On introducing elephants to Australia...

Sometimes scientists suggest the most absurd things. In the news last week (with thanks to an Australian friend for tipping me off) was a paper published in the prestigious journal Nature that suggested in the text and headline that Australia should introduce elephants to control an invasive grass that originally came from Africa: Gamba grass, Andropogon guyanus. The author made a number of sound observations: Australia (like too much of the world) is riddled with invasive species, has suffered a massive extinction of it's native mammal population and has had some pretty nasty wildfires in the last few years. But how you get from those observations to suggesting elephants (and even rhinos) should be introduced to the savannas of Australia is a story worth looking into.

Invasive species can be extremely damaging to biodiversity, farmers and even human health. Here in Africa we have our own set of problem species, highest priority in East Africa at the moment is surely Parthenium hysterophorus, a weed native to the American tropics that is currently invading savannah habitats across East Africa, after introduction for the cut flower trade. In Ethiopia, where the species is already established, the grazing potential of rangelands has already been reduced by 30% as the weed is toxic to most mammals. It's also causing unknown damage to human health, as many (most?) people coming into regular contact with the plant develop allergic reactions to the pollen and sap. In South Africa it is estimated that invasive weeks cost the economy around 6.5billion Rand ($800Million) per year. So invasive weeds can certainly be a major problem. Most scientists recognize that these plants, which may be perfectly innocuous in their own environment, become serious pests in the areas they're are introduced, because they are freed from their natural predators that help keep the populations in check in the native range. One of the best solutions, therefore, is the introduction of a suitable plant predator, to keep the population under control. Some great examples of this exist: the Prickly Pear (Opuntia) cactus, for example, is native in the Americas but a weed across Africa and Australia. A moth was identified that is a specific herbivore of the prickly pear and has been introduced in Australia and parts of Africa with often dramatic success. After introduction to Australia, the moth (Cactoblastis - the perfect name for an effective cactus eradication agent!) rapidly infested many of the cactus plants with the effect that today the infestation has been eliminated. Prickly pear is no-longer considered a serious threat in Australia.

White rhino seem to shun gamba grass...

Biological control, however, is not a straightforward option – to deliberately introduce one species to control the population of another species is a rather risky option. You don't want to end up like the old woman who swallowed a fly, then a spider to catch the fly, then a bird, then a cat, etc., etc. Many examples of apparently good ideas have gone horribly wrong, a favorite example of mine being the introduction of predatory snailsto eat the introduced giant land snail on Tahiti where, once introduced, the predatory snail decided the native snail populations were far tastier than than invasive species, and rapidly drove the native species to extinction in the wild (happily a few were saved and are bred in captivity to be, hopefully, reintroduced once the predator has been dealt with. Somehow...). Recognising this, conservation organisations got together to come up with a list of guidelines for the introduction of species under the umbrella ofIUCN, the World Conservation Union. These guys propose a number of questions and guidelines that governments should consider before making introductions: “What is the probability that the species to be introduced will threaten

the continued existence or stability of populations of native species, whether as a predator, competitor tor food, cover, breeding sites or in any other way? If the introduced species is a carnivore, parasite or specialised herbivore, it should not be introduced if its food includes rare native species that could be adversely affected.” Etc. (Note the implication that you'd only be considering introducing a specialised herbivore!)

They note that “No introduction should be made for which a control does not exist or is not possible. A risk-and-threat analysis should be undertaken including investigation of the availability of methods for the control of the introduction should it expand in a way not predicted or have unpredicted undesirable effects, and the methods of control should be socially acceptable, efficient, should not damage vegetation and fauna, man, his domestic animals or cultivars.”

In this paper David Bowman suggests that the ideal control method for gamba grass is the introduction of elephants. (He further suggests that introduced grasses such as this species, by leading to a build-up of fuel, are responsible for the fires that have killed so many in Australia. This notion is so absurd I don't think it's worth going into: the grasses in question are in the savanna zone of northern Australia, the worst fires in the south. As with all savannas globally, there are and always have been fires in the savanna belt – to remove them would be to cause untild damage to these savanna habitats, etc., etc.) He suggests that other mammals in Australia, such as Asian water buffalo Bubalus bubalis, cattle and the rest are too small to eat the exotic grass, but states that “gamba grass is a great meal for elephants or rhinoceroses”. So, how would this proposal fit the IUCN guidelines? Well, the first thing to do would be to find out if elephants and rhinos really do eat gamba grass. It took me about 10 minutes online to discover thatelephants do eat gamba grass, but they certainly don't select it overother species but rather eat it in proportion to it's abundance during the wet season (and not at all during the dry season). Rhinos I could find less details of: black rhinos are browsers, so we can ignore them for now and look at white rhinos. Contrary to Bowman's assertion, I certainly couldn't find any evidence suggesting gamba grass is a great meal for them, but I did find evidence that they don't like it, with areas covered in gamba grass consideredunsuitable habitat for white rhino reintroductions.

What's more, as anyone who's studied elephant diets would have been able to tell the author, elephants have a mixed feeding strategy – being predominantly grazers in thewet season, and predominantly browsers in the dry season. Even assuming they could be persuaded to eat gamba grass in the Australian wet season, what will they eat in the dry season? If they find something they like, their impacts can be very serious indeed (and if they don't what hope can their introduction have of success?). What's more, if they start having unintended consequences and people wanted to wind the introduction back, I can't see a cost effective method of control ever fitting the “socially acceptable” criterion. Live trapping would be incredibly expensive, and even if possible, where could you put several thousand elephants to live out their lives?!

So, it seems to me that introduction of elephants to Australia could never reduce the fuel loads in southern Australia (particularly not around habitation, where fires are most dangerous). Nor is it likely the animals would even do the job they're being promoted for. There's no way the introductions could ever come close to meeting the IUCN guidelines. And it would be a totally crazy thing to do, as anyone with a bit of knowledge of elephants and rhinos could tell you. So why did this get written? And then why did it get published? Sometimes I wonder about my colleagues...

Main reference:

Bowman, D. (2012). Conservation: Bring elephants to Australia? Nature, 482 (7383), 30-30 DOI: 10.1038/482030a