快猫短视频

MATT DUTKIEWICZ was feeling confident as he led his team of firefighters towards the cottage they had been sent to protect. The approaching bush fire was big but the ground around the cottage had been grazed bare and was hard and dry after months of drought. 鈥淚 said to the crew, 鈥業鈥檒l give you 20 bucks if you can get something to burn,'鈥� he recalls.

Moments later everything changed. 鈥淭he wind stopped. Then it started pulling in from the valley behind us. The hairs on the back of my neck stood up and I thought, 鈥楾here鈥檚 something not right here.鈥� Then this furnace just hit us. The trees in front of us snapped off and the mirrors of the truck smashed. The flames were 2 metres off the ground to about 8 metres in the air and were weird blues and oranges. It was like the air was on fire.鈥�

A few kilometres away, firefighter Neil Cooper was surrounded by burning vegetation and sought refuge in a bare field locally called the Oval. In theory it would be the safest place to sit out the blaze, but when he got there he saw a 鈥渂lanket of flame over the paddock about a metre high, shimmering like oil on water鈥�. It was like nothing he had seen in his 20 years on the job.

The that ravaged Australia鈥檚 capital in January 2003 surprised everyone. No one expected the blaze to reach the city and enter the suburbs with such force, killing four people and destroying almost 500 homes. But for some scientists the biggest mystery of all is what firefighters like Cooper and Dutkiewicz saw. Bare earth shouldn鈥檛 be able to sustain a fire of such ferocity at ground level, let alone metres up in the air, and flames from burning vegetation should be yellow or orange, not blue. As far as most scientists are concerned, what the firefighters saw was impossible.

In the past, observations like these have been dismissed as the result of extreme stress distorting people鈥檚 memory of events. But now a team of British and Australian researchers think they can explain not only the Canberra stories but also other reports of seemingly impossible fires around the world. If they are right, it will make fighting wildfires a whole lot safer.

The basic principles of bush fires are well understood. When vegetation is heated to about 430 掳C the plants鈥� carbohydrates, oils and other organic constituents rapidly vaporise and escape as a mixture of water vapour and flammable gases such as hydrogen and carbon monoxide. If the temperature rises past about 1200 掳C, and if enough oxygen is available, these 鈥減yrolysis products鈥� will ignite and burn with the characteristic yellowy-orange flame of wildfires. Assuming the fire radiates enough heat back onto unburnt vegetation, the pyrolysis process continues, releasing more gases that feed the flames and help the fire spread.

So how did the flames erupt across areas where there was no vegetation to release pyrolysis gases? John Dold, a combustion researcher at the University of Manchester, UK, has a new theory to explain the strange events observed during the Canberra fire. He says gases produced by pyrolysis must have somehow escaped from the fire front and accumulated away from the flames, forming an invisible and highly flammable mixture with the air. Flames can propagate so quickly through such a mixture that the changes in pressure create shock waves, leading to an explosion. For Dold, the snapped trees, blue flames and strange shimmering fire all add up to one thing 鈥� the advancing flames must have caused just such a violent combustion. 鈥淭he evidence seems quite strong that it did happen,鈥� he says.

Now he is trying to work out how. Explosive combustion of unburnt gases certainly happens in house fires, where furniture and other household items are heated in a closed space with limited oxygen. When hot pyrolysis products from furniture can鈥檛 find enough oxygen to burn in a flame, they rise and concentrate near the ceiling. If more oxygen enters the room, say from a door opening or someone smashing a window, the gases can ignite, causing a potentially deadly flashover. But could this happen in open air?

There is no shortage of accounts that seem to support the idea. After the Canberra fire, recovery teams reported finding sheep that had been singed only on their backs, and trees that were only burnt around their middles, from about a metre above the ground. At first, these snippets, plus witness testimonies, were all Dold had to go on. Then he heard about an extensive study of wildfires in the northern Rocky mountains.

In 1998 a team of wildfire researchers at the National Center for Atmospheric Research (NCAR) in Boulder, Colorado, the US Department of Agriculture鈥檚 Forest Service in Washington DC and NASA鈥檚 Ames Research Center in California set out to track wildfires using infrared and microwave sensors mounted on aircraft. These allowed them to 鈥渟ee鈥� through thick black smoke and watch how fires developed.

In one instance, near McDonald Creek in Glacier National Park, Montana, they recorded a sudden explosion from the fire front that shot forward 150 metres at 100 kilometres per hour, then receded almost as quickly, leaving small spot fires in its wake. The team dubbed it the 鈥溾�� and explained it as an explosive response to the extreme heat and heavy concentration of fuel. However, Dold thinks it is evidence of the kind of flashover observed by Dutkiewicz. 鈥淚 can only explain it in terms of vapours from the fire that have accumulated but not burned, and then the flame has propagated through them,鈥� he says.

鈥淎 sudden explosion from the fire shot forward 150 metres at 100 kilometres per hour鈥�

Janice Coen of NCAR鈥檚 Wildland Fire Initiative, one of the researchers who studied the McDonald Creek fire, disagrees. She has modelled the finger of death and other examples of the same phenomenon and believes that it can be explained by vortices generated by the heat of the fire.

As hot, burning air from a wildfire rises above the surrounding air, this upward motion can set the gas column rotating. If two of these vortices occur next to each other, she says, the swirls will be drawn together, which causes them to connect at the top and tip over. 鈥淭he rules governing fluids 鈥� including air 鈥� require that they tilt over into a ball and burst forward,鈥� Coen says. Although she agrees that this process is partly governed by the speed at which flames can travel through flammable gas, she doesn鈥檛 think the accumulation of escaped gases has anything to do with it.

鈥淢y analysis and modelling of fire behaviour shows that pyrolysis products would be drawn into the base of and lifted in the plume of the fire or recirculated from the plume down into the back of the fire. The one place that escaped pyrolysis products definitely are not is ahead of the fire where the finger goes,鈥� Coen says.

With no ground-level observations or witness reports, it is impossible to tell whether the McDonald Creek fire looked anything like the Canberra event, and it could be a different phenomenon altogether. All the same, Dold and his colleague Rodney Weber of the University of New South Wales and the Bushfire Cooperative Research Centre in Melbourne, Victoria, are confident that an open-air flashover is possible under certain conditions.

Although the exact location of the McDonald Creek finger of death is not clear, Dold and Weber know exactly where Dutkiewicz and Cooper were on the day of the Canberra fire, and they believe the terrain gives important clues to what could have happened. 鈥淢att and Neil were both uphill from an incised river valley,鈥� says Weber. 鈥淚f oxygen was cut off from the fire as it burned in the valley, gases would have accumulated. Then, if they were blown up the hill, they could ignite.鈥�

It is well known that as a fire burns up a slope it leans forward, rather than rising vertically. If the fire is burning up a confined space, such as a canyon or valley, then air can only be drawn into the fire from above. Burning can only occur where there are supplies of both oxygen and pyrolysis products. Oxygen below the flames is quickly used up, leaving an accumulation of pyrolysis products trapped between the ground and the flame. What鈥檚 more, the flames often grow very large, radiating heat to produce more and more pyrolysis products.

鈥淚 think that this sort of phenomenon is likely to happen when you have some confinement, as in a canyon,鈥� says Dold. Combined with a phenomenon called radiative extinction, where flames get so big that they extinguish themselves because they dissipate heat faster than the fuel can sustain them, this could explain how flammable gases might build up and then escape. 鈥淲hat you鈥檙e left with is no flame and copious quantities of pyrolysis products ahead of the fire,鈥� he says.

There are other reports of what could be this kind of fireball on similar terrain. In 1994 a fire in South Canyon near Glenwood Springs, Colorado, killed 14 firefighters who tried to escape as the fire raced up the canyon towards them. The circumstances of the deaths and the testimonies of survivors hint at a sudden and possibly explosive event. One survivor reported feeling a blast of hot air and smoke hit him from behind, knocking him to the ground. He and 12 of those who died had already got over the ridge at the top of the canyon. The official report into the deaths says the firefighters killed were 鈥渟till in line, many with their packs still on, and only a few showed signs that they had tried to deploy their fire shelters鈥�. Something clearly happened very fast, but was it an explosive flame?

The report considers three scenarios to explain the speed of the disaster. One is 鈥渢he possibility of a fuel-air explosion鈥�, which would explain 鈥渢he apparent suddenness with which the fire overran the 12 firefighters鈥�. The report dismisses this as unlikely given the air turbulence that accompanies wildfires, which would stop gases accumulating, but Dold thinks this is wrong. He points out that survivors reported the air above the ridge was calm, not turbulent, so the gases could have flowed over the ridge, accumulating and mixing with air, ready to ignite.

To win the argument, Dold and Weber plan to model mathematically how unburnt pyrolysis products could escape from fires and accumulate in the open air. They are even planning a series of real-world experiments with colleagues in Corsica and Portugal, where bush fires are a growing problem. They hope to show that pyrolysed gases can escape burning vegetation and gather in concentrations high enough to burn. Until then, however, the freak fireballs remain unexplained.

For firefighters like Cooper, understanding these strange events could make the difference between life and death. 鈥淲henever you are firefighting you identify a safe zone 鈥� where if it all turns to custard, you鈥檝e got somewhere to go back to,鈥� he says. 鈥淚n 2003 I chose the Oval. Had I been aware of this I鈥檇 have immediately known that was not a good place to be.鈥�