Fires of the apocalypse
New Scientist vol 178 issue 2391 - 19 April 2003, page 32
The asteroid that wiped out the dinosaurs triggered such terrifying infernos that experts wondered how life survived at all. Ivan Semeniuk finds out where to hide if it happens again
IT IS a calm day on the island continent of India. Deep in the Cretaceous forest, small plant-eating dinosaurs are quietly nibbling on tender shoots, keeping a watchful eye out for predators. Suddenly every pair of reptilian eyes looks up. The sky is beginning to glow. The lush green leaves of the forest canopy droop and shrivel. As the air temperature suddenly soars to several hundreds of degrees, the dinosaurs collapse and lose consciousness. Like tinder in a furnace the forest suddenly bursts into flames. Within hours, all of India is burnt to a crisp.
Or that's how David Kring sees it - and he should know. In September last year, Kring, who is based at the University of Arizona's Lunar and Planetary Lab in Tucson, published the most detailed account we have of what happened when an asteroid the size of Manhattan hit Earth 65 million years ago. His analysis is based on a computer simulation carried out with Dan Durda of the Southwest Research Institute in Boulder, Colorado. And that simulation could also tell us where to head if we want to survive a future impact.
The event Kring and Durda studied is widely considered to be the worst disaster to hit the planet in the past quarter of a billion years. It began when an asteroid or comet struck Earth off the southern coast of ancient North America. The Chicxulub crater, the remnant of the impact discovered buried under Mexico's Yucatan peninsula a decade ago, is at least 180 kilometres across, and dates back to the Cretaceous-Tertiary (K/T) boundary, a geological layer of material laid down around the time that the dinosaurs went extinct.
Early calculations suggested a vast amount of pulverised rock must have been blasted into the atmosphere and beyond, before falling back to Earth all around the globe. As the dust and debris re-entered the Earth's atmosphere, the searing heat caused by friction spelled doom: it heated the air to a temperature at which whole continents simply ignited. Three-quarters of all plant and animal species, including the dinosaurs, were wiped out.
But something about this scenario has long puzzled geologists. "I don't doubt the dinosaurs roasted in their tracks," says Jay Melosh of the University of Arizona, who has carried out his own extensive studies of this event. "My problem has been understanding how anything at all survived." Now Kring and Durda may have found the answer - and with it a possible means of escape.
The idea that the K/T impact - and thus any future impact - could ignite fires worldwide surfaced in the mid-1980s when Wendy Wolbach, then a graduate student at the University of Chicago, noticed an unexpectedly large quantity of carbon in a section of the K/T boundary from Denmark. Using a scanning electron microscope, Wolbach revealed the carbon was made up of tiny particles of soot, the telltale signature of an ancient forest fire. Similar particles were soon turning up in K/T sections from all over the world, including the bottom of the Pacific Ocean. The amount of soot implied fires had raged on a massive scale. "One of the questions that came up after the soot was discovered is how do you generate this much fire? How do you ignite forests all over the world?" says Wolbach, now a professor of chemistry at DePaul University in Chicago. "We didn't have an answer for that."
Then, in 1990, Melosh published the results of a series of calculations that showed how impact debris in the form of tiny grains of melted rock - meteors made on Earth - would be carried around the world on ballistic trajectories (Nature, vol 343, p 251). As they re-entered the atmosphere, these particles would have delivered a remarkable amount of heat energy. "People are used to thinking of meteors as harmless," says Melosh. "But in this case you have to imagine the atmosphere so full of meteors the sky turns red hot."
Melosh and his collaborators found that the amount of heat radiating from the atmosphere would cause vegetation on the ground to smoulder and eventually combust. Like paper under a magnifying glass, forests would suddenly burst into flames. But the resulting fires would not be like the small forest fires of today, Melosh said. They would ignite spontaneously everywhere at once.
So how did anything survive? To answer that question, Kring needed to produce a more detailed picture of the devastation caused by impact-generated wildfires. And that is why he recruited the help of Dan Durda, a planetary scientist at the Southwest Research Institute in Boulder, Colorado. Durda is an asteroid impact specialist, but not of the kind that you would expect. He is an expert on impacts on the surfaces of distant asteroids, something he picked up after NASA's Galileo spacecraft returned images of Ida, a 60-kilometre chunk of rock with a heavily cratered surface.
Ida's pockmarked features bear witness to the many blows it has been dealt by smaller objects. Durda developed a computer simulation to determine how the debris ejected from all these craters eventually settled across Ida's irregular contours. The program, powerful enough to trace the intricate ballet of impact debris on Ida, was just what Kring needed to study how debris from the Chicxulub impact crater showered over Earth.
Before they could start, however, Kring and Durda had to get Earth's layout correct. At the end of the Cretaceous period, the continents did not have the same shapes and relative locations they do today, thanks to the relentless creep of the Earth's crust. Central America did not exist, and India was still an island. Even the Chicxulub crater has moved significantly since then.
"We had to reconstruct the geography, including the latitude and longitude of the impact site at that time," says Durda. "That became the location from which we launched the debris." With the continents and impact site in their proper places Kring and Durda were at last able to run their simulation.
What emerged was a dramatic view of how the world caught fire 65 million years ago. As the asteroid strikes, roughly 10,000 cubic kilometres of the Earth's crust are converted into hot vapour and blast debris, spraying outward from the impact. Some of it flies halfway to the Moon before falling back to Earth.
Although the details depend on a few as-yet undetermined variables, such as impact trajectory, the initial effects are always felt most strongly in southern half of North America, the land mass nearest Chicxulub. This becomes a hot spot within moments of the impact. Temperatures are quickly driven up past the combustion point of vegetation, as a mixture of liquid and solid material that Kring calls low-energy ejecta is spattered outward from Chicxulub. This material can be found in these areas today, in the form of a solid layer about 1 centimetre thick.
"This is the stuff that most people think of when they picture an impact event," says Kring. "It's that cone of material that seems to leap out of the crater, similar to the water that splashes when you throw a rock in a pond."
But dramatic enough as that might be, Kring and Durda's simulation shows a surprise. It is the lighter, high-energy ejecta that cause the global catastrophe. While the remains of this debris forms a layer only a few millimetres thick at most, the high-energy ejecta from Chicxulub appears to have settled all over the globe. In Kring's scenario it is made up of a few thousand cubic kilometres of material that was vaporised and lifted out of the crater at high speeds. Within minutes this vapour-rich plume rose far beyond the atmosphere and expanded like a cloud until it enveloped the entire Earth. And wherever it fell back in sufficient quantity, the wildfires began.
Less than an hour after the impact, a second hot spot appears, this time on the other side of the world, near India. This antipodal effect was predicted in Melosh's original calculation: it is the natural result of many trajectories radiating outward from the Chicxulub impact and then re-converging due to the gravitational field and rotation at the opposite point on the globe. This is why India is likely to have been especially hard hit by fire.
However, over the following hours the Earth's eastward rotation causes these two hot spots to drift westward on parallel tracks of destruction. The northern track crosses the Pacific and eventually hits southern Asia. Meanwhile, the southern track moves from India, across the southern half of Africa and on over the Atlantic to scorch the middle of South America. And so it goes on for days.
"Some of the debris is thrown halfway to the Moon," says Kring. "We learned that it takes three or four days for this material to re-accrete to the Earth, and during that time some places got hot not once but three or four times."
These repeated scorchings ensure that anything left unburnt the first time in places like India would eventually be in flames as the hot spot returned every twenty-four hours. But just as interesting to Kring and Durda was the revelation that some parts of the world, including Europe and northern Asia, may have escaped the deadly fires. These, they reason, should be the places where life survived.
Their results are backed up by recent detective work by Art Sweet and Dennis Braman, pollen experts with the Geological Survey of Canada and the Royal Tyrell Museum respectively. In 2001 they published a comprehensive survey of changes in the North American pollen record across the K/T boundary (Canadian Journal of Earth Sciences, vol 38, p 249).
On the side of the boundary prior to the impact, Sweet and Braman found an abundance of pollen that is typical of a diverse forest ecosystem dominated by coniferous trees. But after the boundary, the tree pollen is abruptly replaced with the pollen of fast-growing ferns.
"Big trees disappeared, or at least their capacity to produce pollen disappeared," says Sweet. "One interpretation is that fire stripped off the canopy while plant species in the under-storey of the forest survived and were actually given better growing conditions, in the sense of not being shaded."
The "fern spike" in the pollen record is not new to palaeobotanists. It was first spotted in the Raton Basin, in southern Colorado and northern New Mexico, nearly two decades ago. But Sweet and Braman found the change from tree to fern pollen became noticeably less pronounced as they followed the K/T boundary northward, as far as the Mackenzie River in Canada's Northwest Territories. This fits well with Kring and Durda's simulation, which demonstrates that fire is more likely to have spread westward rather than northward as the Earth rotated.
But while the pollen record gives some clues, Kring and Durda are eagerly awaiting the results of studies by other research groups that may help refine and test the accuracy of their simulation. One study in particular should provide crucial information. It is being led by Margaret Collinson at Royal Holloway, University of London, an expert in palaeobotany. She and her colleagues are studying fossilised charcoal from the K/T boundary, which could provide the most direct evidence of where fires occurred after the impact. Unlike soot, which was carried on the wind and deposited around the globe, charcoal should only appear in places where fires actually happened.
Careful analysis of the charcoal - the size and shape of fragments, for example - can reveal a great deal of information about what kind of vegetation was around, and exactly how it burnt and broke up into pieces. The Royal Holloway researchers are currently trying to determine how various burning conditions produce different kinds of charcoal. They are doing this by making their own charcoal, burning wood and plants in furnaces and trying to get a sense of the microscopic charring patterns produced by different temperatures and heating durations. "The cell walls of plant cells are built in layers," says Collinson, "And we can see the layers homogenise as they heat up. That helps us determine what parts of plants were burning and at what temperature."
Once they have that information, Collinson's team should be able to deduce an enormous amount about the details of the K/T wildfires from the fossilised charcoal samples and the locations in which they were found. The analysis should be complete within a few months. Once that is done, Kring and Durda should have a much clearer picture of not only where in the world fires occurred but how hot and for how long they burned.
Of course, Kring and Durda's simulation does not cover all the effects of the Chicxulub impact: wildfires are only part of the story of how the impact relates to the mass extinction. The fires would have generated large quantities of carbon dioxide, carbon monoxide, methane and various toxins whose effect on the biosphere can only be gauged in conjunction with the dust and chemical pollutants tossed up by the impact itself. The possible side effects include severe acid rain and global cooling (due to dust blocking sunlight) followed by a prolonged period of global warming. "It's every single environmental catastrophe you can think of happening at the same time," says Durda.
It was not all bad, however. To Kring, the simulated fire burning across his computer screen represents the crucible of human evolution: "Without it, we wouldn't be here," he points out. And the K/T boundary is a perfect place to study how life on Earth responds to a major upheaval. "It strikes me as a beautiful test case for looking at how the whole system breaks down and how it recovers," says oceanographer Steven D'Hondt of the University of Rhode Island in Kingston. "In terms of the interplay between organisms and the planet I don't think we've got a better case in Earth history."
But that does not mean we want a repeat of the experience. If we see an asteroid coming and cannot deflect it, can Kring and Durda's model tell us the likely effects of any future impact? Can it tell us where on Earth we will be safe? Perhaps it can - if we have details of the trajectory. Kring and Durda are hoping to achieve this - albeit in reverse - to end the argument over the trajectory of the K/T impact. There is plenty of evidence for a shallow strike at Chicxulub: the crater is shaped like a horseshoe, open to the north-west. "The most reasonable explanation is an oblique impact angle," says oceanographer D'Hondt. But although there is broad consensus that the impact was shallow, different researchers have widely differing views on the direction the object came from.
Kring and Durda's approach is to link the pattern of fires seen in the fossil record to a particular trajectory. A shallow impact angle sprays debris preferentially in one direction, and this alters the pattern of wildfires across the continents - although North America and India still burn no matter what.
Turn that around, and the model could tell us where to head to survive a future impact. It is a tantalising prospect: we could map out Armageddon before it happens, and take steps to minimise the loss of life. And, of course, if the model gets it wrong, there will be no one to sue Kring and Durda for damages.
|
|