Are Fences The Only Way to Save Africa’s Lions?
Posted by Richard Conniff on April 3, 2014
Is building fences the best way to protect wildlife from people, and people from wildlife? For a lot of wildlife enthusiasts, the question conjures up memories of zebra and wildebeest carcasses piled high in the 1980s, when cattle fences cut off ancient migration routes in Botswana’s Kalahari Desert. But this time, some biologists think that fences might just be the only way to save Africa’s rapidly disappearing lions, which have lost half their population just since 2000.
And while Africa is the focus of the argument, the debate has extended to the idea of building fences to separate wildlife from people and livestock even around habitats as remote and sacrosanct as Yellowstone National Park.
The argument this time got started in March 2013 when lion biologist Craig Packer and more than 50 coauthors published an article in the journal Ecology Letters noting that Africa’s lions have already lost 75 percent of their original habitat. It predicted that almost half of the remaining unfenced lion populations “may decline to near extinction over the next 20-40 years.” On the other hand, “every fenced population is expected to remain close to its carrying capacity for the next century,” largely because fences reduce habitat loss, poaching, illegal grazing, and other problems caused by the rapidly increasing human population. According to Packer and his coauthors, fences also make it possible to conserve lions for an annual cost of just $500 a square kilometer, versus $2,000 for unfenced lions.
That article provoked a response in the same journal from predator biologist Scott Creel and an equally large team of co-authors, pointing out that 10 of the 17 fenced reserves in Packer’s study had a population of five or fewer lions and yet were counted as success stories. Meanwhile, Tanzania’s Selous Game Reserve, with 640 lions, “would be considered a failure.” The fenced reserves were also less than a tenth the size of the unfenced ones, and in unfenced reserves, each management dollar conserved many more lions.
A new article appearing today in the leading journal Science takes the fence argument to a broader audience. Sarah Durant from the Zoological Society of London and two co-authors argue that fenced reserves inevitably cause dramatic changes in vegetation and marked declines in wild herbivores. The antelopes are hammered from the top-down by lions and other predators, and from the bottom-up, because they can no longer find enough vegetation to eat. Fences also frequently fail to deliver the promised results. In South India, for instance, a study of 37 fences found that elephants broke through almost half of them.
When I spoke to Packer at his research station in Tanzania’s Serengeti National Park, he argued that it might ultimately be necessary to fence in even the Selous, a Switzerland-sized area, which has been the source of most of the 1,000 lion attacks in Tanzania on humans over a recent 15-year period. But in other cases, it might be more practical to fence in villages within protected areas, as in Mozambique’s Niassa National Reserve. Or it might pay to fence in certain remote pastures to reduce conflicts with wolves outside Yellowstone National Park.
Packer also had one fence more immediately in mind. A largely agricultural population of 600,000 people now farms and grazes along the western border of the Serengeti, he said, and that population is doubling every 20 years. In a paper to be published soon, one of his graduate students has found that bushmeat hunting there consumes 100,000 wildebeests a year. About 250,000 head of cattle also graze illegally in the park.
Sarah Durant, a long-time cheetah researcher in the Serengeti, acknowledged the problem on the western border there. But while a fence might seem like the simplest solution, she said, “to me, it’s a bit of a regression to ‘fortress conservation,’ where you had protected areas and kept people out.” The more recent trend in conservation has been to give people a stake in protected areas and develop mutually beneficial ways for local humans to co-exist with wildlife.
A fence on the western side of the Serengeti would ultimately need to run more than 300 kilometers, at a cost of $3,000 per kilometer. Apart from the question of who pays that bill, local communities would almost certainly oppose it, not least, said Durant, because they have come to depend on the park for bushmeat. Nor would a fence necessarily change things all that much. At other parks, people have simply cut through the fences—and used the wire to make new snares.
If money were spent instead on developing ecotourism and other community benefits, said Durant, “you might end up with a community that’s more tolerant of wildlife and more engaged” with the park. She also warned that once protected areas choose to fence out neighboring communities, “there’s no going back.” People lose their coping skills for living with wildlife and their connection to the landscape.
It’s a measure of just how complex and divisive the fence issue can be that wildlife officials from Kenya and Tanzania appear on opposite sides of the debate. Also lining up against one another are biologists working on the same species, in the same habitat, and even for the same for the same organizations, including the Wildlife Conservation Society and Panthera.
“We all acknowledge that there’s a line where fencing may be the last resort,” said Durant. “But we differ in where we draw that line.” The Science paper says fences can sometimes be valuable in “last-ditch attempts to preserve wildlife areas already isolated by human development.” Fences have served successfully in Kenya, for instance, to separate critically endangered hirola antelope from predators, and in Australia to protect native marsupials from invasive species. But fences should remain just that, said Durant—a last resort.
Packer countered that the population of Africa is on track to quadruple to four billion people over the remainder of this century. That could just mean that the last resort is already here.