Last July I wrote about a Science paper of ours demonstrating that there was a climate-change signal in the overall extinction pattern of megafauna across the Northern Hemisphere between about 50,000 and 10,000 years ago. In that case, it didn’t have anything to do with ice ages (sorry, Blue Sky Studios); rather, it was abrupt warming periods that exacerbated the extinction pulse instigated by human hunting.
Contrary to some appallingly researched media reports, we never claimed that these extinctions arose only from warming, because the evidence is more than clear that humans were the dominant drivers across North America, Europe and northern Asia; we simply demonstrated that warming periods had a role to play too.
A cursory glance at the title of this post without appreciating the complexity of how extinctions happen might lead you to think that we’re all over the shop with the role of climate change. Nothing could be farther from the truth.
Instead, we report what the evidence actually says, instead of making up stories to suit our preconceptions.
So it is with great pleasure that I report our new paper just out in Nature Communications, led by my affable French postdoc, Dr Frédérik Saltré: Climate change not to blame for late Quaternary megafauna extinctions in Australia.
Of course, it was a huge collaborative effort by a crack team of ecologists, palaeontologists, geochronologists, paleo-climatologists, archaeologists and geneticists. Only by combining the efforts of this diverse and transdisciplinary team could we have hoped to achieve what we did.
One of the main reasons we put so much effort and care into this paper was that some previous, high-profile studies have confidently suggested that the extinction of some 20-odd large species in Australia’s ‘recent’ past (last few hundred thousand years) was entirely due to shifts in the climate. Although we’ve debunked the principal tenet of the most recent of those papers, a full analysis was required to demonstrate just how wrong1 they were2.
Enter our latest paper.
When reconstructing prehistoric extinction chronologies, there are some pretty basic steps you have to follow before you can go shooting off your mouth about what caused them.
- The first thing is to check the quality and reliability of the dates you have from your unearthed specimens. As it turns out, a lot of fossil material is dated very poorly, such that many dates are about as reliable as a junkie guarding your wallet. If you do not quality rate each and every date, you will most certainly include some dodgy chronologies.
- The second thing you have to do is correct for the Signor-Lipps effect — a phenomenon that arises because the youngest fossil has a nearly zero chance of heralding the species actual timing of extinction. I’ve reported before how we’ve corrected for this effect via a suite of sophisticated mathematical models.
- Finally, if you’re going to claim that climate change had something to do with the extinction chronology, you’d better have some good paleo-climate proxies, and you bloody well better correct for the sampling bias inherent in things like ice-core data (i.e., because of compression, sampling becomes less precise the deeper in time you go; but wait for a more complete demonstration of this problem within the next month or so2).
Guess what? Those that claim climate change is entirely responsible have done none of these.
But we did, and not surprisingly we found something completely different.
First, most megafauna extinctions demonstrated a clear peak around 42 thousand years ago, some 13 thousand years after humans arrived and set up shop in Australia. Second, six separate, independent climate-proxy timelines revealed absolutely no correlation to the timing of those extinctions.
Jeez-Louise, it’s amazing what you conclude when you actually do science instead of make up stories!
Now of course, these are continent-wide extinction events, so our conclusions are necessarily restricted to that spatial scale. No one, especially not us, is claiming that climate change had no role to play whatsoever over more restricted spatial scales (e.g., regions or catchments within the greater continent of Australia). However, the data we have — despite being the best that anyone has yet examined in this context — simply do not yet lend themselves to that kind of fine-scale investigation.
Suffice it to say that the evidence is pretty solid now (see also this for independent evidence) that climate change, if it had a role at all, was a minor component in the Australian megafauna extinctions. Instead, the finger is pointing fairly and squarely at human beings, which of course should surprise no one given what we know happened elsewhere in the world (both historically and prehistorically).
1A paper so egregiously bad that I would say it borders on fraud instead of just outright sloppiness and subjective bias.
2A more thorough debunking of their paper will appear soon in Proceedings of the Royal Society B.