STANDING at ground zero in the centre of Haughton crater, it is difficult to grasp the enormity of what occurred here 23 million years ago. Today, Devon Island in the Canadian High Arctic is a polar desert with hardly any plant or animal life. But 23 million years ago, it was a very different place. Lush forests grew, giant rabbits hopped around, and small rhinoceros grazed.
All this changed when a comet or asteroid more than a kilometre wide slammed into the forest. The impact excavated between 70 and 100 billion tonnes of rock, forming a crater 24 kilometres wide. In the blink of an eye, all life in an area only a little smaller than Scotland ended. Plants, animals, rock and soil at the point of impact were vaporised. Further away from the crater, the shock wave, air blast and heat produced finished off most other living things.
But not for good. A few millennia after the impact, Haughton was teeming with life again. The warmth of the impact heated groundwater, creating hydrothermal systems, perfect homes for intrepid colonising organisms, such as bacteria and algae. As the crater cooled, a lake formed and sediments began to accumulate. Thanks to fossils in these sediments, we now know it was gradual climate change, not the asteroid impact, that turned Haughton back into the polar desert we see today.
Advertisement
Geologists are telling similar stories about other sites. Impact craters still deserve their reputations as scenes of devastation, but as they cool, they become ideal spots for life to re-emerge. And this has led some people to wonder if impact craters on the early Earth provided the environment for life to emerge in the first place.
Studies of the terrestrial 鈥渢ree of life鈥� suggest that of the three biological kingdoms of Eukarya, Bacteria and Archaea, the Archaea represent some of the earliest forms of life. Importantly, the Archaea are thermophilic, and these heat-loving organisms only grow at temperatures in excess of 60掳C to 80掳C. So early life on Earth must have originated and evolved in pools or systems of water that were warmed by hot rocks underground.
The hydrothermal systems active today are volcanic, for example, in Yellowstone National Park in Wyoming. But given that impact craters can also provide the two most important components of a hydrothermal system 鈥� heat and water 鈥� it seems that volcanoes are not the only contenders for giving life a place to stay.
Back in the 1970s, geologists looking at rocks that had been melted by impact craters and then cooled, realised that some of these melt rocks were not fresh. The minerals they expected to see were replaced by clays, carbonates, zeolites and iron oxides, that is, rust. These formed by precipitating out of warm water that had dissolved the original minerals. Over time, similar evidence was found at so many different craters that Horton Newsom of the University of New Mexico suggested that the formation of hydrothermal systems might be inevitable at impact craters.
But it is only recently that we have discovered what they looked like. In 2000, Pascal Lee of the SETI Institute in California, John Spray of the University of New Brunswick and I found evidence for hydrothermal activity at Haughton crater. All around the crater, piles of beautiful transparent selenite crystals 鈥� formed from hydrated calcium sulphate 鈥� and bright orange patches of rusty, hydrothermally altered rocks stand as a testament to Haughton鈥檚 warm, wet past. Mapping out the distribution of these deposits around the impact crater, something that had never been done before, helped us see what the original hydrothermal system might have looked like.
For example, Lee and I found some unusual structures around the faulted rim of the Haughton crater which we think are fossilised hydrothermal vents, places where boiling groundwater or steam that has been moving through hot rock suddenly escapes to the surface (see Diagram). By comparing our findings to active hydrothermal systems associated with volcanic areas, it seems likely that these vents probably discharged at the surface in hot springs. And since life thrives in the hot springs in Yellowstone National Park, it does not take a huge leap of the imagination to suggest that life could also have flourished in the springs at Haughton 23 million years ago.
The main heat source for Haughton鈥檚 hydrothermal system would have been the 200 to 300-metre-thick layer of molten rock that filled the central area of the crater during the impact. Groundwater underneath this sizzling mass boiled and rose upwards through the rock, escaping wherever it could. As it came towards the surface, the molten layer acted like a lid, forcing the steam to move sideways until it reached fault lines at the edge of the crater that are still there today, and are where we found the fossilised vents. As the crater cooled further, sediments in the crater bed point to the formation of a lake.
Hadean Earth
Such impact-associated hydrothermal activity has been found at nearly every impact crater that has been studied in depth, around 70 out of a total of 170 known on Earth. If almost every impact into a water-bearing planet generates a hydrothermal system, then the early Earth must have been covered with them. Despite controversy over when exactly early life appeared (快猫短视频, 22 February, p 28) many geologists date the first chemical traces of life to around 3.8 billion years ago. By examining the craters on the moon, we know that asteroid or comet impacts were 15 times as frequent then as they are now. This epoch has even been called 鈥淗adean Earth鈥�, for this reason. That represents a puzzle 鈥� why should life evolve at the harshest, most inhospitable time in Earth鈥檚 history?
But if the new thinking on impact craters is right, then although devastating impacts were common, so too were fertile hydrothermal systems. We can calculate that these systems probably outnumbered those due to volcanic activity, making them more likely contenders to house the growth and evolution of life.
But as Lee and Charles Cockell of the British Antarctic Survey in Cambridge recently pointed out, we don鈥檛 know how long the hydrothermal systems in impact craters last. If they fizzle out quickly, they won鈥檛 be much good to emerging life, which presumably needs millions of years to evolve in peace. We know the heat sources in impact craters will cool much faster than their volcanic counterparts. The hot rocks in an impact crater are at the surface, while those in a volcano are a few kilometres down. What is more, the sizzling rock source in a crater is typically several cubic kilometres in size, whereas the magma chamber of an underground volcano can hold thousands.
Kalle Kirsim盲e and colleagues at the University of Tartu, Estonia, used a computer model of hydrothermal activity to reach an estimated cooling time for the 7-kilometre-wide K盲rdla crater in Estonia. Their results, presented at the Biological Processes in Impact Craters conference in Cambridge, UK, in March, suggest that hydrothermal conditions lasted several thousands of years following the impact. But K盲rdla is a small crater. For larger craters, hydrothermal activity will last much longer, as the heat sources take longer to cool. Doreen Ames and colleagues from the Geological Survey of Canada calculate that hydrothermal activity associated with the 250-kilometre-wide crater at Sudbury in Ontario may have lasted nearly a million years. This is getting towards the kind of timescale life might need to emerge. But with only a million years to play with, life would need to develop a way to get from a cooling crater to a new, warmer one. One way to do this may be through groundwater or via springs.
There are other reasons to think impact craters were ideal for early life. Impacts lead to an increase in the size and number of pores, or holes, in rocks. At Haughton, we鈥檝e found cyanobacteria living in rocks affected by the impact. These 鈥渆ndolithic鈥� organisms actually live in tiny pores around a millimetre down into the rocks. But there are no such bacteria in similar rocks not affected by the impact.
Why do these organisms live inside rocks? Spend a few weeks in the Canadian High Arctic, especially during the winter, and the reason becomes apparent. Haughton lies in a polar desert environment that is very dry with only a few millimetres of rain annually. It has average annual temperature of -11 掳C and suffers from high levels of UV radiation. Cockell and his colleagues found that organisms living in rocks have a moist, UV-shielded and relatively warm habitat. From the Earth鈥檚 formation 4.5 billion years ago to the arrival of the first life forms, shocked rocks would be far more prevalent. Endolithic habitats would have been widely available and would have helped to shield early life from the UV ravages of an Earth that did not yet have an ozone layer.
The hunt is now on for fossil evidence of life in an ancient impact structure. But none has been found so far. Finding direct evidence that life did exist in some of these hydrothermal systems would be a huge boost. Unfortunately, it is impossible to test our hypothesis that impact craters were the origin of life, unless we happen to stumble on an ancient crater. Any craters older than 2 billion years seem to have been lost through erosion and plate tectonics, which constantly recycle the Earth鈥檚 surface.
However, plate tectonics did not occur on Mars. And the formation of hydrothermal systems in impact craters need not be confined to Earth. Mars has also experienced high numbers of impact events early in its history. Data from unmanned orbiters suggests that water ice exists in the planet鈥檚 crust and may even be present today in the Martian subsurface. If so, impact-associated hydrothermal systems may have been widespread on Mars, and these hot springs and endolithic refuges would have provided the best opportunity for any life on the Red Planet. NASA鈥檚 Spirit Rover is due to land in the Gusev crater on 4 January 2004. There are impact craters on Mars that date back to between 3.8 and 4.5 billion years ago. If these craters generated hydrothermal systems, some may hold evidence of life. Finding evidence of past life on Mars may be the only way of answering questions about the origin of life on Earth.
For now at least, impact craters are finally shaking off the bad press they have received since the discovery of a 200-kilometre-wide crater in Chicxulub, Mexico, and the coincidence in timing between its birth and the extinction of the dinosaurs 65 million years ago. While Chicxulub destroyed three quarters of the life on the planet, it seems that it also upset the balance of the global ecosystem, making way for mammals in the long run. This dual pattern of disaster and opportunity has existed throughout the history of life, maybe even at its inception.