New Duke-led study presents new way of tracking giant pandas

Although giant pandas are perceived as fluffy and docile, they can be elusive—something that has made conservation efforts difficult. 

However, a newly published Duke-led study suggests a solution: footprints. 

The Chinese government has been successful in giant panda captive breeding programs, but has struggled to find success in reintroduction efforts because of a lack of a cost-effective and reliable way to track the furry creatures once they are released to the wild. The new technology, Footprint Identification Technique, needs just a camera-equipped smartphone with a program from JMP Software and a ruler to correctly identify a panda and its sex more than 90 percent of the time. 

China’s researchers previously relied on testing fecal DNA, which was costly, or measuring bamboo stalks that pandas had munched on, a crude and rudimentary method that was inaccurate, explained Sky Alibhai, principal research associate at JMP Software and adjunct faculty member at Duke’s Nicholas School of the Environment. The Chinese government was also opposed to the traditional method of collaring pandas to track them. 

But the Footprint Identification Technique's low-cost, accuracy and ease of use give researchers a leg up. 

“You have probably five orders of magnitude more probability to get footprints than you have to find the animal itself,” said Zoe Jewell, also a principal research associate at JMP Software and an adjunct faculty member at the Nicholas School. 

Jewell noted that species such as the cheetah can take up to 500,000 steps per day. However, plodding pandas don’t take quite as many. 

“Endangered species are elusive, but they leave footprints everywhere they go,” she said. “If all of us carried our smartphone and every time we saw a footprint we took a picture and submitted it, we would soon...have the data we need to protect them.”

Alibhai said that FIT takes into account many features of the footprint, from the tips of the toes and the pads of the foot. It takes measures like distances, angles and areas in order to help the researchers discriminate between individual pandas, he explained.

By taking all these variables into account, the research team hopes that it can help make a difference in protecting the giant panda. 

“Conservation biology desperately needs reliable data on the distribution of endangered species,” Jewell said. “If you don’t have data on where your animals are and how many there are, you can’t begin to protect them.”

The idea for the research began with Jewell and Alibhai’s tracking research for black rhinos in Africa. The duo collaborated with indigenous trackers, who said that collars were unnecessary and invasive. All the information they needed was in a rhino’s footprints, they said. 

“That was a revelation to us,” Jewell said. “We always thought high technology and collaring were the answer.”

Binbin Li, assistant professor of environmental sciences at Duke Kunshan University and eventual leader of the research, suggested that they apply that same technique to giant pandas.

The research team now hopes that the new technology can also allow regular people—not just researchers—to become a part of the process to protect the vulnerable species.

“We want people to stop thinking that wildlife monitoring is the preserved job of professionals,” Jewell said. “We want everybody to engage in it. And we potentially have the ability to have everyone who has a smartphone going out there and taking pictures and make this a global effort for conservation.” 

---