Panda Eats Shoots, Leaves—and Meat?!?

Pandas are vegetarians, right? Well, new findings by Conservancy scientists suggest the issue isn’t as black and white (or, er, as green and blood red) as once thought.

Motion sensor cameras were set up this summer in the soon-to-be established Motianling County Land Trust Reserve in northern Sichuan by The Nature Conservancy, Peking University and local government partners. In November they captured images of a giant panda consuming the carcass of a takin, a Himalayan goat-antelope. These photos provide visual confirmation that pandas at least occasionally eat meat in addition to their customary staple of bamboo leaves. (See the amazing images captured by remote camera.)

While this isn’t news to scientists — evidence in feces has shown that pandas do sometimes eat meat — very few photos exist of a panda actually consuming it.

But the panda’s no killer; scientists confirmed that the takin had died of natural causes several days before it was discovered by the panda. “These images show that there is still so much we don’t know about their behavior,” says Zhao Peng, the Motianling project lead for the Conservancy. “They really are an incredible species.”

But the question remains: Is the panda portrayed in Kung Fu Panda closer to real life than the cuddly ball of fur we all love and adore? To get to the bottom of this question we reached out to Matt Durnin, the Conservancy’s Asia-Pacific Conservation Science Director. Matt has been studying pandas for more than a decade and conducted his Ph.D. research on the wild giant pandas of the Wolong Nature Reserve in Sichuan province.


Q: Where did the researchers find this panda, and what were they investigating with their camera traps?

Matt Durnin: The work is being conducted in the Motianling Land Trust Reserve a 110km2 area in Pingwu County, Sichuan Province. The reserve is a vital, healthy habitat for conservation target species including the endangered giant panda, and is adjacent to two existing giant panda reserves – Baishuijiang and Tangjiahe National Nature Reserves.

We use remote cameras as a “non-invasive” way to monitor species presence in the reserve. Remote cameras are a now widely used and well proven methodology to gather information not only on the presence of species but — as is evidenced in these photos — on behavior (e.g. feeding or scent marking).

In previous work done on giant pandas in the wild, I was able to capture photos of pandas at “scent trees,” sniffing the scent of other pandas as well as leaving their own scent on trees. Remote cameras have also been used to identify the presence of previously unknown species or species that have been considered extirpated from an area.

Q: Was this news a surprise to you?

Matt Durnin: This news is not so much a surprise because researchers have previously found the remains of animals in panda feces. But it is exciting and significant. I’ve only ever heard of one other incident of a panda being photographed on a carcass and those photos have never been published and were taken with a mobile phone.

What makes these photos significant is the number and quality of them, as well as the systematic way in which they were obtained. Researchers came across the carcass (so were able to estimate how long since its death) and placed the camera there to photograph any animals that might come along and feed on it. I don’t think anyone expected that it would be a panda but rather some other carnivore. This is a panda feeding on a carcass over a 6-hour period; it’s the most extensive photographic footage of a panda in the wild doing so.

Q: What can you learn from these photos? How common is meat-eating in pandas, and is something new happening here that might be increasing the behavior?

Matt Durnin: Pandas are technically carnivores. And we know from finding feces in the field that contained animal remains — as well as anecdotally from conversations with locals living and spending time in panda habitat — that they do eat meat from time to time. So this is not a “new” behavior — but it is, we believe, very uncommon. So documenting it with such a large number of high quality photos is an important result of this research.

From the photographs we have what appears to be a very healthy panda (it’s not possible to say if it’s a male or female) feeding for approximately 6 hours on the remains of a takin. There is plenty of its primary food, bamboo, in the area. So we can assume it was not starving from lack of access to bamboo but rather it was hungry, found a “fresh” carcass and so did what carnivores do and ate the meat.

However, there is only a small amount of data supporting that wild pandas do eat meat, so we still consider this a rare behavior. I’ve collected hundreds — perhaps thousands — of feces for DNA and bite-fragment analysis in my research over the years and have found only one feces with anything other than bamboo in it.

These 600 photos tell us that, despite decades of research on pandas in the wild, we’re still learning.

Q: When most people think of panda bears, they think of sweet, seemingly cuddly creatures. Is that popular image true to the facts, or do pandas have a not-so-cuddly side as well?

Matt Durnin: The famous field biologist George Schaller was once chased and climbed a tree to get away from a female panda that he was observing. Just like any large carnivore, they have very powerful jaws, sharp teeth and claws — if they were to get a hold of a person or other animal, they could do a lot of harm. Zookeepers and zoo guests (that have jumped into enclosures or stuck their arms through cage bars) have been injured and mauled to death by captive pandas. Whlie there’s no evidence that anyone has ever been killed by a panda in the wild, they are wild animals — they are extremely strong — and if threatened can be as lethal as any other large carnivore out there. They are also, despite most people’s image, very fast albeit in short bursts.

However, like most wild animals they do whatever they can to avoid humans and the chances of someone being mauled by a panda are infinitesimally small.

Another little known fact is that their fur is quite bristly and not at all soft as many imagine.

Q: This takin was already dead when the panda started eating it. Given the photos, do you think it’s possible a panda would kill an animal for meat if it were hungry enough?

Matt Durnin: I believe if a panda had to it could catch and kill prey, but it’s not designed for long-distance running. So pandas would need to do any “hunting” by waiting and ambushing a passing animal.

However, in the early 1980s, there was a huge bamboo die-off in a large area of Sichuan inhabited by pandas. If there was ever a time for pandas to resort to hunting prey, it would have been then. But as far as I know, there is no evidence any did so, even though in theory they are capable. So there’s no evidence that they hunt and kill their own prey.

The evidence we have shows that when they do eat meat, it’s carrion.

Q: This panda was photographed on a land trust reserve — the first organized form of private land conservation in China. The United States has certainly had a long history of private land conservation, but not so in China. What can these private initiatives add to China’s nature reserve system?

Matt Durnin: The nature reserve system in China, again as in many places around the world, is sorely underfunded. So tapping into the potential that “private initiatives” have is critical to successful conservation in China.

The overall goal of this project is to overcome existing barriers (lack of funding being one of the biggest barriers) to effective conservation in China by introducing the land trust model, which enables participation by all sectors of society (non-government as well as government) in protecting critical lands while also incorporating sustainable development opportunities for struggling local communities.

The land trust model is a “new” concept to not only China, but much of Asia. Research findings like this help us gain support for piloting this new model.

(Image: Panda eating meat captured by motion sensor cameras stationed on the Motianling County Land Trust Reserve in northern Sichuan. Image credit: TNC. View more photos from the remote camera here.)

If you believe in the work we’re doing, please lend a hand.


  1. This might be a preparation strain for the evolution of next kind of species within Panda’s.Understandably climate changes are leading to eating habit reconditioning of the Panda’s.Hence a necessary check has to be undertaken.

  2. I agree, I believe this has to due with environment contributors, ie (global warming)

  3. I was very interested in this article about a panda eating meat. I was very disappointed when now of the highlighted words didn’t connect to the photos as suggested in the article?? Where may I find them?

  4. An interesting discussion is definitely worth comment.
    I do believe that you need to publish more on this subject, it
    might not be a taboo matter but usually people don’t speak about such issues. To the next! Kind regards!!

  5. This is when I lose all faith in the scientific community. You observe almost all pandas eating exclusively bamboo and yet you insist they are carnivores? One or two observed photos of pandas eating meat or examined feces with meat in it does not making them “carnivore”. Nor does their sharp teeth and short intestinal track. They are the exception, clearly not the rule. Clearly, as a whole they are herbivores.
    I wish I could find honest scientists, biologists, and zoologists.

    1. Hi Michelle, Thank you for pointing this out! It could have been explained in more depth in this post. Carnivore has a couple of different definitions. The most common one is an animal that eats (primarily) meat. However, sometimes people use the term carnivore for species that belong to the order Carnivora, like the giant panda:

Add a Comment