© Tourism Western Australia
The Wolfe Creek crater in Western Australia.
Only recently has science recognised craters for what they really are: evidence of sudden impacts from long ago. Geologists Alex Bevan and Ken McNamara explore our rich heritage of fossil collisions.

Travelling south from Halls Creek in Western Australia, the hilly country of the southeastern Kimberley quickly gives way to the flat sand plains of the Great Sandy Desert.

Some 90 km south of Halls Creek, we see on the horizon a break in the monotony: an apparently flat-topped hill. In these endless plains, it is hard to judge the hill's height and distance, but after another 10 km we are almost there.

The fascinating story of the Wolfe Creek Crater begins to be revealed as we approach the slopes of the hill: the quartzite country rock becomes increasingly broken and disarranged. Rusty-red areas of iron oxide soils, which cap the quartzite, become increasingly fragmented.

Then, curious objects begin to appear. Close to the top of the hill, on its western slopes, rusty balls of rock lie scattered on the ground, sometimes fused into the laterite, and at other times lying loose.

Reaching the top of the hill, we gasp from something other than shortness of breath - for before us lies one of the most startling geological features in Australia: Wolfe Creek Crater.

Between 870 and 950 m in diameter, Wolfe Creek Crater is almost circular. Originally it would have been 120 m deep, but is now largely filled with sand and is only 25 m below the plains of sand.

There are thousands of circular structures on Earth's land surface, and many of these can be explained by the action of well-understood geological processes such as volcanism.

A number of these structures do not occur in volcanic terrains, nor are they associated with volcanic material. In the past, scientists described them as 'cryptovolcanic' or 'cryptoexplosion' structures, believing they were the result of explosive eruptive activity or that the cause of the explosion is unknown.

In the past 50 years, many features thought to be volcanic have now been shown to have an impact origin.

In 1965, researchers found 1,343 grams of iron meteorites some 3.9 km southwest of Wolfe Creek Crater, making it one of only five craters in Australia where meteorites have been found.

Meteorites only survive if the impact is small, producing a crater only a few hundreds of metres across. In larger impacts, the projectile is completely melted and vaporised. So, without the meteorite itself, what other than the circularity of such structures leads us to believe they were formed by impact?

The telltale evidence of a meteoritic origin falls into three main categories: structural, mineralogical and chemical. Geophysical surveys of many suspected impact structures show that they do not have deep-seated roots.

For example, Gosses Bluff, about 175 km west of Alice Springs, has a limit to the depth of severely disrupted rocks at around 4 km below the ground, indicating that the cause of the disruption could not have come from below, as in volcanic eruptions.

But the vital piece of evidence that distinguishes impact craters from other geological formations is the presence of shock-altered minerals. Shock waves cause microscopic transformations to occur in certain minerals.

The diagnostic features of true impact structures include multiple sets of microscopic planar deformation features in quartz, a mineral that in its unaltered form has no natural cleavage.

Significantly, impact cratering is the only geological process known to produce these so-called 'shock-metamorphic' effects abundantly.

Other features include impact glasses - melted and rapidly solidified rock - and the rare minerals stishovite and coesite, which were formed by the intense compression of the mineral quartz.

In some impact structures - such as Popigai in Russia and the Nördlinger Ries structure in Germany - the rocks are shot with tiny diamonds formed by the shock-compression of graphite (or carbon); some of these diamonds may even have condensed from the vapour created by the impact.

Since they survive over great periods of geological time, impact diamonds are additional useful indicators in the identification of impact structures.

Recently, tiny diamonds of impact origin have been recognised within impact glass at the Henbury craters, 13 to 14 craters about 150 km southwest of Alice Springs.

On a larger scale, other diagnostic features of intensely shocked rocks are shatter cones, which are produced by the sliding of rocks along cone-shaped fractures. Like accusing fingers, the apex of each cone always points toward the point of impact.

Shatter cones have not been found at Wolfe Creek, but it is likely that erosion has not yet penetrated deeply enough to expose the shatter cones, which may lie beneath the crater.

Evidence of giant impacts occasionally presents itself in places other than at the site of collision. Rocks thrown great distances from craters go on to land in alien terrain, and flying glass such as tektite (far from the crater) or impactite (close to the crater), remain as incriminating evidence of impacts.

Asteroids hitting seas and oceans raise giant tsunamis that break on land, depositing muddy sediments and other debris as they flow across hundreds of kilometres. Ejecta in the form of breccias - broken bits of rock cemented together - are either hurled, or flow considerably far from the crater from which they were ejected.

Pseudotachylite is an unusual kind of impact breccia that occurs deep in the target rocks of some large impact structures. In composition, pseudotachylites correspond closely to their adjacent host rocks, indicating that they formed from them by grinding, and sometimes frictional, melting.

Pseudotachylites can occur as thin, millimetre-sized veins, or bodies up to many tens of metres thick. They can contain numerous large and small rounded inclusions of milled target rock, set in a dense matrix that is generally black to blackish-green in colour. Pseudotachylite-like rocks are not exclusive to impact structures but also occur in zones of intense deformation, such as faults. They can result from sharp tectonic movements.

Bodies of tectonic pseudotachylite tend to be linear and less than a few metres thick, whereas impact-produced pseudotachylites can form large, more irregular bodies developed over wide areas.

Pseudotachylites then, are themselves not necessarily diagnostic of impacts unless they are accompanied by other evidence of shock-metamorphism. When pseudotachylites are present in impact structures, the melt-rich varieties offer a means of absolute age dating of the impact event.

In many impact craters, gas from the vapourised projectile can be injected into the target rocks, leaving a distinctive chemical signature.

The type of signature depends on the make-up of the meteoritic culprit, and it has been possible in some cases to identify the type of impacting meteorite from analysis of the melted rocks that form the impact structure.

An example of telltale signs of a giant impact occurring in a place other than the impact site is the recognition of the Lake Acraman structure in the Gawler Ranges of South Australia as a huge scar of extraterrestrial origin.

In 1983, geologists, including Vic Gostin working in the Flinders Ranges some 300 km east of Lake Acraman, discovered a layer of shattered volcanic rock fragments, some up to 30 cm across, encased in 600-million-year-old shales in the Pichi Richi Pass and other spots throughout the ranges.

Dating of the 'foreign' rocks showed that they were around one billion years older than the shales in which they had become embedded.

Recognising that the rock fragments were typical of ancient volcanic rocks found at the Gawler Ranges, near Lake Acraman, the geologists deduced that they must have come from there: but how?

Independently, in 1979, another geologist, George Williams, had discovered abundant evidence of shocked mineral grains and shatter cones in the volcanic rocks exposed on the shores of the nearly circular Lake Acraman, and concluded that it was an impact structure.

Eventually, the geologists compared notes and realised that the rock fragments found in the Flinders Ranges represented some of the debris hurled from the impact at Lake Acraman. The estimated age of the host shales pinned down the time of the impact to around 600 million years ago.

At the time of the Lake Acraman impact, a shallow sea existed in the area now occupied by the Flinders Ranges. Debris from the impact rained down on the sea and sank into the muddy seabed, which later became buried by other sediments.

Subsequent upheavals raised and contorted the accumulated sediments that now form the Flinders Ranges, and erosion has exposed the 'fossilised' remains of the fallout from the Acraman impact.

The original Acraman crater, which formed some one or two kilometres above the present land surface, was at least 30 km in diameter; but circular fractures around the impact indicate that the final collapse crater could have been as much as 40-90 km in diameter. This makes it one of the 10 largest known impact structures in the world, and the energy released during its formation was around a million times greater than at Wolfe Creek.

Recently, satellite photography and extensive geophysical exploration have revealed some 200 circular structures that may have been formed by giant meteorite impacts on Earth's surface and beneath the oceans.

Many circular features, some up to 100 km or more in diameter, have been observed as ghostly outlines in some of the oldest rocks on Earth.

Like Australia, other ancient continental landscapes such as the Canadian Shield, South Africa, Siberia and the Scandinavian Shield display faint circular scars that testify to the impact of huge asteroids millions of years ago.

Once earth's geological history has been detailed, it can reveal how often these potentially destructive bodies strike Earth. Crater-forming meteoroids - those weighing a few hundred tonnes or more - would be expected to strike Earth's land surface every 20 years or so.

Theoretically, a meteoroid big enough to produce a crater the size of Wolfe Creek - tens of thousands of tonnes - should arrive on Earth about once every 5,000 years.

However, a crater-forming impact on this scale has not occurred in historical times. The reason is that Earth is not without its natural defences, and there are a number of factors that work in our favour.

Since three quarters of Earth's surface is covered by water, most meteorites fall into the sea. Also, the majority of meteorites are brittle objects and break up in the atmosphere to fall as many small objects rather than one large one.

Studies of craters show that many were made by the impact of iron meteorites, which are less prone to break-up in the atmosphere, but far less common than stony meteorites.

When all the factors are taken into account, impact-producing structures on the scale of Wolfe Creek are predicted to occur perhaps once every 25,000 years, while collisions on the Lake Acraman and Gosses Bluff scale only about once every 15 million years. Potentially catastrophic global events like the Chicxulub impact, which might cause extinctions, may only occur on a time scale of 50-100 million years.

Realistically, what are the chances of a catastrophic impact on Earth? If the fossil-cratering record is anything to go by, over the vast time-scale of geological history, catastrophic impacts have certainly occurred and are indeed likely to occur again in the future.

Humanity's written history is very short - a few thousand years or so; this pales into insignificance against the 4.6 billion years of Earth's geological history. And just because large-scale cratering has not occurred during historical times, this does not mean it will not happen in the future.

Since 1998, there's been a coordinated hunt for Near Earth Objects, or NEOs: asteroids or comets that cross the Earth's orbit and are up to 195 million km from the Sun (Earth is 150 million km from the Sun).

Australian astronomers use the 0.5 m Uppsala Schmidt Telescope at Siding Spring Observatory, in Coonabarabran, New South Wales - the only survey in the southern hemisphere.

Multiple images of the same patch of sky are taken 15 minutes apart and compared - enough time to spot a NEO crossing the frame. Unfortunately, only NEOs 1 km in diameter or more are easily detectable, and the work is funded by NASA since Australian funding dried up over a decade ago.

To date, the surveys have found 7,679 NEOs. Of these, 1,180 asteroids are classed as 'potentially hazardous': passing within 7.5 million kilometres of Earth and being at least 110 m in diameter. Each new NEO's orbit is calculated to establish if it will collide with Earth in the future.

If one was identified as a threat, existing technology might be able to deflect it. One technique suggested is to set off hydrogen bombs above its surface: high-speed neutrons from the blast would irradiate one side of the asteroid, causing material to expand and blow off.

The asteroid would recoil, producing a small change in velocity - enough to make it eventually miss Earth. Blowing up the asteroid entirely would create an even bigger problem, since all the pieces would still be headed for Earth. Another idea is to attach large solar sails to the object, allowing the pressure of sunlight to redirect the object away from Earth.

There may even be a commercial spin-off from NEO studies: the potentially hazardous comets and asteroids are the ones that could most easily be captured in Earth's orbit and mined for metals such as platinum and cobalt.

It's estimated that a metallic asteroid 1.6 km in diameter might contain more than US$20 trillion worth of exploitable resources. - Nina Pace