Read about new data from lake-bed sediments, which suggest that a major impact around 12.8 thousand years ago may have triggered a return to glacial conditions at the start of the Younger Dryas.
Read about new data from lake-bed sediments, which suggest that a major impact around 12.8 thousand years ago may have triggered a return to glacial conditions at the start of the Younger Dryas.
Note: Earth-Pages will be closing as of early July, but will continue in another form at Earth-logs
The New Yorker magazine normally features journalism, commentary, criticism, essays, fiction, satire, cartoons, and poetry. So it is odd that this Condé Nast glossy for the chattering classes snaffled online what may be the geological scoop of the 21st century so far (Preston, D. 2019. The day the dinosaurs died. The New Yorker 8 April 2019 issue). The paper that lies at the centre of the story had not been published and nor had the issue of The New Yorker in which Douglas Preston’s story was scheduled for publication. The very day (29 March 2019) that Britain was thwarted of its Brexit moment the world’s media was frothing with news about the end of another era; the Mesozoic. The paper itself was published online on April Fools’ Day with a title that is superficially arcane (DePalma, R.A. and 11 others 2019. A seismically induced onshore surge deposit at the KPg boundary, North Dakota. Proceedings of the National Academy of Science, early online publication;p DOI: 10.1073/pnas.1817407116). But its contents are the stuff of dreams for any aspiring graduate student of palaeontology; the Indiana Jones opportunity.
An ‘onshore surge deposit’ occurs at many Western Hemisphere sites where the K-Pg boundary outcrops in terrestrial or shallow-marine sediments. The closer to the Chicxulub crater north of Mexico’s Yucatan Peninsula the more obvious they are, for they result from the tsunamis that immediately followed the asteroid impact. Lead author Robert DePalma, now of the University of Kansas, became focussed on the dinosaur-rich, Late Cretaceous Hell Creek Formation of North Dakota as an undergraduate. Accepted for graduate studies he was directed to a project on the fauna of lacustrine sediments close to the K-Pg boundary layer, which is well-known in the area, and that’s what he has been engaged with ever since. In 2012 he was guided to a remarkable locality by a rockhound, disappointed because it exposed extremely fossil-rich sediments but was so soft that none could be extracted intact with a hammer and chisel. It turned out to have resulted from a surge along a sinuous river that had washed debris onto a point-bar deposit at the inside of a meander. The debris includes remains of both marine and terrestrial organisms and shows clear signs of having been swept upriver, i.e. from the sea and possibly the result of a tsunami. Being capped by a thin, iridium-rich layer of impactite, the 1.5 metre surge deposit is part of the K-Pg boundary layer, and probably represented only a few hours before being blanketed by ejecta.
This Event Deposit comprises two graded, fining-upwards units and thus two distinct surges, with a thin mat of vegetation fragments immediately below the Ir-rich clay cap that also contains sparse shocked quartz grains. The Event Deposit contains altered glass spherules throughout, which cgradually become smaller higher in the 1.5 m sequence. Some of the larger spherules produced ‘micro-craters’ in the sediments. Fossils include marine ammonite fragments (some still nacreous) and freshwater fish (paddlefish and sturgeon). The fish are so complete as to suggest an absence of scavengers. The paper itself contains little of the information that dominated Preston’s New Yorker article and the global media coverage. This included clear evidence that the fish ingested spherules, found clogging their gills and possible causing their death. There are examples of spherules embedded in amber formed from plant sap, which suggests sub-aerial fall of ejecta, and among the marine faunal samples are teeth of fish and reptiles (see DePalma et al’s Supplemental Data). The most startling finds reported by Preston are nowhere to be found in DePalma et al’s paper or its supplement. These include possible dinosaur feathers; a fragment of ceratopsian dinosaur skin attached to a hip bone; a burrow containing a mammal jaw that penetrates the K-Pg boundary layer; dinosaur remains, including an egg (complete with embryo) and hatchlings of dinosaurian groups found at deeper levels in the Hell Creek Formation. Previously, palaeontologists had found no dinosaur remains less than 3 m below the K-Pg boundary layer anywhere on Earth, prompting the suggestion that they had become extinct before the near-instantaneous effects of Chicxulub, and were perhaps victims of the general effects of the Deccan Trap volcanism. If verified in later peer-reviewed publications, DePalma et al’s work would help resolve the gradual vs sudden hypotheses for the end-Cretaceous mass extinction.
Preston reports some academic scepticism about DePalma’s work, and emphasises his showmanship at conferences; for instance, he named the site ‘Tanis’ after the ancient city in Egypt featured in the 1981 film Raiders of the Lost Ark. There are geophysical queries too. If the inundation was by the on-shore effects of a tsunami it doesn’t tally with the abundance of ejecta fallout of glass spherules: tsunamis propagate in shallow seawater at speeds less than 50 km h-1 and more slowly still in channels, whereas impact ejecta travel much faster. This is acknowledged in the paper’s supplement, and the paper refers to a seiche wave activated by seismic waves associated with the Chicxulub impact which could have arrived in North Dakota at about the same time as its ejecta blanket. The paper’s authorship includes the imprimatur of other authorities in different geoscientific fields, including Walter Alvarez, jointly famed with his father Luis for the discovery of the K-Pg boundary horizon and its impact connections in 1981. So it carries considerable weight. No doubt further comment and further papers on the Tanis site will emerge: DePalma has yet to complete his PhD. It may become the lagerstätte of the K-Pg extinction; in DePalma’s words ‘It’s like finding the Holy Grail clutched in the bony fingers of Jimmy Hoffa, sitting on top of the Lost Ark.’ …
These days reports of geological evidence for asteroid impacts are not regarded with a mixture of disbelief, wonder and foreboding: well, not by geologists anyway. But for such a small area as Britain now to have three of widely different ages and in easily accessible places is pretty good for its brand as the place to visit for practically every aspect of Earth history. The first to be discovered lies at the base of Triassic mudstones near Bristol (see Britain’s own impact) and would need some serious grubbing around at a former construction site. The next to emerge was located in one of the best geological districts in the country at several easily accessed coastal exposures in Northwest Scotland. A glass-rich ejecta layer occurs in the basal Torridonian Stoer Group on Stac Fada, Stoer, Sutherland (UK National Grid Reference 203300, 928400). The most recently found (Drake, S.N. and 8 others 2018. Discovery of a meteoritic ejecta layer containing unmelted impactor fragments at the base of Paleocene lavas, Isle of Skye, Scotland. Geology, v. 46, p. 171-174; doi:10.1130/G39452.1) is on the Inner Hebridean island of Skye at the base of its famous Palaeocene flood basalt sequence (UK National Grid Reference 155371,821112).
The last is perhaps the most spectacular of the three, as it contains the full gamut of provenance, matched only by material from the drill core into the 66 million year-old Chicxulub crater. The 0.9 m thick debris layer rests directly on mid-Jurassic sandstones beneath Palaeocene basalts of the North Atlantic Igneous Province (NAIP). The layer contains a basalt clast dated at 61.54 Ma, but is dominantly reminiscent of a pyroclastic ignimbrite flow as it contains glass shards. But there the resemblance ends for the bulk of small clasts are of quartz and K-feldspar, sandstone and gneiss. Zircons extracted from the debris show shock lamellae and give Archaean and Proterozoic ages commensurate with the local basement, but also with the bulk of the Scandinavian and Canadian Shields. So the impact could have been anywhere in such widespread terrains, although the enclosed basalt narrows this down to areas where basement is overlain by lavas of the NAIP. The Skye impactite contains unmelted meteorite fragments in the form of titanium nitrides alloyed with vanadium and niobium, metallic iron-silicon alloy containing exsolved carbon, and manganese sulfide.
Although it may be coincidental, the situation of the ejecta layer immediately beneath the Skye lavas, its containing a clast of basalt whose age corresponds to the oldest flows anywhere in the NAIP is fascinating. But the actual impact site is, as yet, unknown. Even so, the layer provokes thoughts about whether an impact may have been more than spatially related to the large NAIP flood basalt pile, preserved on either side of the North Atlantic. If the event was large, then surely the ejecta should be preserved near the base of the flood basalts elsewhere in NW Britain and further afield
One of the most eagerly followed ocean-floor drilling projects has just released some results. Its target is 46 km radially away from the centre of the geophysical anomaly associated with the Chixculub impact structure just to the north of Mexico’s Yucatan Peninsula. In the case of large lunar impact craters the centre is often surrounded by a ring of peaks. Modelling suggests such features are produced by the deep penetration of immense seismic shock waves. In the first minute these excavate and fling out debris to leave a cavity penetrating deep into the crust. Within three minutes the cavity walls collapse inwards creating a rebound superficially similar to the drop flung upwards after an object is dropped in liquid. This, in turn, collapses outwards to emplace smashed and partially melted deep crustal material on top of what were once surface materials, creating a crustal inversion beneath a mountainous ring of Himalayan dimensions that surrounds a by-now shallow crater. That is the story modelled from what is known about well-studied, big craters on the Moon and Mercury. Chixculub is different because the impact was into the sea and involved debris-charged tsunamis that finally plastered the actual impact scar with sediments. The drilling was funded for several reasons, some palaeontological others relating to the testing of theories of impact processes and their products. Chixculub is probably the only intact impact crater on Earth, and the first reports of findings are in the second category (Morgan, J.V. and 37 others 2016. The formation of peak rings in large impact craters. Science, v. 354, p. 878-882; doi: 10.1126/science.aah6561).
The drill core, reaching down to about 1.3 km below the sea floor penetrates post-impact Cenozoic sediments into a 100 m thick zone of breccias containing fragments of impact melt rock, probably the infill of the central crater immediately following the first few minutes of impact. Beneath that are coarse grained granites representing the middle continental crust from original depths around 10 km. The granite is intensely fractured and riven by dykes and pods of impact melt, and contains intensely shocked grains that typify impacts that produce a transient pressure of ~60 GPa – around six hundred thousand times atmospheric pressure. From seismic reflection surveys this crustal material overlies as yet un-drilled Mesozoic sedimentary rocks. Its density is significantly less than that of unshocked granite – averaging 2.4 compared with 2.6 g cm3. So it is probably filled with microfractures and sufficiently permeable for water to have penetrated once the impact site had cooled. This poses the question, yet to be addressed in print, of whether or not this near-surface layer became colonised by microorganisms in the aftermath (Barton, P. 2016. Revealing the dynamics of a large impact. Science, v. 354, p. 836-837). That is, was the surrounding ocean sterilised at the time of the K-T (K-Pg) mass extinction?; an issue whose resolution is awaited with bated breath by the palaeobiology audience. OK; so theory about the physical process of cratering has been validated to some extent, but will later results be more interesting, outside the planetary sciences community?
The work done by an asteroid or a comet that hits the Earth is most obviously demonstrated by the size of the crater that it creates on impact, should it have survived erosion and/or burial by sediments. Since some is done in flinging material away from the impact, the furthest point at which ejecta land is also a rough measure of the power of the hit. All this and much more derived from the kinetic energy of the object, which from Newton’s laws of motion amounts to half the product of the body’s mass and the square of its speed (mv2/2). It’s the speed that confers most energy; doubling the speed quadruples the energy. At a minimum, the speed of an object from far-off in space is that due to acceleration by the Earth’s gravitational field; the same as Earth’s escape velocity (about 11.2 km s-1). In March 1989 Earth had a close encounter with Newton’s laws writ large; an asteroid about 500 m across passed us with just half a million kilometres to spare. Moving at 20 km s-1 it carried kinetic energy of around 4 x 1019J. Had it hit, all of this immense amount would have been delivered in about a second giving a power of 4 x 1019 W. That is more than two hundred times greater than the power of solar heating of the day-side of the Earth. A small part of that power would melt quite a lot of rock.
As well as the glass spherules that are one of the hallmarks of impact ejecta on Earth and more so on the Moon’s surface, some of the larger known impact craters are associated with various kinds of glassy rock produced by instantaneous melting. Some of this melt-rock occurs in thin dykes, but sometimes there is an entire layer of once molten ‘country’ rock at the impact site. The most spectacular is in the Manicougan crater in Quebec, Canada. In fact a 1 km thick impact-melt sheet dominates most of the 90 km wide structure and it is reputed to be the most homogeneous large rock mass known, being a chemical average of every rock type involved in the Triassic asteroid strike. Not all craters are so well endowed with an actual sheet of melt-rock. This has puzzled some geologists, especially those who studied the much larger (160 km) Vredfort Dome in South Africa, which formed around 2 billion years ago. As the name suggests this is now a positive circular topographic anomaly, probably due to rebound and erosional unloading, the structure extending down 20 km into the ancient continental lithosphere of the Kaapvaal craton. Vredfort has some cracking dykes of pseudotachylite but apparently no impact melt sheet. It has vanished, probably through erosion, but a relic has been found (Cupelli, C.L. et al. 2014. Discovery of mafic impact melt in the centre of the Vredfort dome: Archetype for continental residua of early Earth cratering? Geology, v. 42, p. 403-406). One reason for it having gone undiscovered until now is that it is mafic in composition, and resembles an igneous gabbro intrusion. Isotope geochemistry refutes that mundane origin. It is far younger than the rocks that were zapped, and may well have formed as huge energy penetrated to the lower crust and even the upper mantle to melt a sizeable percentage of 2.7 to 3.0 Ga old mafic and ultramafic rock.
Oddly, the same issue of Geology contains an article that also bears on the Vredfort Dome structure (Huber, M.S. et al. 2014. Impact spherules from Karelia, Russia: Possible ejecta from the 2.02 Ga Vredfort impact event. Geology, v. 42, p. 375-378). Drill core from a Palaeoproterozoic limestone revealed millimetre-sized glass droplets containing excess iridium – an element at high concentration in a variety of meteorites. The link to Vredfort is the age of the sediments, which are between 1.98 and 2.05 Ga, neatly bracketing the timing of the large South African impact. Using reasonably well-constrained palaeogeographic positions at that time for Karelia and the Kaapvaal craton suggests that the glassy ejecta, if indeed they are from Vredfort, must have been flung over 2500 km.
From the days when advocates of impacts by extraterrestrial objects as explanations of geological features were widely regarded as ‘whizz-bang artistes’ a great many hats have probably been eaten, albeit in closely guarded privacy. In 1986, when beds of glassy spherules similar to those found in lunar soil and in the K-T boundary sequence were reported from early Archaean greenstone belts in Australia and South Africa, and deduced to have formed by an impact, the authors, Donald Lowe of Stanford University, USA and colleagues, were pounced on by those who thought they could plausibly explain the very odd rocks by unremarkable, Earthly processes. Subsequent work on their geochemistry overwhelmingly supported their formation by an impact of a large carbonaceous chondrite asteroid. And at one site, the Barberton Mountain Land greenstone belt in northeastern South Africa, there was evidence for at least three such impacts formed in a 20 Ma period. In hindsight, given the lunar bombardment history that peaked between 4 and 3.8 Ga, early Archaean rocks were a great deal more likely to contain materials formed by giant impacts than less antiquated ones.
Lowe has been steadily working on his original idea since then, his enthusiasm drawing in others. The latest focus is on evidence for other likely consequences in the Archaean record of the vast power unleashed by incoming asteroids travelling at speeds around 15 km s-1 (Sleep, N.H. & Lowe, D.R. 2014. Physics of crustal fracturing and chertdike formation triggered by asteroid impact, ~3.26 Ga, Barbertongreenstone belt, South Africa. Geochemistry, Geophysics, Geosystems, v. 15, doi:10.1002/2014GC005229). The damage at Barberton not only produced spherule beds but opened fractures on the shallow sea bed into which liquefied sediments, including some spherules, were injected. These swarms of up to 10 m wide cherty dykes extend up to 100 m below what was then the sea floor strewn with impact spherules, and contain evidence of successive pulses of sediment injection.
Sleep and Lowe explain these dyke swarms as fractures caused by seismicity associated with a major impact. Their complexity suggests extreme shaking for upwards of 100 seconds; far longer than that from large, tectonic earthquakes. The fact that cracks opened to accommodate the sedimentary dykes indicates extension of the affected crust, which the authors suggest resulted from gravitational sliding of the shocked surface sediments down a gentle slope. Possibly the sediments, including the direct products of impact, the spherules, were swept into the cracks by currents associated with tsunamis induced by the impact.
Interestingly, the spherules and dykes formed upon crust largely formed of mafic to ultramafic lavas, yet volcanism following close on the heels of the impact event was of felsic composition. Did the impact trigger a shift locally from oceanic magmatism to that characteristic of island arcs; that is, did it start a new subduction zone?