignitions.
This pattern holds up globally, where human activities cause roughly 85 percent of wildfires. But if you count by area burned rather than the sheer number of fires, the picture completely flips.
Research comparing wildfires in New South Wales and Victoria highlighted this contrast. Lightning-caused fires that destroyed houses had a median size of 26,314 hectares. Compare that to just 3,222 hectares for human-caused fires. That is an eightfold difference.
The reason for this gap is geography. Human ignitions tend to cluster near population centres. Because people are around, someone usually notices the smoke within minutes. Fire crews can arrive quickly while the fire is still small and manageable.
Lightning is different. It strikes remote ridgelines and deep bush, often during dry storms that spark multiple ignitions simultaneously. These fires can smoulder undetected for hours or even days. Once spotted, crews often have to hike or helicopter in to reach them. By the time suppression actually begins, the fire has usually established itself across terrain that makes containment difficult.
The 2019–20 Black Summer fires illustrated this starkly. The largest and most destructive blazes were lightning-ignited fires in remote forests that burned for weeks before eventually threatening populated areas.
Terrain has a dramatic effect on how fire spreads
Fire loves a hill. It moves faster when it is climbing, and the difference is pretty dramatic.
In Australia, fire agencies use a standard rule of thumb to calculate this. For every ten degrees of slope, the speed of the fire doubles. If a fire is moving at 5 kilometers per hour on flat ground, it will hit 10 km/h once it hits a 10-degree slope. On a 20-degree slope, you are looking at 20 km/h.
The physics here is surprisingly simple. When flames burn uphill, they lean into the slope. This puts them physically closer to the unburned fuel waiting just above them. The radiant heat acts like a pre-heat cycle, drying out the moisture in that vegetation before the flames even touch it.
You also get a chimney effect. Hot air naturally rises along the slope, feeding oxygen to the fire and carrying embers upward. The steeper the hill, the stronger this effect becomes. Lab tests show that once a slope passes 25 degrees, the acceleration is massive.
But going downhill is a different story. The fire loses all those advantages. The flames lean away from the fuel, and the radiant heat dissipates into the open air rather than pre-heating the scrub. Gravity is also working against the natural rise of hot gases.
Generally, downhill spread slows to about 60 percent of the speed on flat ground.
Building homes on steep slopes in fire prone areas is risky. If you have a house at the top of a 20-degree incline, it is facing a fire moving four times faster than it would on the flat. And it is hitting with much higher intensity.
Its raining men embers
Embers, rather than the main wall of flames, are responsible for the majority of property loss. These burning fragments of bark, twigs, and leaves account for 75 to 80 percent of property loss in Australian bushfires.
This process is called spotting. Embers rise in the superheated plume above a fire and catch the horizontal winds. They rain down ahead of the main front and ignite new fires. This creates a dangerous situation where firefighters can be trapped between blazes, or structures can be overwhelmed long before the actual fire front arrives.
The distances involved are significant. There have been reports of embers traveling up to 20 kilometers ahead of a fire front. During the 2009 Black Saturday fires at Kilmore East, spotfires were reported 30 to 40 kilometers downwind.
Australian eucalypts are notorious embers factories. Two bark types cause particular trouble: ribbon bark and stringybark.
Ribbon-barked trees, like the candlebark, produce strips of bark with long burn-out times. These can be transported up to 30 kilometers. Stringybark trees operate differently, producing massive amounts of smaller embers that travel shorter distances, usually up to 2 kilometres.
CSIRO researchers used a vertical wind tunnel to test this. They examined how long strips of burning ribbon bark from manna gum could sustain combustion while falling at terminal velocity. They found that tightly curled bark cylinders burned for more than seven minutes on average.
The maximum burnout time for internally convoluted cylinders reached 1,304 seconds. That is long enough to travel 20 kilometres if the wind is blowing at 60 km/h. The danger lies in the shape. As the bark dries, it curls longitudinally. This creates an insulated combustion chamber inside the cylinder that extends the burn time dramatically.
Once embers land, they find their way into structures through remarkably small openings. Direct ignition happens when embers enter properties through gaps, vents, and windows.
In Australia, evaporative air conditioners are a known weak point. The filter pads catch fire, and the flames spread straight into the roof space, destroying the home.
Ignition risks extend beyond gaps. Gutters clogged with dry leaves function as ignition trays. Timber decking, firewood stacks against walls, doormats, or outdoor furniture can all catch an ember and transfer the fire to the building.
Research from UNSW Canberra suggests that current models regarding ember transport may be incomplete. While science often assumes embers travel through the air, observation shows many actually roll and tumble along the ground. They accumulate in corners and against foundations where they eventually ignite. This intensity can overwhelm defenses. During the 2019 Currowan fire in NSW, a fire truck caught fire in an ember attack, though fortunately, the crew escaped.
- https://en.wikipedia.org/wiki/Ember_attack
- https://www.unsw.edu.au/canberra/our-research/research-centres-institutes/unsw-bushfire/understanding-bushfires/ember-storms-explained
- https://ecos.csiro.au/firespotting/
- https://www.researchgate.net/publication/320301566_Messmate_stringybark_bark_ignitability_and_burning_sustainability_in_relation_to_fragment_dimensions_hazard_score_and_time_since_fire
- https://www.publish.csiro.au/wf/wf15031
When fires are big enough, they start to create their own weather, and its usually pretty wild
Large fires do more than just respond to the weather. They actually create it.
When a bushfire burns hot enough, a column of superheated air punches through the lower atmosphere with updrafts exceeding 160 kilometres per hour. As that plume rises, it cools and expands. Moisture condenses into cloud. If the conditions align (unstable atmosphere, dry mid-levels, and sufficient heat release from below) that cloud keeps building. It can reach the upper troposphere or even the stratosphere, 15 kilometres or more above the ground.
Up in those frigid upper reaches, ice particles collide and build up electrical charge. Eventually, it discharges as lightning. The fire has effectively made its own thunderstorm. We call this a pyrocumulonimbus, or pyroCb.
These fire-generated storms are dangerous in ways ordinary thunderstorms aren’t. They produce lightning that can ignite new fires kilometres away from the original blaze. But unlike normal storms, the rain often evaporates before it hits the ground. The heat and dry air near the fire surface simply boil it away.
The physics get violent here. The updrafts loft burning embers to extraordinary heights and scatter them far downwind. Then you have downdrafts punching outward from the storm base. These generate erratic winds that can shift the fire’s direction without warning.
Fire tornadoes are not the work of science fiction
In extreme cases, the rotating updraft spawns fire tornadoes, yep that’s right fire-nados. The 2003 Canberra fires produced an F3-rated fire tornado. It was the first confirmed violent fire tornado on record.
During Black Summer, pyrocumulonimbus clouds over East Gippsland reached altitudes above 16 kilometres. Near Jingellic, a fire tornado from a pyroCb overturned two fire trucks, one weighing up to 12 tonnes. Tragically, a firefighter lost their life.
The frequency of these events is rising. Of the 144 pyroCb events recorded in Australia’s register since satellite monitoring began, 135 have occurred since 2003. We saw 45 during the 2019–20 fire season alone.
Researchers at the Bureau of Meteorology have developed a Pyrocumulonimbus Firepower Threshold tool to help forecasters predict when these storms might form. But prediction only goes so far when the fire itself is rewriting the weather.
Eyes in the sky, satellites are constantly watching the world in a science called Earth Observation
A constellation of satellites watches the Earth around the clock. The satellites are taking imagery in a multitude of bands of the light spectrum, both visible and infra-red/ultraviolet. These images are process to produce a dataset of "thermal anomalies," which are points where infrared radiation spikes above the background temperature of the landscape.
NASA’s FIRMS system processes data from MODIS and VIIRS sensors. These fly on polar-orbiting satellites and detect hotspots globally within three hours. In Australia, the Digital Earth Australia (DEA) Hotspots system takes it a step further. It combines those feeds with data from Japan’s Himawari-9 satellite. Because Himawari-9 is geostationary, it allows for updates every ten minutes during fire season, but the downside is it’s much coarser resolution.
But this thermal anomalies dataset detects heat, not fire. A thermal spike might be a bushfire racing through scrub. It could also be a gas flare at an oil refinery, a steel smelter, a volcanic vent, or even a large solar farm. While algorithms try to flag persistent industrial sources to reduce false positives, distinguishing fire types from a thermal signature alone is impossible. Analysts have to infer context from location and timing.
There are also physical limitations. Clouds block detection entirely. Furthermore, polar-orbiting satellites only pass over a specific point a few times a day. A fast-moving fire can ignite, spread, and burn out completely between observations. Consequently, these systems provide broad situational awareness rather than real-time tactical guidance. Emergency services treat hotspot data as just one input among many, never the sole basis for safety-of-life decisions.
Fire prediction is an imprecise art but good enough is the name of the game
Predicting where a wildfire will spread is a sophisticated discipline that bridges physical science and hard-earned field experience. It is the domain of Fire Behaviour Analysts (FBANs), highly skilled specialists who apply their knowledge to forecast the movement of an inherently chaotic force.
While modern software supports them by ingesting dozens of variables, from fuel type and vegetation curing to wind speed and elevation, the human expertise required to interpret these outputs is irreplaceable.
The same fire models don’t typically work in multiple regions given the difference in vegetation and climate
The Australian model In Australia, the science rests on work done in the 1960s by CSIRO scientist A.G. McArthur. After studying over 800 experimental fires, he developed the Forest Fire Danger Index. He combined temperature, humidity, wind speed, and drought factors into a single rating.
It was groundbreaking work, but it had a flaw. McArthur calibrated his system using low-intensity test fires. He set the catastrophic 1939 Black Friday fires as the theoretical maximum of 100. When the Black Saturday fires hit in 2009, the index exceeded 100 entirely. It revealed the limits of a system designed for conditions that no longer represent the extremes we face today.
These calculations now feed into Phoenix RapidFire. This is a simulation tool developed by the University of Melbourne and the Bushfire Cooperative Research Centre. It combines the danger indices with terrain mapping and weather forecasts to predict flame height, ember density, and spotting distance.
This is now being superseded by Spark Operational, a next-generation capability from CSIRO and AFAC that offers more flexibility across different landscapes.
The Canadian approach Canada built its system on even older foundations. The Canadian Forest Fire Danger Rating System evolved from field research starting in the 1930s. The modern version was formalized by C.E. Van Wagner in 1987.
The Canadian system is unique because it tracks fuel moisture across three specific timescales:
- Surface litter drying out in hours.
- Organic duff responding over days.
- Deep layers changing over weeks.
This allows them to estimate crown fire likelihood with high accuracy. This logic is codified in simulation tools like Prometheus and the open-source Wildfire Intelligence Simulation Engine (WISE). It has become the most widely applied system in the world, adapted for use from New Zealand to Portugal.
Models vs. reality All these systems share one major weakness. They model what fires should do based on averaged relationships from past observations.
They cannot account for the outliers. They cannot predict a micro-gust pushing embers kilometers ahead of the front. They cannot foresee a pyroconvective column generating its own weather system. And once crews start fighting the fire-cutting lines or dropping retardant, the predictions diverge even further from reality.
This is why agencies treat these outputs as scenarios rather than forecasts. They run multiple simulations and plan for the worst reasonable case. The software saves towns by focusing resources where they matter most, but every fire commander knows the models are just approximations. Ultimately, the fire will do what the fire will do.
Links & Whitepapers
Canadian Fire Danger Rating System
Van Wagner, C.E. (1987). Development and structure of the Canadian Forest Fire Weather Index System.
- Forestry Technical Report 35, Canadian Forestry Service, Ottawa
- PDF: https://cfs.nrcan.gc.ca/pubwarehouse/pdfs/19927.pdf
- The foundational document for the Fire Weather Index system, describing the six-component structure based on fuel moisture codes and fire behavior indices
Forestry Canada Fire Danger Group (1992). Development and structure of the Canadian Forest Fire Behaviour Prediction System.
- Information Report ST-X-3, Forestry Canada, Ottawa
- Repository: https://ostrnrcan-dostrncan.canada.ca/handle/1845/235421
- Documents the FBP System based on measurements from experimental fires across Canadian fuel types
Tymstra, C., Bryce, R.W., Wotton, B.M., Taylor, S.W., Armitage, O.B. (2010). Development and Structure of Prometheus: The Canadian Wildland Fire Growth Simulation Model.
- Information Report NOR-X-417, Natural Resources Canada, Northern Forestry Centre, Edmonton
- Government of Canada Publications: https://publications.gc.ca/site/eng/9.568367/publication.html
- CFS Publications: https://cfs.nrcan.gc.ca/publications?id=31775
- Documents the Prometheus fire growth model structure and vector propagation technique
WISE (Wildfire Intelligence Simulation Engine)
- Open-source successor to Prometheus
- FireGrowthModel.ca: https://firegrowthmodel.ca/
- FRAMES webinar: https://www.frames.gov/catalog/67109
Canada CFFDRS
- Natural Resources Canada CFFDRS page: https://natural-resources.canada.ca/forest-forestry/wildland-fires/canadian-forest-fire-danger-rating-system
- Canadian Wildland Fire Information System: https://cwfis.cfs.nrcan.gc.ca/background/summary/fdr
- R package (cffdrs): https://cran.r-project.org/web/packages/cffdrs/cffdrs.pdf
Australian Fire Danger Rating System
McArthur, A.G. (1967). Fire Behaviour in Eucalypt Forests.
- Forestry and Timber Bureau Leaflet 107, Commonwealth of Australia, Canberra
- The original empirical work underlying Australia’s fire danger rating
- Available via National Library of Australia: https://nla.gov.au/nla.cat-vn2275488
Noble, I.R., Bary, G.A.V., Gill, A.M. (1980). McArthur’s fire-danger meters expressed as equations.
- Australian Journal of Ecology 5: 201-203
- Converted McArthur’s circular slide-rule meters into mathematical equations
- PDF: https://courses.seas.harvard.edu/climate/eli/Courses/global-change-debates/Sources/Forest-fires/aridity-indices/Nobel-etal-1980-australian-forest-fire-danger-index.pdf
Phoenix RapidFire
- Developed by University of Melbourne, Bushfire CRC, and DELWP
- Fire Prediction Services: https://firepredictionservices.com.au/
- Knowledge base: https://firepredictions.atlassian.net/wiki/spaces/PH/overview
CSIRO Spark
- Next-generation national bushfire simulation capability
- CSIRO Spark homepage: https://research.csiro.au/spark/
- Documentation: https://research.csiro.au/spark/resources/documentation/
- User guide PDF: https://research.csiro.au/static/spark/Spark_applications_user_guide_v112.pdf
- Technical documentation: https://research.csiro.au/static/spark/docs/docspark.html
FBAns: The people who read fire
Software models can crunch the numbers, but they cannot make the final call. That responsibility falls to Fire Behaviour Analysts, or FBANs and they are some of the smartest people you’ll meet.
These specialists sit right in the middle of incident management teams. They collect local weather observations, go out to assess fuel conditions on the ground, and run simulations to predict where a fire is heading.
Their job is to take the semi-processed and raw data from prediction software (things like probability maps, arrival times, and ember projections), validate it and translate it into something useful. Incident controllers rely on this intelligence to decide where to position crews, which communities to warn, and when to pull firefighters back from dangerous positions.
The path to expertise
Becoming an FBAN requires years of accumulated experience on the fireground. In Australia, formal training for this role began in 2007. Candidates start with e-learning modules covering fuel assessment, weather interpretation, and fire behavior calculations before moving on to practical seminars run through AFAC.
The Emergency Management Professionalisation Scheme currently offers two credential levels:
- Registered Fire Behaviour Analyst: For those proficient in Level 2 incidents of moderate complexity.
- Certified Fire Behaviour Analyst: For experts capable of handling Level 3 incidents of high impact and complexity.
It is a demanding role. A 2014 CSIRO review of Australian capacity noted that technical competence is just the baseline. The real skill lies in the ability to communicate complex fire behavior information under intense operational pressure. They have to translate scientific uncertainty into clear, confident guidance when lives are on the line.
The arsonist’s paradox
Fire agencies live with an uncomfortable truth. The same predictive intelligence that helps firefighters save lives could, in the wrong hands, help an arsonist take them.
We make fire danger ratings, fuel moisture maps, and wind forecasts publicly available because communities need that information to survive. If you don’t know the danger is coming, you can’t prepare or evacuate. But a sophisticated arsonist could use that exact same data to pick the ignition points, timing, and conditions that would cause maximum destruction.
Agencies are well aware of this tension. It is why some operational tools are kept strictly within professional circles. You generally need credentials to access the precise algorithms behind spread simulators or real-time fuel state mapping.
But the fundamental trade-off remains. You cannot warn a community about catastrophic fire danger without also signaling to malicious actors that conditions are ripe.
In practice, fire investigators have found that most arson is opportunistic rather than calculated. It is usually a disturbed individual with a lighter, not a technician with a laptop. While the few cases where arsonists have demonstrated tactical awareness are deeply troubling, they haven’t shifted the consensus. We still believe that an informed community is a safer community, even knowing the information cuts both ways.
Ever heard of Dry Firefighting? Me neither until I worked with the people who do it
When most people picture firefighting, they imagine water. But in the Australian bush, water is often impractical. Hoses cannot run through kilometers of rugged terrain, and no tanker carries enough to douse a fire front stretching for miles. So crews rely on "dry firefighting." Instead of putting the fire out with liquid, they remove the fuel before the flames can reach it.
The goal is to construct control lines. These are strips of ground scraped clean of vegetation down to bare mineral earth. On accessible terrain, bulldozers and graders do the heavy lifting by pushing aside trees and undergrowth to create a barrier the fire cannot cross. Some agencies now use rake attachments instead of standard blades to keep topsoil in place, but in rugged country inaccessible to machinery, specialist crews do the same work by hand using rakehoes and chainsaws.
It is brutal, dangerous work. Heavy machinery operating near active fire faces rollovers on steep ground, falling trees, radiant heat, and the risk of being overrun if conditions change. Modern fire dozers carry protective systems like heat-resistant glass, and rollover bars. But even with all that technology, the work remains inherently hazardous.
Aerial firefighting is an amazing feat but there are some environmental considerations
Aerial firefighting relies on two main types of suppressants. You have the short-term options, which are essentially detergent-based foams that coat vegetation to slow down evaporation. They work well enough, but their effectiveness drops off when you are up against high-intensity fires.
Then there are the long-term retardants. These are the bright red slurries you often see dropped from planes. They are a mix of fertilizer (ammonium phosphate), thickeners, and corrosion inhibitors. Unlike the foams, these stay effective even after the water in the mix evaporates because they chemically alter how the fuel burns.
On paper, they look fairly safe. After the 2019–20 bushfires, the NSW Environment Protection Authority tested these retardants and concluded the environmental impact was minimal. There are also no known long-term health effects for humans.
But if you dig into the peer-reviewed research, the story becomes a bit more complex.
The main issue is the fertilizer. Australia has incredibly old, phosphorus-impoverished soils. Our native ecosystems are actually adapted to these low-nutrient conditions. When you dump a fertilizer-based retardant on them, you shock the system.
Studies on Victorian heathlands showed that Phos-Chek caused direct mortality in key native species like Banksia and Leptospermum. It effectively poisons the plants that aren’t built to handle that much nutrient load.
It also creates a long-term problem with recovery. Because the soil is now fertilized, it becomes the perfect breeding ground for invasive weeds which outcompete the natives. This can actually increase fuel loads for future fires. The chemicals can also reduce seed viability in native plants, making it even harder for the bush to recover naturally.
There is also the water to worry about. Foams are moderately toxic to aquatic life, which is why pilots generally try to avoid dropping near waterways.
In an emergency, we often have to use imperfect tools. If a fire is threatening a town, the environmental cost of the retardant is a secondary concern. But we need to be realistic about that cost. The science suggests the impact is significant, particularly in our sensitive ecosystems and at the volumes we are now seeing.
Mountain Waves are an interesting and terrifying effect
Mountain waves are a deadly hazard for aerial firefighters. Like water flowing over a submerged rock, air flowing over a ridge creates oscillating waves and violent "rotors" downstream.
These rotors can push air down at 5,000 feet per minute. That is faster than most planes can climb. Air tankers are particularly vulnerable because they fly low, often just 200 feet off the ground.
The risk is highest right after a drop. The aircraft is lighter, but it is flying through hot, thin air where performance is already reduced. If it hits a downdraft, it can be slammed into the terrain instantly.
This exact scenario happened to a C-130 Hercules during Australia’s Black Summer fires in 2020. The crash killed all three American crew members. The Australian Transport Safety Bureau investigation found that strong gusting winds and mountain wave activity producing turbulence were both forecast and present at the drop site. After completing a partial retardant drop, the aircraft likely aerodynamically stalled while flying through hazardous conditions that included windshear and an increasing tailwind.
- https://www.atsb.gov.au/media/news-items/2022/large-air-tanker-accident
- https://www.atsb.gov.au/publications/2005/mountain_wave_turbulence
Microbursts are even more nightmare fuel for aviators
A microburst is a violent, localized downdraft that often strikes without warning. It starts inside a thunderstorm when a column of cold air becomes denser than its surroundings. This column descends rapidly toward the ground and accelerates as it falls.
When that column hits the surface, the wind spreads out in all directions. This creates damaging outflow winds that can exceed 100 knots. The entire event happens within an area less than four kilometres across and usually lasts less than five minutes.
For a pilot, the sequence of events is a trap. You first encounter a strong headwind as you fly into the outflow. This increases your lift and airspeed, which might tempt you to reduce power to maintain your glide path.
Then the trap snaps shut. You enter the downdraft, which slams the aircraft toward the ground. Simultaneously, the wind shifts to a tailwind. This steals your airspeed and your ability to climb away.
The numbers are pretty scary. Downdrafts can force an aircraft to sink at 6,000 feet per minute. The horizontal wind shear across the microburst can reach 90 knots. Since 1943, these events have killed more than 1,400 people worldwide.
Firefighting aircraft are particularly vulnerable because they operate exactly where microbursts are most lethal. They fly at low altitude, with slow airspeeds and heavy loads, often near the thunderstorms that accompany fire weather.
On 1 July 2012, a MAFFS-equipped C-130 from the North Carolina Air National Guard flew into a microburst while fighting the White Draw Fire in South Dakota. The aircraft was forced into the ground. Four aircrew were killed, and two survived with injuries.
The investigation revealed that the aircraft had experienced unexplained airspeed loss during a previous drop, despite running at full power. A lead plane flying just half a mile ahead was pushed to within 10 feet of the ground by the same microburst. It was the first fatal crash in the 40-year history of the MAFFS program.
These events come in two forms, which makes them difficult to manage during fire operations.
Wet microbursts occur within active thunderstorms. They are generally visible and relatively easy to avoid. Dry microbursts are harder to spot. They form in clear air beneath virga, which is that wispy curtain of rain that evaporates before it reaches the ground.
Without precipitation reaching the surface, a dry microburst is invisible unless the winds kick up a dust ring on the ground. A tanker crew might see nothing but clear sky and a distant rain shaft, only to fly directly into an invisible wall of descending wind. These dry events are common in the Rockies and other mountainous areas of the western United States during the summer, which coincides exactly with the peak of fire season.
- https://wildfiretoday.com/air-force-report-says-micorburst-caused-crash-of-maffs-air-tanker/
- https://www.dvidshub.net/news/97848/investigation-report-details-causes-fire-fighting-c-130-accident
- https://journals.ametsoc.org/view/journals/bams/103/12/BAMS-D-22-0038.1.xml
- https://pilotinstitute.com/microbursts/
Not all wildfires need to be extinguished
You would be forgiven for thinking all wildfires need to be suppressed, but many ecosystems require fire to encourage reproduction. Take the Australian Banksia which store seeds in woody follicles sealed with resin that only melts in the heat of a passing fire. Similarly the Logepole Pines in Yellowstone require fire to melt the resin in their cones to release the seeds. Even smoke has been proven to have a positive effect on seed germination and plant growth when exposed to plant-derived smoke. So the goal isn’t really to extinguish all wildfires, in the case of NSWRFS its goal is to protect people, property and environment.
NSWRFS Priorities of Operations
| Priority | Description |
|---|---|
| Overriding Priority | Firefighter safety |
| First Priority | Protect people |
| Second Priority | Protect property |
| Third Priority | Protect the Environment |
| Fourth Priority | Help restore normality |