Correction: An earlier version of this story incorrectly stated that IBM’s Watson would be used to analyse the data. IBM’s Internet of Things systems will assist the Wageningen University in analyzing the data collected.
The internet of things could become the Internet of animals too, as data is harnessed to save Africa’s wildlife.
A combination of IBM’s Internet of Things, MTN’s 3G and 4G networks and a herd of zebras could be the latest defense against rhino poaching. On Sept. 19 at the Welgevonden game reserve in South Africa, IBM and MTN began a new data analysis program that they hope will prevent rhino poaching.
This isn’t the first time technological innovation has been recruited in the fight against poaching. Drones have been used to guard rhinos from above, while rangers have been kitted out with thermal imaging and infrared goggles. In some cases, the rhinos themselves have been fitted with microchips to monitor their movement.
This time, however, other members of the animal kingdom will be helping out. Zebra and buck species like eland, wildebeest and impala will be fitted with collars to track their movements and responses to danger. Their activity will be measured against data patterns and research collected by the Wageningen University in the Netherlands, which shows that animals move differently depending on the threat they’re facing. For example, they’re likely to move as a herd if a lion approached, but scatter chaotically if a hunter with a rifle is spotted. Their response time is also usually much faster than a one-ton rhino’s.
“Because they’re much more skittish, they’re movement is monitored,” said Craig Holmes, vice president for cognitive industry solutions in IBM Middle East and Africa.
The smaller animals act as sentinels, with their response patterns acting an early warning system. Using the mobile data network, the collars themselves emit a signal four times lower than the average cellphone and so should, in theory, not disturb the animals.
The collars make use of LORA technology, long range low power signals to send the data to IBM’s Internet of Things where it will be managed, to help the university analyzed and model the data to inform rangers of whether the rhinos grazing near the buck are in danger, says Holmes.
“For the data to be collected and analyzed you need to respond within 400 milliseconds or less,” explained Petrus Greyvenstein, IBM’s executive architect. The data they’ll be checking for include GPS coordinates, ambient temperature, acceleration and speed. So far 135 buck and zebra have been fitted with the collars, and the aim is to collar more animals to expand the dataset and machine-learning capabilities.
Rhino populations around the continent have dwindled as poachers hunt them for their horns. Across the continent there are about 5,000 critically endangered black rhinos left and approximately less than 30,000 white rhinos, putting the animal at the “near threatened” stage. South Africa may be home to most rhinos, but they’re still not safe with poaching increasing from just 13 rhinos killed in 2007, to 1,215 in 2014, according to the World Wildlife Fund.