New study shows that the asteroid, rather than volcanism, likely caused the end-Cretaceous dinosaur extinction

July 22, 2020 • 8:45 am

Over the past few decades, there have been many arguments about what caused the extinction of the dinosaurs at the end of the Cretaceous—about 66 million years ago. One theory is that volcanic eruptions from the Deccan traps, a volcanic complex in what is now India, spewed carbon dioxide and sulfur dioxide into the atmosphere. This weakened the Sun’s radiation striking our planet, cooling the Earth and leading to the destruction of habitat suitable for dinosaurs.  Some support offered for this theory is that volcanic eruptions were likely responsible for the end-Permian extinction about 252 million years ago. That was the most dramatic extinction in life’s history, wiping about 96% of all marine species, 70% of terrestrial vertebrate species, and 83% of all genera. Three-quarters of all the species on Earth went extinct in a relatively short period of time.

Most paleobiologists, though, think the end-Cretaceous extinction was due to the collision of an asteroid with the earth right before the dinos started going extinct: the famous 10 km asteroid whose collision left a 180-200-km diameter crater in the present Gulf of Mexico. This spewed dust, ash, sulfates, and other matter into the atmosphere, also causing global cooling (the “impact winter”) and habitat loss for dinos. This extinction was very rapid.

Here’s a reconstruction of that impact, with the Wikipedia caption, “An artist’s rendering of an asteroid a few kilometers across colliding with the Earth. Such an impact can release the equivalent energy of several million nuclear weapons detonating simultaneously.”

Several million nuclear weapons!

Of courses these causes could have acted together, but a new paper in PNAS (click on screenshot below, pdf here; reference at bottom), used complicated models to judge which of the two causes was most important. They conclude that the asteroid was the primary factor, and in fact the volcanism could have even tempered the global cooling caused by the asteroid.

Now judging the climate models, which coupled estimates of the effect of aerosols and other ejecta on climate change, and then the effect of climate change on suitable habitat for dinosaurs, is beyond my pay grade. But the authors do conclude that the Deccan volcanism was highly unlikely to have contributed to the end-Cretaceous extinction, with the asteroid impact more likely. (It’s not just the models in this paper, either; other recent work suggests that the volcanism was too prolonged to account for the relatively quick extinction, nor was it likely to have caused sufficient climate change that would affect habitats in the required ways.)

The asteroid models show a relatively quick onset of subfreezing temperatures that would have lasted 3-16 years, with temperature recovering only after 30 years. That would have accounted for the geologically instantaneous extinction of the dinosaurs. (Some aquatic reptiles like turtles and crocodiles managed to find refuges that kept their lineages going.) Further, volcanic activity could have created a kind of “greenhouse effect” that actually ameliorated the effect of the asteroid.

Here are a few figures from the models showing the effect of volcanism vs. asteroids on suitable dino habitat (the redder the color, the more suitable the habitat would have been for non-avian dinosaurs). The first three show a control without Sun dimming (A) and then a 5% dimming (B) and a 10% dimming (C). 5% was enough to do them in.

These two figures show the same effect on habitat with the author’s Deccan-trap volcanism model. As you see, there’s plenty of suitable habitat left (in red):

And the effect of the asteroid. (F) assumes that the volcanoes were inactive, and (H) that they were active. No dino habitat in either case:

You can judge the models for yourself if it’s within your pay grade. Suffice it to say that, despite the assertiveness of the paper’s title, this won’t settle matters, as one can always quibble about the details of the model. But assuming the models are reasonable, it does support the consensus view that it was an asteroid, not volcanoes, that brought the dinosaurs to grief.


Chiarenza, A. A., A. Farnsworth, P. D. Mannion, D. J. Lunt, P. J. Valdes, J. V. Morgan, and P. A. Allison. 2020. Asteroid impact, not volcanism, caused the end-Cretaceous dinosaur extinction. Proceedings of the National Academy of Sciences 117:17084-17093.

65 thoughts on “New study shows that the asteroid, rather than volcanism, likely caused the end-Cretaceous dinosaur extinction

    1. Douglas Preston:

      “If, on a certain evening about sixty-­six million years ago, you had stood somewhere in North America and looked up at the sky, you would have soon made out what appeared to be a star. If you watched for an hour or two, the star would have seemed to grow in brightness, although it barely moved. That’s because it was not a star but an asteroid, and it was headed directly for Earth at about forty-five thousand miles an hour. Sixty hours later, the asteroid hit. The air in front was compressed and violently heated, and it blasted a hole through the atmosphere, generating a supersonic shock wave. The asteroid struck a shallow sea where the Yucatán peninsula is today. In that moment, the Cretaceous period ended and the Paleogene period began.”

      1. 45,000 miles per hour is an astonishing speed, and for something that large! Now, they say that is very unlikely to happen again. If it ever does, we should be able to predict it years in advance, and perhaps, do something about it. We are advised not to lose any sleep over it, but it can come at you in dreams.

        1. The speed at 20 km/s is lower than Earth orbital speed at about 30 km/s, so Earth chased the impactor and the impact velocity was low in the usual distribution.

          When the dimension of the impactor is comparable to the atmosphere scale height (~ 8 km) there is not much of a brake distance …

  1. I just want to throw my two cents in regarding the noted “assertiveness” of the paper’s title: I really wish that scientific papers, at least, wouldn’t fall prey to the tendency toward unwarranted certainty (or the pretension to it) that’s so prevalent in so many other facets of society. It would be so refreshing to have read, “Our models strongly suggest that asteroid impact, not volcanism, caused the end-Cretaceous dinosaur extinction,” or something along those lines.

    1. I tend to lean the other way.

      First in news media, I hate it when articles have titles essentially saying “X may cause Y.” In most cases, we already knew that it was a possibility; what we are looking for is something stronger than that.

      Second, in article titles, I read them as asserting what the article contains and shows, not the final conclusions of the scientific community on the subject. This article does show the asteroid being the cause; that the title says so is helpful to me, not a problem.

      1. I do get where you’re coming from, but I think erring on the side of cautious wording is just better in a scientific paper. I was reading a blog by a physicist yesterday who compared an old article from the early 20th century, with appropriate caveats on the conclusions and generalizability of the research involved, to modern articles that all too often claim to be groundbreaking and revolutionary – whether or not those claims are REALLY justified.

        The paper above cannot CONCLUSIVELY say that “the asteroid did it,” not as convincingly as, say, “On the Electrodynamics of Moving Bodies” demonstrated the principles of Special Relativity.

        The fact that we crave certainty, or something close to it, is no excuse for scientists to pander to that craving. And in the media in general, I think if they erred more on the side of caution presenting scientific findings, especially in medicine, the public would be less likely to be dismissive of the science of, say, nutrition or epidemiology, because they wouldn’t have been told “definitively” that, for instance, eating eggs is bad for you, only to have that conclusion put in doubt by future, better information.

          1. If the reality of the situation is a qualified one, then to use qualifiers is simply to make the most accurate choice of words. This seems to be a reasonable choice for a scientific endeavor.

    2. “New study … likely caused … ” Well that seemed to cover the bases, without any over dramatisation you normally read in the media. I actually thought the Yucatan Peninsula meteor impact zone was already assumed to be the conclusive precursor to the rapid cooling event, and this was just another study to verify this. I thought this was really well written.

      1. Yes, it is the scientific consensus. However, a researcher named Gerta Keller managed a resurgence for the Deccan trap theory (which she advocates) by presenting herself as a brilliant but discriminated scientist coming from an Oppressed Group (i.e. women), and supporters of the impact theory as oppressive patriarchs.

  2. I have wondered about the asteroid’s angle of incidence- the pattern near the Yucatán peninsula appears circular, as if the asteroid was a direct hit. That’s astonishing if my surmise is true – usually I thought there would be some angular approach.

      1. That paper is awesome

        This must be a variable that NASA considers in the problem of NEO impacts – the angle. If they can guide an NEO to a low angle and get it to skim a large tract of uninhabited area, or a rapidly evacuable (evacuatable?) area, that’d be better … maybe.. than a direct hit.

    1. The most likely impact angle is 45 degrees, and the 60 degree paper that savage links to have 45 degree as almost as good fit – it is a good fit to the impact distribution in any case. As I pointed out above it is the low impact velocity of ~ 20 km/s that is odd, so instead of a head on angle Earth likely chased the impactor.

      In any case, for hypervelocity impactors – impactors that come in faster than the sound speed in rocks or ~2-3 km/s – there is a shock wave that dissipate energy (and spallate away crust, here to the dinosaur’s detriment – it was sulfurous and calciferous rocks). So you are naturally looking for a round remain of shock waves and other “don’t-care-much-about-the template” energy transport.

      IIRC it is *very* hard to see large elliptical impact craters, such an impactor had to be travelling almost parallel with the crust. There are a few on Mars that has been taken as evidence of deorbited earlier moons.

  3. I think I read that iridium deposits found concentric with the Yucatan crater also was
    strong evidence for the asteroid theory.

    1. The ‘iridium spike’ is a feature of the thin layer of material found worldwide at the K-T boundary. The presence if iridium is evidence that this distinctive layer was fallout from an asteroid impact since this element is rare on earth, but more abundant in some asteroids.

      1. “ … an asteroid impact …”

        is there something that rules out more than one impact? Is the distribution even?

        1. I thought I read somewhere that the thickness of the deposit and some other details were found to be greater in the Western Hemisphere, indicating that the impact crater would be on that side of the world. The crater itself was discovered by accident.

  4. Some aquatic reptiles like turtles and crocodiles managed to find refuges that kept their lineages going.

    I imagine that migratory species adapted to the median latitudes could just remain more in the tropics. Not that they would know to do this, but if their instincts are influenced by things like water or air temperature, it would be a case of just continuing to go to the temperature x latitudes every summer…only x has changed location.

      1. The Canadian Museum of Nature presents the asteroid theory regarding dinosaur extinction as absolute certainty, which surprises me, as I also felt that not all experts agree. I remember seeing a tv program on dinosaurs some years ago in which well-known palaeontologist Robert T Bakker (he of the omnipresent cowboy hat) stated that he felt disease (dare I say a pandemic?) was responsible. I don’t know whether he may have changed his mind on that. According to Wikipedia, Bakker sees no conflict between science and religion. Uh oh.

        1. To me, Bakker’s idea that disease was a main factor shows that one can be a talented paleontologist, but not have a good grasp of other fields, like epidemiology. The idea that diease wiped out such a broad and diverse (land animals and marine; plants and animals)spectrum of life is unlikely to the point of ridiculous (even if acounting for cascading effects in collapse of ecosystems). It just doesn’t pass the smell test. Combined with the complete lack of any evidence of disease, I don’t see why this idea gets the attention it does.

          1. One of the more…ill-thought-out….ideas in paleontology. We had a case-study of what diseases do to populations when they intermix in our own species. Yes, diseases exchange between populations–but they certainly didn’t cause an extinction! Limiting ourselves to fossil evidence you have the closing of the Isthmus of Panama, which again caused some extinctions but not wholesale destruction on the scale of the K/Pg extinction. At best, were Bakker right, we’d see dinosaurs from one continent come to dominate both areas.

            It’s an interesting idea to speculate on, but not tenable given even a cursory examination of the data.

          2. Bakker in 1985: “They know next to nothing about how real animals evolve, live and become extinct. But despite their ignorance, the geochemists feel that all you have to do is crank up some fancy machine and you’ve revolutionized science. The real reasons for the dinosaur extinctions have to do with temperature and sea-level changes, the spread of diseases by migration and other complex events.”

            Earlier ideas were more interesting:

            1910: […] Woodward argued that traits like large size, spiny coverings and lack of teeth seen in some later dinosaurs were signs that the group was approaching its inevitable end.

            1917: Franz Nopcsa suggested that dinosaurs may have developed overactive pituitary glands that led them to become pathologically gigantic […]. He also suggested that a “[d]iminution of sexual activity” may have played a role in their demise.

            1925: Paleobotanist George Wieland hypothesized that Tyrannosaurus rex survived on a diet of eggs. He argued that feeding its great bulk would have led it to consume the entire last generation of dinosaurs before they could even hatch, leading to their extinction. He also suggested that mammals may have driven the dinosaurs extinct by eating all of their eggs.

            * P. E. Raymond suggested that dinosaur brain size diminished over the course of the Mesozoic until, in effect, they became too stupid to live and went extinct.
            * W. E. Swinton argued that dinosaurs were driven extinct when the lakes and swamps they inhabited dried up.

            1962: S. E. Flanders suggested that at the end of the Cretaceous caterpillars began multiplying until they had so denuded the contemporary plant life that nothing was left for the dinosaurs, who starved to death


        2. It also disusses how cladistically speaking that’s not so- modern birds are dinosaurs.

          So, what did they have that the others lacked? (Or better, what did the others have or lacked that doomed them?)

          1. My understanding is that the line between birds and dinosaurs must always be drawn in such a way that Archaeopteryx can be classified as a bird.

    1. That an asteroid hit the Earth is more or less settled–the iridium and the impact crater sort of make doubting that difficult. That it caused the extinction is accepted by almost everyone who studies the events. There is a strong holdout of people who absolutely refuse to consider the idea of an impact generating a mass extinction, however. I don’t want to lump them in with Creationists, because these folks are at least doing good science (mostly marine microfauna)–but they’re doing so in service of a pre-established conclusion and are ignoring a tremendous amount of evidence to do so. Their methods are good, their intent is less so. Comes from having science done by humans.

      This isn’t a controversy. The issue is settled among the vast majority of scientists who research this topic (I say this as someone who researched this topic, before moving on to Quaternary mammals). It’s a “controversy” in the sense that there’s a group trying to create a controversy.

      1. Alvarez himself was controversial, was he not? It probably doesn’t hurt to humor the rebels for a while.

    1. I can access the Nature article. The Clovis First debate is in many ways similar to the Asteroid/Vulcanism debate for the end Cretaceaous.

      The Nature paper, while interesting, does not have unequivocal evidence of human occupation (stone tools in a well-dated layer) at 30,000 years ago. They do have dates as much as 20,000 years ago (definitely pre-Clovis), and the site is at high altitude, which suggests humans were in the Americas long before they ever arrived at this site. Genetic studies also place the arrival closer to 30,000 years ago.

      But, nothing short of a well-dated unequivocal occupation layer at a site will convince the few remaining Clovis first deadenders. While an impressive discovery, this site falls a bit short of that.

      I don’t have a dog in this fight (I’m not an Americanist prehistorian). But I’ve followed it closely. For the record, I am certain people arrived in the Americas well before 15,000 years ago. I am fascinated with the theories (kelp highway) that have been developed since Clovis days. But, I’d love to see some more sites (they will be found)with rock-solid dates and evidence of occupation.

      1. Some early criticism to the two works here: .

        Prof David Meltzer, from Southern Methodist University in Dallas, Texas, who was not involved in the research, said the findings were “interesting”. But he explained: “It is not enough to argue that the stone specimens could be cultural (artefacts), one has to show that they are not natural.” Natural processes could mimic some types of stone tool, Prof Meltzer said.

        Second, he explained: “With a stone tool tradition that long-lasting, one expects it would have been far more widespread in the region, raising the question of why that technology hasn’t been spotted elsewhere,” adding: “Perhaps more important, with modern humans one expects to see evidence of technological and cultural change over such a long span of time.”

        Finally, he said, “the cave is 1,000 metres above the valley floor, but leaving aside the issue of why not camp closer to the valley floor, why keep coming back to that same place on ‘a relatively constant basis’ over such a long period of time? I find that curious. Not many sites have that kind of long term repeated occupation, unless there is something quite useful / available at the spot”.

        I would add that, as the paper’s admit, there isn’t any genetic evidence (yet) either in Beringia or further south that such an early gene flow would have taken place. The current evidence, which has helped dethrone the Clovis first hypothesis, is of migration – likely along the kelp highway [ ] – ~ 17 – 20 kyrs from Beringia and quickly flooding the continents.

        1. Oops. ” paper’s” = papers.

          But I’m confusing the papers here. The criticism published in BBC (which is early, they have not looked at the stone tools at a guess and if so shouldn’t criticize that part) is of the find paper. It is not criticizing the bayesian dating model I think (which is an interesting work all on its own, if the find datings stand up).

          It was likely the find paper that said they didn’t find any DNA, nor have any extant sequence data supporting the find.

  5. So did vulcanism cool the earth, as proponents of the vulcanism extinction claim, or warm the earth as the authors claim? Given the vast amounts of CO2 that must have been added to the atmosphere, I am inclined to believe vulcanism should have warmed the earth, ameliorating the impact winter caused by the asteroid impact, as the authors claim. But who knows? Since the Deccan vulcanism began before the impact, what do we know, if anything, about what was happening to the climate before the impact?

    1. It did both. At certain stages it warmed the atmosphere, at others it cooled the atmosphere. The impact was incredibly complicated, and we’re still working out exactly what was involved. The lithology of the rock it hit, how long the material stayed in the upper atmosphere, upper-atmospheric wind patterns, and a host of other factors need to be taken into account. Then, of course, you have to dig up the evidence that your model is correct (which means figuring out what evidence can be dug up–not an easy task).

      In terms of gross effect, I think warming won out, but I may be mistaken. But the devil’s in the details.

      1. It amazes me how paleontologists and earth scientists can reconstruct events that happened countless millions of years ago. A couple of hundred years ago, we would have been clueless how to do it.

  6. This is not my area of expertise, but I had heard a theory that the Deccan Flats activity, whcih is more or less on the opposite side of the earth from the asteroid impact, might in fact have been activated by the asteroid strike.

    From Wikipedia:
    Work published in 2014 by geologist Gerta Keller and others on the timing of the Deccan volcanism suggests the extinction may have been caused by both the volcanism and the impact event.[10][11] This was followed by a similar study in 2015, both of which consider the hypothesis that the impact exacerbated or induced the Deccan volcanism, since the events occur at antipodes.[12][13]

    1. Not my area of expertise either, but I am curious as to whether or not there is a significant amount of volcanic ash deposit accompanying the “iridium spike” layer.

    2. The published outflow datings are often a bad fit to the impact (and extinction) event.

      The liquid core of Earth dampens the deep seismic waves. And the surface waves are quickly dampened too, I thought, within 1000 kms they would be too weak to trigger anything (and I don’t think increased volcanism is seen outside the local area affected by the impact). But I admit I can’t find that reference again. I found this, which is recent (2019):

      The dates, and the increase in lava volume after the impact, also line up with a previous suggestion by Sprain’s team, including her former adviser, Paul Renne, a geochronologist at the University of California, Berkeley, that the two events are directly related: The impact might have struck the planet so hard that it sent the Deccan Traps into eruptive high gear.

      The second study used a different method to date the eruptions. A team including Keller and led by Blair Schoene, a geochronologist at Princeton, looked at zircon crystals trapped between layers of basalt. These zircons can be precisely dated using the decay of uranium to lead, providing time stamps for the layers bracketing the eruptions. The zircons are also rare: It was a full-time job, lasting several years, to sift them out from the rocks at the 140 sites they sampled.

      The dates recovered from the crystals suggest that the Deccan Traps erupted in four intense pulses rather than continuously, as Sprain suggests. One pulse occurred right before the asteroid strike. That suggests the impact did not trigger the eruptions, he says. Instead, it’s possible this big volcanic pulse before the asteroid impact did play a role in the extinction, Schoene says. “It’s very tempting to say.” But, he adds, there’s never been a clear idea of how exactly these eruptions could directly cause such extinctions.

      Though the two studies differ, they largely agree on the overall timing of the Deccan eruptions, Schoene says. “If you plot the data sets over each other, there’s almost perfect agreement.”

      [ ]

      So I guess YMMV at this time.

  7. I didn’t read the paper but I remember reading of the possibility that the asteroid strike in Yucatan caused the volcanic activity in the Deccan Flats. A shock wave would have traveled from one side of the earth to the other, converging with great intensity at the spot roughly opposite to the Yucatan Peninsula.

  8. The huge amount of kinetic energy converted to heat (more than a billion times the energy of the atomic bombings of Hiroshima and Nagasaki) produces a huge essentially point explosion that creates a circular crater over a wide range of angles of incidence (10km asteroid versus 200 km diameter crater). In other words a circular crater is, by far, the most probable result of this large an impact explosion.

  9. FWIW, one of the sites used to correlate the iridium layer to the extinction – all the dinosaur fossils are beneath it – is Makoshika State Park in the Badlands just E of Glendive MT. Fascinating terrain!

  10. That was the most dramatic extinction in life’s history, wiping about 96% of all marine species, 70% of terrestrial vertebrate species, and 83% of all genera. Three-quarters of all the species on Earth went extinct in a relatively short period of time..

    I’m a novice on these matters, but I know natural selection is an indelible factor of life once it takes hold (at least on earth); metaphors abound. Cobb’s brain book is a perfect example of how just one aspect of evolution has only been “solved” by metaphor. Yet, even after such catastrophic species’ extinctions, here we are. Evolution also tells us we probably won’t live as a species indefinitely. Long enough though for me and you and generations and such… Ceiling Cat and Covid willing (wink). Humans insatiably seek for more life; is that a feature or a bug? Probably a mixture of both.

    1. Perhaps the meteor was the most dramatic, but I’d say the Snowball Earth (hypothesized) extinctions, would be close rivals, although not as sudden.
      I’m now reading Cobb’s brain book, and yes, he traces all the metaphors for the activity of the brain. From hydraulic, to the telegraph, and the computer analogues.

  11. This has been an interesting debate since the discovery of the asteroid impact that the Deccan Traps was the Antipode, via concussive waves through the Earth. India was moving into the underbelly of Asia via Plate Tectonics, building the Himalayas. The problem has been radiometric dating and correlation with the impact. The span of time between the two has been difficult to get less than several million years. The Traps cover atleast 1,000,000km2 of area and up to 2000km in thickness at minimum. Erosion has plagued this area significantly due to climate conditions. Some think this area was vastly greater, going well into what is now the Indian Ocean. 3D FEM/FEA modelling has suggested that the Deccan Traps was not the Antipode, due to the severity of the expanse of the Traps. The pressure should also show signs of deformation culminating with the Traps, but at this time there is no evidence to support that concept. I do have degrees in Geophysics and Geologic Oceanography and have conversed with my Undergraduate Professor at the U of Minn. on this subject (on and off again) for nearly 20 years.

  12. Thanks! I read the paper at the time, but it is always useful to see other takes.

    other recent work suggests that the volcanism was too prolonged to account for the relatively quick extinction,

    It even looks like, which IIRC the paper covered, the Deccan volcanism came in two pulses, and also mostly before and after the extinction event. Not a good fit.

Leave a Reply