ConservationBytes.com

A few months ago I asked Chris Johnson of the University of Tasmania to put together a post on his recent Science paper regarding Australian megafaunal extinctions. It seems that it stirred, yet again, some controversy among those who refuse to accept (mainly archaeologists) that humans could have had anything to do with pre-European extinctions. Indeed, how could humans possibly have anything to do with extinctions?!

—

Like Corey, I am mainly interested in current environmental problems. But now and then I wade into the debate over the extinction of Australia’s Pleistocene megafauna [editor’s note: Chris literally wrote the book on Australian mammal extinctions over the last 50,000 years], those huge animals that wandered over the Australian landscape until about 40,000 years ago.

This is an endlessly fascinating topic. The creatures were wonderful and bizarre – it’s great fun doing work that lets you think about marsupial lions, giant kangaroos, geese bigger than emus, echidnas the size of wombats, and the rest. The cause of their extinction is perhaps the biggest mystery, and the most vexed controversy, in the environmental history of Australia. And for reasons that I will explain in a minute, solving this mystery is profoundly important for our understanding of contemporary Australian ecology.

The latest bit of work on this is a paper that a group of us (including Corey’s close colleague, Barry Brook) published in Science. You can see it here (if you don’t have access to Science, email me for a copy). So far, research on this problem has concentrated on dating fossils to find out when megafauna species went extinct. Several recent studies have found evidence for extinction between 40,000 and 50,000 years ago, which is about when people first came to Australia. But the conclusion that people caused a mass extinction of megafauna has been strenuously criticised, because so far it is based on only a few species with good collections of dates. The critics argue that other species disappeared before humans arrived, maybe in an extended series of extinctions caused by something else, like a deteriorating climate.

This argument over fossils will be with us for a long time. Because finding and dating fossils is such hard, slow work, the fossil record will inevitably give a seriously incomplete picture of what happened. One way around this problem would be to analyse the fossil record using mathematical approaches that take into account the problem of incomplete sampling. Corey is lead author of a recent paper that introduced a great new set of tools for this, and we are part of a group that is currently assembling a complete database of all recent dates on Australian fossils so that we can analyse them using these tools. Stay tuned for the result.In our recent study we took a different tack, using an ecological proxy that provides a continuous record of the abundance of big herbivores. We looked at Sporormiella, a fungus that produces spores in the dung of herbivores. Big herbivores produce lots of dung and therefore lots of spores, making Sporormiella a neat proxy for the relative abundance of megafauna. We counted spores in swamp sediments at Lynch’s Crater in northeast Queensland, sampling the last 130,000 years of environmental change at the site. Our results show that megafaunal abundance was stable, despite dramatic shifts in climate, until it crashed about 41,000 years ago, which is about when people appeared in the area. We analysed vegetation as well, and found no change in vegetation leading up to, or coinciding with, the megafauna crash. This makes it clear that climate change was not involved, because if climate had caused the extinction it would have transformed the vegetation too.

And this is where it gets really interesting. The vegetation did change at Lynch’s Crater, and fire increased as well, but not until just after the megafauna declined. These events were well known from previous study of the site, but had always been attributed to landscape burning by people. Our results suggest that they are best explained by the removal of megafauna. This did not surprise me, because I have always suspected that Australia’s megafauna had a large impact on vegetation. This is shown by the structure of the plants themselves. Many Australian Acacia, for example, have spines and densely tangled branches that are classic adaptations for defence of leaves against large browsing mammals. Some trees have preposterous growth strategies whereby they maintain this form until they reach a ghost browse line at about three metres above ground, and then adopt a different, non-defensive, leaf and branch structure.

The existence of these traits suggests that browsing by large mammals drove the evolution of Australian plants. For that to be true, big animals must have had strong effects on the fitness of individual plants. If mega-herbivory did that, it must also have shaped the structure of vegetation communities. The sudden removal of big animals by some external factor (like the appearance of people) would therefore cause a major ecosystem shift. That, I argue, is what we’ve described at Lynch’s Crater.

The reason this should matter to ecologists now is that we need to come to terms with the fact that large herbivores were once a major control on Australian environments, but the continent has recently been transformed into a land without very large animals. That loss provides part of the explanation for why Australia environments are they way they are, and it tells us there is no reason to think that environments like Acacia scrublands, that evolved with big herbivores but are now bereft of them, are in a natural or equilibrial state. One unsettling implication of this knowledge is the idea that, if important interactions between Australian plants and animals were lost with megafaunal extinction, we might be justified in introducing alien species to reinstate those interactions.

This would be a radical and almost heretical proposition if it had not already happened. Europeans have introduced many large mammalian herbivores that have become well-established as wild species in Australia. Some of them seem to make a poor fit with Australian environments, but in other cases that is not quite so clear. For example, one large chunk of the megafauna was made up of large, dry-country kangaroos that browsed on the tough leaves of shrubs and small trees. That ecological role disappeared when those kangaroos went extinct, but it may have been partly re-taken by goats and camels.

While there is no doubt that both species do environmental harm when they are over-abundant, they are also capable of providing environmental benefits, for example by controlling woody weeds. At the Ecological Society of Australia conference in Hobart in December 2011, the central Australian botanist Peter Latz gave a talk arguing that Acacia woodlands were in a healthier condition when browsed by camels than when not. But at present, Australian ecologists and conservation managers see goats and camels only as destructive pests, and would eradicate them if they could. That can’t be done, so the goal of management is usually to reduce their population densities as far as possible.

I suggest that we might manage Australian environments better if we took the long history of large herbivores in Australia to heart [editor’s note: a proposition analogous to our ideas about letting dingos do the feral management for us], and re-evaluated the ecological potential of invasive large herbivores. The key to this new thinking would be to estimate, for species like goats and camels, the population densities at which they bring more environmental benefit than harm, then aim to manage populations to hold them close to those densities. I suspect this would be a for more achievable proposition than current pest control operations, because it might often be that the densities at which those species provide ecological benefit are not far below the high densities that result in clear environmental damage. It’s a fine balance and will require a lot of work to ascertain, but it’s probably better than the ad hoc way we manage these species now.

Chris Johnson