<link rel="stylesheet" id="wp-block-library-css" href="https://c0.wp.com/c/5.8.2/wp-includes/css/dist/block-library/style.min.css" type="text/css" media="all">
loader image

Roll back the clock 50,000 years and the world was a very different place. Up to 5 species of human were living side by side, including the Neanderthals and the hobbit, but now they’re all extinct (except us). That’s something I talk about a lot on this blog. Something I talk about less often is just how much the rest of the ecosystem has changed. Mammoths, woolly rhinos and dozens of other species of megafauna that lived alongside the Neanderthals had disappeared by around 10,000 years ago1. So where did all the mammoths go?

Were the captured by Hollywood for use in bad movies?

The term “megafauna” might conjure up impressive images of giants roaming the landscape, but really it basically means any animal larger than a deer. And that definition might make the animals seem a tad less impressive, but it makes these extinctions pretty mind-blowing. Some of the continents lost more than 95% of their megafauna2. Australia lost everything, apparently as a result of humans burning everything to the ground whilst hunting4. In some places it was a bit less severe, with Eurasia loosing just above 40% of its megafauna2

The rate at which megafauna went extinct around the world, with the darkness of the red meaning more exinction

The percentage of megafauna that went extinct; with the darkness of the red meaning more exinction

These extinctions coincide with two important events. First, modern humans (us) began leaving Africa for the first time and spreading throughout the world. Shortly after this was the ice age (actually one of many, so should be called the “last glacial maximum”). So for quite a while now scientists have been arguing back and forth whether it was us or the ice age responsible for killing off the mammoths1.

As with almost anything in science, it seems the answer is “it’s complicated”. For a while a consensus has been emerging amongst researchers: in places where animals had encountered humans (or our relatives, like the Neanderthals) before they had adapted to us. This includes places like Africa and Eurasia. As such these extinctions were less severe, and those which did occur could likely be attributed to the environment. Meanwhile the places that had never seen a hairless upright ape before, such as the Americas and Australia, were hit hard by the arrival of humans and we wiped out a lot of the megafauna living there2.

Bare with, this is perhaps the most difficult graph to explain. Basically the red box represents when modern humans arrived at various places, the blue bit represents climate change. The numbers above these time lines refer to when x number of species of megafauna went extinct. Finally, based on all this, the colour of the picture of megafauna represents

Bare with, this is perhaps the most difficult graph to explain. Basically the red box represents when modern humans arrived at various places, the blue bit represents climate change. The numbers above these time lines refer to when x number of species of megafauna went extinct. Finally, based on all this, the colour of the picture of megafauna represents what can be blamed for the extinction. Red means its our fault, blue is the environment. The brown animal means “uncertain”

So where did all the mammoths go? In Europe they’d been living alongside Neanderthals for hundreds of thousands of years, so when we showed up they knew to avoid our spears. So they managed to survive for quite a bit longer, with some groups surviving until the pyramids were built. But climate change eventually finished them off. Meanwhile the mammoths in America didn’t know how to deal with humans; so were quickly finished off by the first people to arrive.

The end? Not quite, as research published earlier this summer challenges this nice little narrative. A team from Denmark began examining a whole range of extinction data from the period, not just restricting themselves to megafauna. This didn’t change the data too much, with Eurasia still having less extinction than America. However, this larger data set meant they could do a more rigorous analysis; analysing each country with data and checking to see if these extinctions are correlated with either the climate or human arrival3.

This investigation had two very interesting results. First: Uruguay has lost the most large mammals over the past hundred thousand years. Second: this research failed to find any correlation between the change in climate and the extinction of large mammals. Sure, in some places the extinction was less strongly correlated with the arrival of humans; but even in these places it wasn’t linked to the environment. In other words, it simply took longer for humans to wipe out the animals that were used to them (like European mammoths); not that the climate was responsible instead. Where the animals weren’t adapted to us, we could commit our megafauna genocide much quicker. There may have been some link with climate in Eurasia, but this was relatively minor. Humans are still to blame for most of it3.

I suspect the people behind the “humans only killed them some places” hypothesis (who really need a snappier title for their idea) will hit back at this research. There are a few flaws in the paper that lead me to think it isn’t the final word on the issue. Still, they present a compelling case and it’ll be tough for them to overturn it. So I guess everyone should start feeling bad that we killed all the mammoths.

And maybe you should bookmark this post, for the next time someone starts trying to argue about how people used to live in harmony with the environment and all that drivel.


  1. Barnosky, A.D., 2004. Assessing the Causes of Late Pleistocene Extinctions on the Continents. Science, 306, pp.70-75.
  2. Koch & Barnosky, 2006. Late Quaternary Extinctions: State of the Debate. Annual Review of Ecological Systems, 37, pp.215-250.
  3. Sandom, C., Faurby, S., Sandel, B., & Svenning, J. C. (2014). Global late Quaternary megafauna extinctions linked to humans, not climate change.Proceedings of the Royal Society B: Biological Sciences, 281(1787), 20133254.
  4. Rule SBrook BWHaberle SGTurney CSMKershaw APJohnson CN. 2012 The aftermath of megafaunal extinction: ecosystem transformation in Pleistocene Australia. Science 335, 14831486.

Related posts


DrFinlay · 5th August 2014 at 3:29 pm

What I have read, which felt quite sensible/logical to me suggests that in Eurasia, when the ice came, a reduced number of mammoths and all the other megafauna retreated south. When the ice melted the previous broad leaved fauna was replace by grassland. Like present day elephants, mammoths were broad leaf eaters and struggled to thrive on grass – physically harder for them to eat and nutritionally less rewarding. So the mammoths were not able to re-establish themselves in large numbers and therefore vulnerable to a growing human population with a taste for barbeque. Do you want to megafauna that happy meal?

    Adam Benton · 5th August 2014 at 3:58 pm

    That does seem like a plausible idea; and the multiple glacial advances and retreats could well have “weakened” the mammoths somehow. However, the fact that they were some of the last surviving megafauna suggests to me that this didn’t play a huge role in their extinction

Graham Rutter · 5th August 2014 at 4:58 pm

Thanks! Very interesting; sounds reasonable to me. I guess that the megafauna extinction chart needs a bit of an update then! I’d be very interested to hear what you think the “few flaws” in the paper were though?

    Adam Benton · 5th August 2014 at 10:21 pm

    In the original idea North American extinctions were caused by humans. From what I can tell they divided up America into states for their analysis, so that would weight the data fairly heavily in the “humans killed everything direction”. Now they did analyse each country individually, so I don’t think this is a huge problem (which is why I didn’t bother really talking about it) but I think it is worth considering. Also, all their examples of the environment remaining stable whilst megafauna still went extinct are from those places where we already thought humans were responsible, like North America. Again, a minor quibble, but enough to the point where I don’t think the discussion is over just yet.

Wyrd Smythe · 5th August 2014 at 6:09 pm

So humans have been changing their environment for a very, very long time. Clever damned apes that we are, now we have the power to really change things! o_O

    Adam Benton · 5th August 2014 at 10:10 pm

    If humans were capable of wiping out 78% of large mammals with fire and spears, imagine what we’re doing now.

Cynthia Echterling · 5th August 2014 at 7:36 pm

I read somewhere that, with the last few glaciations, they came and went so rapidly that populations in refugia were not able to recover. They became very inbred and so did Neandertals. So with that and the new top predator, it was a double whammy. I have also wondered about humans coming into the Americas and Australia with their dogs. What about deseases spread by their fleas?

    Adam Benton · 5th August 2014 at 10:09 pm

    I know that was the case with Neanderthals, not so sure about mammoths (but then haven’t read much about them). Refugia weren’t really refuges they retreated too, just places where the species didn’t go extinct. So the population really crashed whenever this happened, and that could influence their extinction. If this was the case I wonder whether you would expect to see a correlation between extinction and environment. After all, it might take millenia for the effects to really be felt. Could this influence these results?

      Abel Gone · 20th August 2014 at 4:23 am

      I have been impressed by the extinction of the Moa in New Zeeland. Human landed in ~ 1400 and Moa, the largest bird on land during our era, were gone by ~1600. They had probably no knowledge of the danger of humans. It did not take long to eat all that. Then the Maori ended up being cannibals when found by Abel Tasman and then James Cook. This is not long ago: human can definitely wipe out a species with spears.

G "Fats" Montgomery · 7th August 2014 at 8:49 pm

I just feel like when we frame the question with two possible answers one will always be the answer even if it isn’t the correct one. Posing the question as such places a really broad brush on this picture.

Considering that 99.9% of all species have gone extinct and that the vast majority went before man ever set foot on Earth I think we need to allow for some complexity before we draw a straight line from Mammoths to Man.

First, my understanding is that this study only considers megafauna extinction without considering lesser fauna or flora. So if, for example, climate change caused a particular grass to die off then the animals that depended on that grass die off as do the animals that depend on those lesser fauna. Perhaps its bye bye for Dyer Wolves or Sabre Tooths. Further Tigers for example are associated with decreased populations of wolves, not because they eat them but because they out compete them. Assuming human technology, population, and taste remain constant humans were no more a threat than they were before climate change.

It is probably safe to assume that human populations suffered in some locations due to climate change but probably prospered in others on the same account. The atlatl and the bow and arrow were probably the two biggest innovations in UPaleo times and possibly nets but would this be enough of an innovation to decimate a populations of megafauna? Even religion plays a role in this as most all religions proscribe dietary restrictions. Do their mores require certain types of sacrifices and not others?

On top of all that we need to ask if there are genomic factors that allow one species to evolve and adapt more rapidly than others? How did Siberian Tigers adapt to modern day Siberia but other cats did not? It doesn’t seem clear that humans would have preferential hunting for other cats over Tigers.

Thank you I really enjoy your blog.

    Adam Benton · 7th August 2014 at 11:17 pm

    It’s not as though only two factors are being considered; it’s more that after decades of research these are the two which appear the most plausible, chronologically speaking. Of course, we should be careful not to limit ourselves but there’s still a fair few reasons why these are the factors worthy of further examination.

    Also, this latest study expanded out what they were examining to “large mammals” (i.e. mammals over 10kg) rather than just megafauna (mammals over 100kg). They didn’t examine vegetation though; and that would certainly be something to consider.

Les · 7th January 2018 at 6:31 am

Evidence shows a series of meteor impacts around the globe in 2 cataclysmic events lasting up to 9 years each. The events were apparently yearly, coinciding with the end of fall. Nano diamonds, only caused by impacts, spread throughout the continents wide ‘black layer’ corresponding with the extinction. The ‘black layer may not be evidence of human burning but evidence of continent wide firestorms. Nano diamonds in the Greenland ice sheets in the corresponding layer. ‘Kill sites’ attributed to human predation that non-archaeologists judge to be places where animals were trapped by forest fire or carcass depositories of floods. At this same period, the human population bottomed out, rebounded, bottomed out, and rebounded world wide. Humans inhabited North America before this time, went nearly extinct at this time and received an influx of rapidly expanding population after. Archaeological evidence in Australia can be interpreted similarly.

    Adam Benton · 18th February 2018 at 12:24 pm

    I’ve heard of this idea a few times but never quite been convinced of the scale of the impact it had. After all, many kill sites show signs of butchery; an indication humans, not meteors, were involved. That said I won’t pretend to have thoroughly examined it, so any sources you can cite would be appreciated.

Leave your filthy monkey comments here.

This site uses Akismet to reduce spam. Learn how your comment data is processed.