The Great Ordovician meteor shower

Posted on 30 September, 2013 by Metageologist

Between Mars and Jupiter, 470 million years ago, there was a massive collision between two 100km-sized chunks of rock – this solar system’s biggest bang of the last billion years. It created a massive cloud of smaller fragments. Some of these landed on the earth, falling at a rate at least a hundred times greater than at present. These fragments can be found today in sedimentary rocks from that time. More speculatively, this shower from space has been linked to two dramatic events – a set of vast fossil landslides plus a major event in the history of life – a landslide of new fossils known as the Great Ordovician Biodiversification Event.

Grains raining from the sky

Meteorite from Thorsberg Quarry. Image from Lund University

It started with unsightly green blobs in a Swedish limestone quarry. Discarded by the quarrymen, these odd lumps are fossil meteorites. These are incredibly rare – something most geologists will never see. This may explain why they were only identified correctly in the 1980s (by amateur geologist Mario Tassinari). Since then, researchers – notably Birger Schmitz of Lund University in Sweden – have found over 90 meteorites from this one quarry1.

Recognising this as something remarkable, they started looking for other evidence in other rocks of the same age. Dissolving limestone in acid, they were able to pick out tiny grains of chromite. This mineral forms on the earth, but using chemistry they were able to show that these grains could only have come from space. Such grains have been found in China2 and Russia, as well as Sweden. Tiny tiny meteorites (fabulously called cosmic spherules) have been found in Scotland3 and Argentina.

Reading up on this, I was rather excited to realise that rocks of the same age are found in Ireland. I was writing a post about the use of heavy mineral analysis in these rocks, showing variation in the number of chromite grains! Had I just made an exciting connection? The Irish papers interpret the chromite as coming from an eroded ophiolite – were they actually from space? I did some maths and as it turns out, no they weren’t. Maths can be cruel.

The rocks in Sweden contain so many meteorites for two reasons. As well as forming at a time when huge numbers were falling to earth, they also formed extremely slowly. Known as a condensed sequence, each centimetre thickness represents tens of thousands of years of deposition. Sitting on a flat sea-bed, little or no sand was washed in so only limestone mud, a sort of organic dandruff, settled to the sea-bed. That and fragments of a space collision, scientific manna from heaven. The Irish rocks built up thick layers a thousand times faster as sand, gravel and mud spilled from the hills above. If you processed 1000kg of Irish rock, you’d expect to find only a few grains of extraterrestrial chromite. So the many grains counted in the Irish studies could have contained only a small number of space chromites. The most likely small number being zero.

Look to the heavens

All the best scientific stories reach across different disciplines. Studies of the chemistry of this space dust show it to come from a single source – the meteorites are all of a well-known type called “L-chondrites”. Using isotopes, Schmitz was able to assess how long his chromite grains had been floating in space, exposed to cosmic rays. The younger the rock layer, the longer the exposure. All of this suggested the middle Ordovician meteorite shower was caused by a single event, fracturing a large body into many pieces. Independently, a 1964 study of L-chondrite meteorites had identified a ‘shock age’ of around 470 million years ago – providing independent evidence for the collision. Using spectral analysis of asteroids (looking very very carefully at their colour), its possible to identify pieces of the original body that remained in stable orbits – the Gefion family of asteroids. L-chondrites even today form about 20% of the meteorites that reach earth.

Hidden impacts?

In the Earth Sciences, things tend to follow a power-law distribution. For example, tiny earthquakes are very common, moderate ones common, large rare, and very large earthquakes are very rare. Smash a huge asteroid into pieces and you might expect the size of the fragments to follow a power-law distribution . On earth we’ve found the uncountable numbers of tiny chromite grains and a lot of small meteorites – it is entirely reasonable to assume that a few crater-forming-size fragments also hit the earth in the Ordovician.

They’ve found a few – the Lockne crater in Sweden and the Osmussaar breccia in Estonia4 are pretty solidly linked to large impacts by L-chondrite bodies in the Ordovician. However craters are remarkably hard to preserve so maybe there aren’t that many more to find. What is needed is traces of the impacts that affected a large area and might be found in sedimentary rocks of this age.

John Parnell of Aberdeen University has suggested5 that the many impacts at this time caused an unusual series of ‘mass wasting’ events on continental margins – essentially a series of massive landslides. These are not small things – the Buttermere formation6 in England’s Lake District is a 1500m thick sequence of sediments that was sheared and folded as they shifted down an ancient sloping sea-floor. There are another 13 similar deposits of middle Ordovician age around the world.

Sample of the Buttermere Formation Olistostrome from Ian Stimpson

Not everyone agrees7. Massive landslide deposits are not uncommon – the middle Ordovician was also a time of sea-level fall, something that can cause continental slopes to become unstable. Further, all of the examples formed in tectonically active areas. The Lake District rocks formed on a volcanic arc near a subduction zone. There would have been plenty of large earthquakes to trigger a landslide – there is no need to invoke a nearby meteorite impact to explain it.

Change the history of life?

Could the mid-Ordovician impacts have changed the course of life on earth? In 2008 Birger Schmitz 8 linked them to a dramatic event in the history of life: the Great Ordovician Biodiversification Event (GOBE). Spanning 25 million years, this event saw an unprecedented increase in the number of species of fossil animals. Schmitz and co-workers tracked both fossil abundance and the record of extraterrestrial debris on a bed by bed scale.

Figure 3 from Schmitz et al. (2008). “The results are based on bed-by-bed collections at eight localities. Note the dramatic increase in biodiversity (black line) and high extinction (blue line) and origination (red line) levels following the regional Volkhov–Kunda boundary, that is, the same level where extraterrestrial chromite appears and Os isotopes change”

How could the things be linked? They talk of “impact-related environmental perturbations” which feels like one of those CIA euphemisms for murder, meaning as it does “sterilising large areas of the earth”. The key point is, not all of the earth. Once the dust has settled, a habitat empty of inhabitants is a fantastic opportunity for nearby animals to move into. By creating a more varied environment, impacts can actual increase the diversity of species, they argue.

It’s a lovely idea,but one that is far from being proven. Other environmental factors (very high sea-level, lots of islands, changing climate) or biological changes (the GOBE sees planktonic life become important for the first time) are equally or more plausible. In a great recent podcast overview of the GOBE, David Harper (second author on the paper) refers to the meteorite link only after discussing all the other possible causes, and does so with a slightly apologetic tone.

Showing that the GOBE and the remarkable flux of space debris happened at the same time is not enough. What is required is to prove the causal relationship. This is a very hard thing to do. Maybe all of the proposed causes were each partly responsible?

The History of the Earth is history

The Earth Sciences are unusual in being partly a study of past events. Sometimes perhaps we should think more like scholars of human history. Historians studying, say, the origins of the First World War are aware of the importance of multiple causes. The tensions of imperialism, the aggressive German foreign policy, even the inflexibility of railway timetables – are just some of the many proposed ’causes’ of this terrible war. Any educated discussion of this topic would acknowledge that many things were contributory in some way. There is no single ’cause’ for WW1.

Was the extinction of the dinosaurs caused by the eruption of the Deccan Traps or the Chicxulub impact? Yes. Surely both are part of the story in some way9? Thinking a little like historians, the quest to prove one cause is right and the other wrong seems a little foolish.

The events of the middle Ordovician are less dramatic but illustrate the same point. Maybe we will never know if the Buttermere olistrostrome was caused by a meteorite impact, or just a large earthquake and low sea-level. But maybe we will. Historians derive new insights from studying archives of old documents. Our archive is the world itself (and beyond). New rocks, new techniques and novel combinations of the two may one day give more dramatic insights on the Great Ordovician meteor shower.

References

I first heard of the Great Ordovician meteor shower from Ted Nield’s excellent book Incoming, which I recommend to you as it covers all manner of marvellous meteoric matters.

I’ve tried to put web links to copies of the original papers in the footnotes. Formal references to key papers are below.

Schmitz B., Tassinari M. & Peucker-Ehrenbrink B. (2001). A rain of ordinary chondritic meteorites in the early Ordovician, Earth and Planetary Science Letters, 194 (1-2) 1-15. DOI: 10.1016/S0012-821X(01)00559-3
Parnell J. (2008). Global mass wasting at continental margins during Ordovician high meteorite influx, Nature Geoscience, 2 (1) 57-61. DOI: 10.1038/ngeo386
Meinhold G., Arslan A., Lehnert O. & Stampfli G.M. (2011). Global mass wasting during the Middle Ordovician: Meteoritic trigger or plate-tectonic environment?, Gondwana Research, 19 (2) 535-541. DOI: 10.1016/j.gr.2010.07.001
Schmitz B., Harper D.A.T., Peucker-Ehrenbrink B., Stouge S., Alwmark C., Cronholm A., Bergström S.M., Tassinari M. & Xiaofeng W. (2007). Asteroid breakup linked to the Great Ordovician Biodiversification Event, Nature Geoscience, 1 (1) 49-53. DOI: 10.1038/ngeo.2007.37

Some facets of the Geology of Diamonds

Posted on 12 May, 2013 by Metageologist

Originally published on the Scientific American guest blog.

Geoscientists can’t say if diamonds are forever, but they can say that some are already billions of years old. They form in a place we’ll never reach: the deep earth, hundreds of kilometres under our feet. Diamonds tell us much about this hidden world and how it is linked to the surface – and life – in surprising ways.

Diamonds are made of carbon atoms which are densely packed into a structure that is extremely strong. On earth they form only under extreme pressures – under conditions very unfamiliar to us surface-dwellers. Some form in the sudden shock-waves created when material from space hits the earth. The global impact layer found suspiciously close in time to the extinction of the dinosaurs contains countless tiny diamonds. Impact diamonds are rare. Most diamonds, certainly any big enough to put in an engagement ring, form slowly within the deep earth.

Imagine a slab of concrete – about 5cm thick – resting on your chest. The pressure is small, but tangible. The pressure found in the deepest ocean is equivalent to some 80,000 of such slabs. Diamonds form at pressures that are at least 45 times greater still, equivalent to millions of slabs or hundreds of kilometres of rock. The earth’s deep interior is a place where even rocks are transformed by the massive pressure.

Natural diamonds don’t form, Superman-style, by the application of pressure directly to other solid forms of carbon (such as coal). They grow by the interaction between a carbon bearing fluid and rock – typically involving redox reactions such as the breakdown of CO2 or methane. Diamonds show complex patterns that suggest they grow gradually. Studies of diamonds from a single area often show a wide distribution of ages, from over 3 billion years old to a few hundred million.

The ‘Picasso diamond’ shows complex growth patterns highlighted by cathodoluminescence. Image with permission of University of Edinburgh

Diamonds form within the earth’s mantle, the thick layer between the thin crust and earth’s metal core. They are particularly associated with parts of the mantle that are stuck to the bottom of long-lived continental crust. Here the mantle forms stable ‘keels’ and doesn’t take part in the convection-driven movements that happen lower down. The portions of stable crust with keels are called cratons – the largest are found in North America, Africa and Australia -all areas rich in diamond mines.

Cratonic keels are very stable, but are not totally insulated from the dramatic events in the rest of the dynamic earth. Subduction at the edge of cratonic plates allows oceanic crust to sink deep into the mantle underneath the craton. Carbon-bearing fluids from the sinking oceanic crust rise into the cratonic keel and may cause a phase of diamond formation. Mantle plumes, columns of hotter rock rising from the base of the mantle can do likewise.

In contrast to how they form, the way diamonds reach the surface involves one of the quickest and dramatic geological events we know. Most diamonds reach the surface brought up within an odd type of molten rock called Kimberlite. This magma forms at great depth in cratonic keels and is rich in volatile elements such as CO2 which makes it highly pressured. If it is able, it will rise to the surface extremely quickly through vertical fractures. At the surface it forms a carrot-shaped pipe which nowadays is often the site of a large circular diamond mine.

Diamonds and other deep minerals are brought to the surface as fragments within the kimberlite magma. Diamonds are able to survive the rough-and-tumble of the eruption very well, but it helps that the eruption events are very quick. Not just geologist-quick, but normal-folk quick. Estimates are that diamonds travel to the surface in at most months but maybe as quick as a few hours. Diamonds are only stable under surface conditions because they are too cold to change their structure. The speed with which they reach the surface and cool down keeps them beautiful and prevents them from turning into worthless graphite on the way up.

Some diamonds are not conventionally beautiful. They contain blemishes, tiny blebs of fluid or inclusions of other minerals that dim their brilliance. But to geologists these are the most attractive diamonds of all. Listen to them carefully and they will whisper secrets about a place we’ll never reach – the deep earth.

The deep earth is only a few hundred kilometres below your feet, but is completely inaccessible. The deepest hole ever drilled is a puny 12.2 kilometers. At diamond depths the rocks are at temperatures over 1000°C – few man-made materials can survive such conditions.

Fortunately we can tell a lot remotely. Seismologists gather information on the way waves created by earthquakes pass through the earth and they can dimly make out structures at great depths. This ‘seismic tomography’ applies the same principles that PET or MRI scanners use to study a human body. Such tools are useful, but in medicine as in geology, sometimes direct sampling of the interior is required: kimberlites act like biopsies, making samples of the interior available for detailed study.

A tremendous range of experimental techniques have been used to study diamonds and their inclusions. Some have poetic-sounding names (“Raman spectroscopy”) but many do not (“combustion analysis”, “laser ablation ICPMS”). Most are used to measure the elemental composition of the minerals or the isotopic makeup of those elements. These data are not just of interest to chemists.

The chemistry of mineral inclusions can yield information about the pressures and temperatures at which they (and the diamond) formed. Radioactive isotopes can be used to estimate the age of formation.

Stable isotopes tell some of the most remarkable stories in the earth sciences. Particular processes create distinctive isotopic signatures that may be preserved through a whole range of subsequent events. One isotopic signature only forms when ultraviolet light interacts with sulphur in an oxygen-poor environment. This signature has been found in diamonds, meaning that they contain material that was once at the surface (rock is a very good sun-block, so UV reactions do not occur inside the earth). Also, the sulphur was at the surface very early in Earth history, before photosynthesis caused atmospheric Oxygen levels to rise.

Photosynthesis has its own distinctive isotope signature, affecting carbon. Some diamonds contain this ‘light carbon’, meaning they are formed from life itself. They are the most amazing type of ‘fossil’ imaginable. Some living organism ended its life as a smear of black carbon in a sedimentary rock. It was then buried deep by subduction. Some of its atoms rose up again, first in fluid and then as part of a diamond, suddenly flung to the surface for us to find and marvel at. This deep loop of the carbon cycle is small in terms of volume but conceptually it is enormous. The cycling of carbon between plants, animals and the atmosphere is well know. Uncomfortably, we are becoming more aware of the additional link between buried coal, atmospheric carbon and climate. But the far deeper cycling of carbon into the mantle, demonstrated by diamonds is only recently proven. We can never reach the deep earth, yet it is intimately linked to surface via the subduction of oceanic crust.

Not all diamonds form from surface material. Carbon has been part of the mantle since the formation of the earth and this carbon forms diamonds too. Tracing types of mineral inclusions, it is possible to distinguish diamonds formed from subducted material from other types. This reveals an interesting pattern: diamonds that are older than 3 billion years show no trace of subducted material. This suggests – consistent with other evidence – that plate tectonics as we know it was not active in the very early earth. Subduction may only have started 3 billion years ago.

Inclusions of lower-mantle minerals (ferropericalse and MgSi-perovskite) inside a diamond that formed at >600km depth. Image kindly supplied by Prof. Ben Harte, University of Edinburgh.

Most diamonds form in the upper reaches of the mantle but some come from deeper down. These ‘sub-lithospheric diamonds’ form in the part of the mantle that slowly circulates in convection currents. This lower mantle forms the majority of the earth by volume, yet is poorly understood. At these depths only exotic minerals are stable, traces of which are found as tiny inclusions within diamonds. The only other place we can see these materials is in the laboratory. Here ‘anvils’ are used to squeeze tiny samples to tremendous pressures. The material they are made of is very strong, but also transparent, so that observations can be made and lasers fired through it to heat the samples. What are these special anvils made of? Diamonds, of course. These are precious stones indeed.

References

A great open-source review of current knowledge from the Deep Carbon Observatory :
Shirey, S., Cartigny, P., Frost, D., Keshav, S., Nestola, F., Nimis, P., Pearson, D., Sobolev, N., & Walter, M. (2013). Diamonds and the Geology of Mantle Carbon Reviews in Mineralogy and Geochemistry, 75 (1), 355-421 DOI: 10.2138/rmg.2013.75.12

The latest evidence that diamonds are made from life:
Schulze, D., Harte, B., , ., Page, F., Valley, J., Channer, D., & Jaques, A. (2013). Anticorrelation between low 13C of eclogitic diamonds and high 18O of their coesite and garnet inclusions requires a subduction origin Geology, 41 (4), 455-458 DOI: 10.1130/G33839.1

Earth, moon and Mars: connected by meteorites

Posted on 21 November, 2012 by Metageologist

The time immediately after earth’s formation is known as the Hadean Eon. It was a time when earth suffered a heavy bombardment from space. Rocks this age on earth are extremely rare, mostly they have been destroyed by later events – eroded, dissolved, melted or smashed. Scientists searching for rocks formed during these hellish times are starting to turn their eyes away from earth, up to the heavens.

At a recent Lunar Science Event, joint between the Geological Society of London and the Royal Astronomical Society I had a marvellous time. Here’s some of what I learnt.

One of my favourite talks was given by Jay Melosh (Purdue University) who talked about how pieces of planets are thrown off into space and land on other planets. The most unequivocal line of evidence that this happens is the Martian Meteorites – 40 pieces of meteorite that came from Mars but ended up on earth . Dating of these samples suggests 6 ejection episodes, all within the last 20 million years. Magnetic studies suggest that at least some of these samples never reached temperatures greater than 40°C

Apart from Apollo astronauts, the only mechanism that send rock samples through space are large impacts. These are associated with large pressures and temperatures, so how do we explain these samples that were not strongly heated? Numerical modelling, if its very fine-grained, shows that rocks on or near the surface close to the impact are thrown off extremely fast, without being heated. As the shock wave radiates out from the impact site it interacts with the free surface in a process called spallation.

Studies of Martian craters show large sprays of material, up to 1500km from the crater. This material, thrown from the impact can be large enough to form their own craters as they hit the ground again, so called ‘crater secondaries’. It’s a relatively small step to move from throwing material this far to putting it beyond escape velocity and into space. The Zunil crater on Mars is a good example of this and may be a source of earth’s Martian meteorites. The same phenomena is seen on earth – fragments of limestone originally from Germany, near the Ries crater, are found near St Gallen in Switzerland where they were thrown by the impact.

Cluster of Zunil Crater Secondaries. Image: NASA/JPL/University of Arizona

When a probe was due to land on Mars’ moon Phobos , (it never made it), Jay Melosh was asked to model how much Martian material would be on it. They wanted to know if they ought to quarantine Phobos samples in case they contained Martian bugs. The answer was – there would be not a lot, but some.

The clear picture of all this evidence is that impacts on Earth, particularly the early earth when impacts were common, would leave fragments on the moon. Euan Nisbet (Royal Holloway, University of London) gave a whistle-stop tour of the early earth. If pieces of this sit on the moon, we would expect to see zircons (very durable), komatiite lava (olivines with distinctive compositions) and maybe sediments. All very distinctive from lunar rocks.

Dave Waltham (Royal Holloway) talked us through how the earth-moon distance has increased over time. Using an equation created by Charles Darwin’s son and armed with only 3 data points (distance now, zero at moon formation and data from some awesome 620Ma tidal sediments) he constrained distance over time. The moon was very close very early on, but it reached half of the current distance with 10 million years. So the moon was a closer target for bits of the early earth, but not by as much as we’d thought.

One of the joys of the conference was the variety of scientists on display. This ranged from slick suit-wearing committee-men to wild-eyed, wild-haired ‘crazy scientist’ types. There were geochemists who talked about the difficulty of getting good data points, geologists who showed pictures of rocks “because they are pretty” and astronomers who, while restrained, where keen to link events on earth to astronomical causes.

We also had an experimentalist, Mark Burchell (University of Kent). It is a truth universally acknowledged that a man in possession of a good gas gun must be in want of interesting things to shoot out of it. Things such as lumps of shale, yeast-infused samples and seeds are interesting – firing samples at up to 5 kilometers a second from a ‘two-stage light gas gun’ into sand is a pretty good approximation of a piece of earth landing on the moon. Very interestingly, samples of shale retain their biomarkers, chemicals indicative of life. Yeast and bacteria also survive the impact, making ideas of simple life moving between Mars and earth plausible. Plant seeds wouldn’t survive the journey though, so we don’t need to worry about ‘Day of the Triffids’ style alien-plant invasions. Part of me wonders if Mark Burchell wasn’t a little disappointed that the seeds didn’t survive, removing the need to move onto higher forms of life. He strikes me as a man who’d relish the challenge of firing a flea at cosmic velocities.

Layers of lava visible in lunar crater wall. Image: NASA/GSFC/Arizona State University

Finding the evidence

The phrase “the relatively accessible surface of the moon” is definitely something only a scientist would say. For a Professor of Planetary Science such as Ian Crawford (Birkbeck, University of London) the moon is our backyard. He spoke passionately about the scientific benefits of a human presence on the moon.

The samples returned by man to the earth (mostly in Apollo missions) are all from low latitudes on the near side. A return mission to the moon should focus on sampling high latitudes and widening our coverage of lunar material.

Ian Crawford has given much thought to suitable areas to explore lunar geology. He has identified the potential of layers between lava flows as repositories of ancient materials. Consider two overlapping lunar lava flows. Chances are there is a layer of material in between the two that landed on the surface during the gap between eruptions. This material could include earth meteorites, ‘normal’ meteorites, samples of the solar wind and more exotic material from the wider galaxy. Preserved by the upper lava flow, the age of the material can be constrained by dating the surrounding lava flows.

Slightly more prosaically, dating of events on all terrestrial bodies apart from the earth is based on cratering rates derived from the moon. The more craters on a surface on Mars, the older it is. Estimates of its absolute age are derived from studies of lunar cratering. Data of more lava flow surfaces on the moon would help refine our understanding of cratering rate over time and so improve dating of events across the solar system.

Katherine Joy (University of Manchester) spends her time studying rock samples from the moon. Some of these are Apollo samples, some are meteorites found on the earth. She’s already found some amazing things such as a meteorite in a meteorite. A tiny piece of rock floating around the solar system landed on the moon and got incorporated into the surface layers (the regolith). Another large chunk hit the moon and fractured the regolith sending pieces hurtling into space. One of these lunar fragments, containing the older meteorite within it, fell to the earth as a meteorite.

A portion of her research is identifying these fossil meteorites in samples of regolith. She has enough data to identify a suite of 3.9 Ga meteorites that are noticeably different from modern-day meteorites found on the earth. This work involves painstaking analysis to identify the individual fragments and analyse their chemistry. She is able to distinguish between fragments of the moon and pieces of meteorite – these techniques would allow her to recognise a piece of rock from the ancient earth. So one day, maybe right now, a scientist in Manchester will jump up in excitement, eager to share the amazing fact that they have before them a sample from earth’s early history, a little piece of evidence that’s survived four billion years and two journeys through space to bear witness to an otherwise lost time.

Image of Moon from ‘via moi’ on Flickr.

What came from outer space

Posted on 20 August, 2012 by Metageologist

I admit it. I’m geocentric. Not in the old-fashioned sense, I’m not that eccentric. I don’t believe the earth is the physical centre of the universe, but it certainly feels that way. The universe, space, the wonders of the solar system are all very well, but emotionally they are all ‘out there’ somewhere. A Twentieth Century geological education contributed to this – everything on the earth was explained by something else on the earth. For example we were taught about the increasingly odd forms of late Cretaceous ammonites. The implication was that their extinction was due to them becoming decadent and depraved, like the late Roman Empire. When the theory that they and the dinosaurs where killed by a meteorite impact arrived, it felt somehow unnecessary (and yet now seems so obvious).

Which is by way of justifying why my response to Dana Hunter’s Accretionary Wedge call for ‘out of this world’ posts on exogeology has its feet planted firmly on terra firma. One of the most interesting strands of studies of this planet in the last 20 years has been discovering the many terrestrial features caused by external influence. Often ‘it came from outer space’ is science fact, not science fiction.

Meteorites are the most dramatic way in which ‘space’ affects the earth. Lumps of rock and/or iron hitting the earth at cosmic velocities (more than 11 kilometres a second) leave some dramatic traces.

Meteor Crater, Arizona. Courtesy of Scott Tanis on Flickr. http://www.flickr.com/photos/8376919@N02/3905452932

What is striking about meteor craters on the earth is how few there are. There are so many craters on Mars that studies of its geology use their areal density to estimate the age of the surface. A bit like people’s skin, old craggy areas look different to smooth new ones. By contrast, there are fewer than two hundred impacts recognised on earth. Partly this is due to a thicker atmosphere, mostly due to high rates of erosion and geological activity and partly because there are many more traces of impact left to be found.

Most large round features on earth have been investigated. Some small craters no doubt remain to be found but there are other ways to find traces of ancient impacts. Some believe that we should be looking in the sedimentary record and that many more impact craters remain to be found, hidden away in oil company seismic data.

In the last few decades geologists have recognised a range of distinctive features found in rocks that have suffered/enjoyed an impact. Structural features within the impact include shatter-cones, fractured rock and faulting. On a microscopic scale, features such as shocked quartz are distinctive traces of the sudden massive stresses of impact. For large impacts, the energy turns into heat that melts and vaporises rocks. If a big enough hole is made, there will be extensive metamorphic changes to deep rocks due to the pressure release of removing kilometres of rock.

Another way to identify past impacts is to find material thrown out beyond the crater area. This can include thick layers of fragments and glass, called suevite. Molten glass may be thrown up into space, falling again as spheres of glass over a wide area of the earth. These may be quite thin layers in sedimentary sequences. These surely represent the most extensive record of past impacts but not many have been recognised to date, so keep your eyes peeled. For recent impacts, the glass lumps may still lie on the surface, where they are known as tektites. The largest set of tektites is found in a vast ‘strewn field’ stretching from south-east Asia into Antarctica and yet no-one has yet the crater that was surely formed at the same time.

The hunt for impacts continues. In the last few months researchers have published evidence for two more. A huge old one in Greenland has long been eroded away, but there are lots of distinctive features in the deep rocks that sat below it, which is all that remains. A smaller more recent one in Canada shows contorted strata and lots of pretty shatter cones (fractures formed by the impact).

Shatter cones from Prince Albert crater. Courtesy of University of Saskatchewan. http://www.flickr.com/photos/usask/7644732242/in/set-72157630747745754/

The most inventive method of finding evidence of past impacts is to study ancient myths. Some link myths about ‘fire from the sky’ to impacts that occurred in human history and seek to explain flood myths by a cometary impact.

My awareness of my own geocentric bias makes me sceptical of such things. If I was an astrophysicist, aware of the sheer volume of space debris, I would be much more inclined to explain features of the earth in terms of impacts. Perhaps too much so. I read someone once trying to explain the South African Cape Fold belt as caused by the Vredefort impact. This was nonsense.

Of course this is how science progresses, by bringing perspectives from different disciplines together as a way of stimulating new research. The link between Chixulub and the K-T extinction is pretty clear (but still debated). What is interesting about many mass extinctions is that there are just so many plausible mechanisms. The Deccan Traps may have killed the dinosaurs just as much as a meteorite. In a similar way the end-Jurassic extinction is close in time both to an impact and extensive vulcanism. The only way to take this argument further is by careful study of the evidence, most of which sits in the rocks.

Study of earth’s early history removes any lingering doubts that earth can be studied in isolation from its surroundings. The earth formed within a dusty disc around the new sun 4.56 billion years ago. During the earth’s first few 100 million years it was constantly being struck by other pieces of debris. The best current theory for the formation of the moon is the ‘giant impact hypothesis’. This suggests that the proto-earth was struck by another proto-planet the size of Mars. The impact resulted in two separate blobs which formed the earth and the moon. The energy of such an impact left both bodies completely covered in a magma ocean. Any water in the earth would be boiled off, meaning that our atmosphere and oceans are all derived from water from comets that have hit the earth since. We are all made of star dust, but let’s not forget the comet juice.

My favourite link between the earth and beyond is only an idea so far, but a beautiful one. On earth we find rare meteorites that came from Mars and the moon. When one day we study the moon in more detail, perhaps we’ll find pieces of earth on there. The period when the most impacts hit earth (sending bits flying off) is also the time when we have the fewest rocks preserved on earth. What if the oldest earth rock still in existence is actually to be found on the moon?

To end, a look at earth from the outside to remind me that our planet, endlessly fascinating as it is, is only a tiny dot in space. ‎