By Jason Koebler
Hop off a plane in Johannesburg, head six hours east to the Mozambique border and enter the Balule Private Game Reserve, located within South Africa’s Kruger National Park. Once you arrive, computer scientist Tom Snitch says he can show you an endangered rhinoceros within 15 minutes.
Snitch, a researcher at the University of Maryland, is new to the conservation game, but he’s developed an algorithm that could revolutionize how endangered animals—and poachers—are monitored.
“I don’t have to find a poacher, but I have to find a rhino,” Snitch says. “I can take you today to Balule and I can show you to within 200 meters of where every rhino is. We’ve got to put the rangers between the poachers and the rhinos.”
To do that, Snitch has used drone monitoring and park data to create a sophisticated algorithm that predicts rhinoceros movement. The algorithm takes things such as the availability of prickly pear cactus that Snitch says rhinos “treat like candy,” wind patterns, phases of the moon, precipitation, past poaching data, and recent ground intelligence data to devise flight plans for two $21,000 Falcon drones. The flight plans are also used to help position rangers so that they will be more likely to intercept poachers before they’re able to kill the rhinos. There’s little advantage, he says, of using unmanned aerial vehicles to take videos of rhinos getting killed if you can’t actually make an arrest or prevent the poaching.
“Africa,” he says, “Is too big to randomly launch UAVs.”
Last year, Snitch, with funding from the Endangered Wildlife Trust, flew drones in Olifants Nature Reserve in the western part of Kruger National Park. During the 11 days he flew, there were no poachings and there were several “apprehensions of individuals who were where they shouldn’t have been.” It’s hard to say whether those people were would-be poachers, but Snitch says the suspects “were too close to a fence where they had line-of-sight of a rhino at night.” In the week after Snitch left, there were nine rhinos killed.
“I’m not going to attribute it to causality, that just happens to be the number,” Snitch says.
Using drones to help control poaching is a fairly new idea, but it’s one that has caught on quickly. In 2012, Google gave the World Wildlife Fund a $5 million grant to expand the organization’s drone program. The South African government has also used military-grade drones to help surveil poachers in the country, groups which have become highly sophisticated crime syndicates with automatic weapons and night vision goggles.
That sophistication has led to an unprecedented increase in poachings in South Africa. In 2007, 13 rhinos were killed in the country. Every year since then, a new record has been set. Last year, the number topped four digits, with 1,002 dying, many of them in Kruger.
Drones offer hope, but so far, conservationists and park rangers across Africa aren’t sure what kind of impact they’ll have. Even low-end drones can be too expensive for many parks to afford, and most don’t believe that drones will be the cure-all for poaching that some have hoped for.
“Over and over we’ve seen that more ranger booths on the ground is the most important deterrent to poaching,” Eric Dinerstein, vice president of the World Wildlife Fund’s conservation science program says. “You can have UAVs flying, video cameras going, but if you don’t have anyone on the ground to intercept the poachers, it won’t matter.”
Snitch is more optimistic than most that smart drone use can be a huge poaching deterrent, if not the final answer. He’s grown frustrated with big drone companies offering million-dollar drones to small parks that could never afford it and conservationists who want to try setting up stationary camera traps, which he says can be easily stolen, to spy on poachers.
“I think people have got to start believing that simple solutions are better than more sophisticated ones. Kruger is the size of Israel and if you’re going from guys with boots and guns to suddenly having satellites and UAVs, that’s a lot to digest and comprehend,” he says. “It’s tough to get people to think in ‘Africa mode.’ There’s not a lot of money out there to be doing this.”
That’s why he says knowing when and where to fly smaller drones is a much better idea than purchasing more expensive ones with longer flight times. Dinerstein isn’t convinced that what he’s doing is all that revolutionary.
In September, poachers poisoned a watering hole with cyanide in Zimbabwe, killing 80 elephants and hundreds of other animals who fed on the elephant carcasses and also drank from the water. Poachers, Dinerstein says, “have their own algorithms” and are getting more sophisticated. They don’t necessarily have to be near an animal to eventually kill it.
“I think his point is important but it’s a very obvious one,” Dinerstein says. “It doesn’t take an algorithm to know that watering holes are going to be the hot spots, particularly in arid environments.”
Dinerstein and Snitch are at odds with how to best proceed: Snitch says that tracking animals as they move is the most effective, Dinerstein says that in Namibia, where WWF has just started flying their Google-purchased drones, watering hole security is more important. Snitch says it’s important to start in Kruger because it has more rhinos than anywhere else on Earth, and loses hundreds of rhinos to poaching each year. Namibia, on the other hand, loses just a handful.
“You lose five a year in Namibia. You lose two a day in Kruger,” Snitch says. “From a PR point of view, if you cut it poaching by 40 percent in Kruger, you still have hundreds being killed. People like raising funds, and it’s easier to raise funds if you can say you’ve reduced poachings to zero in a year.”
Dinerstein does agree that drone use will eventually need to be expanded to other countries where poaching is more of a problem, but says that WWF has a good relationship with the Namibian government and that they “wanted to start in a place where poaching is pretty well under control and a place where we’ve had good cooperation.”
“Right now, this is just a proof of concept,” Dinerstein says. “Then we’ll move on to other sections.”
Snitch, however, is already pressing forward.
Snitch’s work in Olifants Nature Reserve impressed government officials at Kruger, and he and his team plan to start doing test flights there in March. They’ve gotten government poaching data—a huge step in developing a new algorithm for the area on the South Africa-Mozambique border, where much of the poaching occurs. He says by March, he hopes to create an “aerial curtain” along the border that can be run for half a million dollars per year.
“That’s the price of one rhino horn,” he says. “I think we can cut poaching by about 50-60 percent. That’d be a huge measure of success. But the poachers can just go somewhere else. We’ve got to have a system-wide application in place, otherwise you’re just playing whack-a-mole.”