Brett Smith for redOrbit.com – Your Universe Online
Researchers from Cornell University have been equipping pastoral herders with smartphones, and the herders are using the devices for much greater purposes than updating their Facebook status or perusing Tinder.
Herders in the Cornell program are using an app to guide their herds to the best possible grazing areas based on crowd-sourced data and GPS technology.
“If we can use sophisticated technology to run Amazon we can use sophisticated technology to help people in Africa,” noted lead investigator Carla Gomes, a professor of computer science at Cornell. “We can provide the pastoralists with information on which areas are good for their cattle.”
Pastoralists in Kenya, where the program is based, depend on the health of their herd, and the economy country itself is made of 60 percent livestock products. To keep the herds healthy, these nomads must shuffle them from grazing area to grazing area.
“Where the soils and rainfall are inadequate to grow crops, the only viable livelihood is raising livestock,” explained Christopher Barrett, professor and chair of Cornell’s Dyson School of Applied Economics and Management.
In order to boost their chances for survival and economic success, herders need to know several important pieces of information, such as the locations of good grazing areas, the going price for cattle, and the movements of warmongering groups.
A new era for our herding friends
Satellite images can differentiate vegetation from arid wasteland, but can’t determine what types of plants are present. Herders need to know this information because goats and camels can browse on trees and bushes, but sheep and cattle need grass.
In the new Cornell system, herders can upload mobile phone reports about an area’s vegetation, which are then labeled with GPS coordinates. A computer takes this information is associates it with the spectral “fingerprints” on the satellite images. Ultimately, the computer will be able to determine several kinds of vegetation from the satellite image and that data will be provided back to the pastoralists via their phones. The researchers also plan to use drones to collect data on the ground.
The team is assessing the system with a small number of pastoralists in the town of Isolio in Kenya, who are attracted to the program by the prospect of upgrading to a smartphone. The herders are paid for participating in cell phone minutes, a typical currency in their culture.
“The app is designed so they don’t have to speak the language: it’s all pictures,” said Peter Frazier, assistant professor of operations research and information engineering at Cornell.
Users are able to log information into the system by tapping various icons created to show the different kinds and degrees of vegetation. As a way to confirm their reports, users are prompted to add a photo of the vegetation.
In order to gather information from as much area as possible, researchers incentivize travel to low-information areas by upping the payments for herders who report from these locations. The researchers also track the herds using GPS collars.
“Data from near their home is not that interesting,” said team member Yexiang Xue, a graduate student who worked on the incentive system. “If you set up one place with a very high payment, you lose the benefits of people going to other places.”
If this initial stage produces good results, the Cornell team expects to expend the program to other areas.