Monday, January 25, 2016

Chimps as individuals

At Chimp & See, we ask our volunteers to accomplish two different tasks. First, we need their help classifying many hours of video material from camera traps in Africa in terms of the species and number of wild animals they contain. Secondly, if one or more chimpanzees are found in a video, we want to identify them in order to better estimate population size and answer other ecological and behavioral questions (like association patterns and cultural transmission).

While other citizen science projects like Snapshot Serengeti and WildCam Gorongosa are similar in terms of annotating all animals present, Chimp & See is one of the first projects (another one is Whales as Individuals) to attempt identifying individual animals from camera footage. In the beginning, the science team was not sure if volunteers would be able to identify individual chimps, but is happy that it works very well. We now know that with a bit of effort and guidance many can learn this skill.

Identifying (i.e., matching) chimpanzees means finding out whether a chimp in two different situations (video sequences) is the same chimp or not. The matching is considerably harder than classification itself as everyone in the project can tell. First of all, we are mostly observing chimp communities that are unhabituated to human presence. Little is known about the individual chimpanzees at each research site and so we cannot compare them to a well-lit high-resolution picture and a list of his/her features known in advance. These chimps are new to all of us – and with each site progressing, we try to get to “know” them better and recognize features that enable us to identify them as individuals. A second level of difficulty arises from the chimps themselves. The chimps hardly ever stand still in front of the camera to allow a good look at all sides of their body and face and show the specific individual traits we can rely on.

Achenar is a male adult chimp with a boxy head missing his right ear.
The traits should ideally be permanent, but temporary ones like wounds might be recognizable at least a few days or maybe even weeks. Basically every distinctive feature can be seen as a trait and support the match, starting with general body size and build, missing limbs, fur color, bald spots, number of digits (fingers and toes), and pigmentation of face, hands, and ears. The face can be narrow, round, triangular. Small or bigger scars can be visible on body and face. Ears can be bent, have cuts, or be missing altogether. A recurrent topic is the distinctiveness of the baldness pattern on the forehead that is partly already seen in infants. More behavioral traits like a special gait (how a chimp walks) or handedness are often hard to compare, as camera perspectives are rarely the same, but observing it can support a difficult match. For all these and many more features one could think of, it does not matter whether the volunteers know the correct scientific term or can provide a full list of features the chimp has. A lay description is sufficient to discuss with others and the science team. The wisdom of the crowd generates very good results.

Female chimp Esme is quite old and has a badly damaged left ear
Once an individual chimp is identified, a volunteer who has participated in the matching discussion can propose a name for this chimpanzee. Up to now, we identified and named more than 100 chimps at different research sites.

During the year, I will introduce individual chimps and show how we came up with a specific match. Following is irregular series, you can learn what to look for when comparing two chimps, how to deal with different camera angles and black-and-white footage, and how hard it is to identify infants and often even juveniles.

A first example: Timur

This old and weathered looking male from Quiet Wood (a site in Central Africa) has been identified as part of our “throwback” initiative that aimed to take a second look at the very first research sites up at Chimp & See after some months of experience with the video footage.

Timur helped us a lot with his identification as he always stopped in front of the camera and stayed there for some invaluable seconds just looking around. Even with different lighting – including black-and-white footage from evening twilight – his facial and bodily features were clearly seen: a flat head, pointy and undamaged ears, straighter brows, big eye sockets, and the flat nose with long nostrils. In some footage, the lips look lacerated and an underbite has been discussed. In addition to quite a lot of gray fur at the lower back and legs he has a unique pink patch on the right side of his butt. It is not clear whether this is a wound, scar or something else, but the footage spans over several months and it does not go away.

Full screen picture here
Full screen picture here
Back views (two months apart): same pattern of gray
and a pinkish patch on the right side of his butt.
Timur has been named (by me ) after confirmation of the proposed match by the science team. He has been seen in three different video sequences over the course of four months.
You can join the matching discussions at Chimp & See!

Monday, January 18, 2016

Field Update: Camera Traps in Gabon

Rolland and Ivonne surverying their TRS in Gabon

I'm really happy to announce we are going to try to do a monthly feature on updates from the field (as long as we have good enough internet to do it!)

Our first field update is from Ivonne Kienast from the Batéké Plateau in Gabon where she and Bo Larson are working as temporary research site (TRS) managers for the MPI-EVA Pan African Programme: The Cultured Chimpanzee's collaborative research site with the Aspinall Foundation and l'Agence Nationale Des Parcs Nationaux Du Gabon.

Ivonne writes:
You know it is going to be a specially exciting day. You can feel the tickling in your fingers, the impatience makes you walk faster, and you just want to arrive to the spots you know so well. It's camera trap day ! Meaning... that you are coming back to camp with SD cards full of amazing footage. You know you will have videos of elephants, duikers, leopards, red river hogs, hopefully pangolins, aardvaarks and golden cats. You cross your fingers for a lion video and of course and number 1 in the charts: Chimps ! You are excited to find out if the female with her fluffy baby you saw the other day in the forest will have passed by one of the cameras, or if you are going to see again on a video those 3 amazing huge males sitting in front of the camera studying it carefully. Or maybe the young male you called Tango pushing the camera and throwing a stick at it.
C&S citizen scientists you can look forward to the video this camera caught ;)
 Like every month you arrive excited to the camera. And you know something went wrong, because the camera is not there. It is lying on the ground, ripped in peaces, smashed and crashed. You could give it so many definitions ! And it is also how you feel in that moment. You look at the pieces, shaking your head, feeling frustrated. Resigned you check if you can save the SD card. You think: “Even if you loose the device, please don't loose the data.” Then you start searching for all the pieces of the plastic box, of the camera and all components which were used for the setup. In 13 months it have been 11 cameras, the last 8 ones in the last 3 months ! Elephants are reacting extremely aggressively towards the cameras. You cannot blame them. It's their forest. Your cameras scare them, or they just might have a bad day and decide to make your work a bit more difficult. As you always look at the bright side of life :P, you think: “Ok, better the cameras and not us. Or the cameras AND us”.

Well, thanks to our partner and host PPG (Projet Protection des Gorilles – The Aspinall Foundation), we found a perfect solution. PPG has been working with camera traps for gorilla monitoring for a while, and because gorillas are very touchy and curious apes, they needed to protect their cameras very well. So they designed a metallic box, which is fixed to the trees with wire, and the camera is just placed into this box. The setup is pretty stable. But PPG was having big problems with the humidity, and every month they were taking out cameras which were not working properly.
We did not have problems with the humidity, as our cameras are protected by a plastic box, covered with cling film, and having silica gel in the interior to absorbe humidity. PPG was clever ;). They decided to put their cameras in plastic boxes, covered with cling film, silica gel in the interior and all together in the metallic box. What a great idea ! So, we decided to do the same. We got 15 custom made metallic boxes and we have all our hopes on them. Unification of two different systems used at one site will allow us to get lots of amazing videos without loosing our cameras. For elephants it won't be impossible, but more difficult to destroy the new setup. We hope that out of 10 we might save 9. And we have high chances this will work ! Thanks PPG for this great idea.  

Saturday, January 9, 2016

Best of Chimp & See 2015!

Thank you to everyone who nominated and voted in our Best of 2015 event! I hope you enjoyed the chance to reminisce over our best finds, and maybe enjoy some you had missed before. But now you want to know the results, right? Without further delay.... (insert drumroll here)...

The Best of Chimp & See 2015:

Favorite chimp:

Winner:  🏆 MFInf20 🏆| Runners Up: Sugar and Abile

Funniest clip:

Winner:  🏆 Duiker log fail 🏆| Runners Up: Diana monkey playtime and Field team dancer

Best camera reaction:

Winner:  🏆 Leopard attack 🏆| Runners Up: Baboon camera adjustment and Chimp goes knock knock

Creepiest clip:

Winner:  🏆 Snake stands up 🏆 | Runners Up: Spider dangles in front of camera and Hypnotic cobra

Cutest clip:

Winner (tie): 🏆  Mom, Dad, and a porcupette and Young mangabey opens wide 🏆 | Runner Up: A kiss from Mom

Biggest surprise:

Winner: 🏆 Elephant excitement 🏆| Runners Up: Dodge's drumming display and Our first lions

And that's it! Thank you again, and now it's time to get back out there and see what amazing chimps, funny behaviors, or surprising scenes you can discover. The hunt for the Best of Chimp & See 2016 is just 11 months away! ;-)

See you over at Chimp & See!

Saturday, January 2, 2016

Savannah wildlife

Over time, Chimp & See features different research sites from West via Central to East Africa with diverse habitats and climates to study chimpanzee ecology and behavior. After several West and Central African forests habitats, the current site Dry Lake 11 is the first savannah habitat we worked on. The site and its wildlife presented some new (to us) species like warthogs, green and patas monkeys, crested porcupines, baboons, and for instance our first lions (unfortunately, only two sightings).

The landscape is very different from the dense forests we used to see at previous research sites. The space is more open with grassland and scattered groups of medium-height trees. The environment varies remarkably with distinct seasons. The site is generally drier and warmer with temperatures up to 45 °C in April (the warmest month of the year) and a rainy season peaking in August. With the cameras up at every research site for 12 to 18 months, we could follow the changing seasons very well. During the dry season, multiple species assembled at the subsequently drying out waterholes, warthogs sought relief from the heat in open “caves” and many animals used the cooler hours of the day for foraging and hunting. Wildfires often arise during this time. In July and August, heavy rainfall could be seen in many videos (sometimes very heavy rain will trigger the motion detector of the cameras without animals being present).

A guinea baboon at a waterhole late in the dry season.

A week earlier, this genet still found some water to drink.

Wildfire in a bamboo wood patch
A very welcome attraction of this research site – next to the chimpanzees – is the presence of guinea baboons. They never failed to entertain us – be it at assisting with camera adjustments  or feeding with the help of all four limbs.

Guinea baboons feeding on borassus (palm fruits), hands and feet engaged.

As baboons often travel, feed, and rest together in bigger groups, many social interactions could be observed. The greeting behavior found special interest. In addition to the more familiar hugs, touches, and vocal greetings between two individuals, (mostly) males greet each other with “diddling” (touching each other’s genitals) to assure an “amicable” social arrangement. Baboon infants are an attraction by themselves, not only to the citizen scientist observers, but also for other baboons. They are kind of “kidnapped” (here called “infant handling”) by other group members frequently and sometimes need to be retrieved by their mothers (mostly just from some cuddling).

As this one of the first research sites, SD (standard definition) cameras often with a creaky sound were mostly used here. These cameras with a lower resolution do not constitute a problem for many questions the science team is interested in like species and number of animals recorded, behavioral questions, and e.g., human pressure, but they gave us a hard time to identify and match the chimpanzees present since many features as cuts in ears, scars etc. disappeared in a cloud of coarse pixels. Most of the newer sites however use HD cameras exclusively so this should become less of a problem as newer sites are uploaded to Chimp & See.

Join us at Chimp & See to watch and annotate videos from across Africa!

Friday, January 1, 2016

Happy New Year!

Happy New Year!!! Its been a great first yeat at with 1,723,842 classifications to date and and just over 6600 users! Looking forward to hitting the 2 million mark soon!