Incorporating wildlife into art has always been something of a hobby of mankind (see most cave drawings). Wildlife photography is a bit more recent, and using wildlife photography for scientific purposes is even more recent. But, as amusing as the image of an intrepid scientist standing perfectly still with a Nikon in front of her face—day and night, for weeks on end, hoping to blend in and waiting for animals to walk past—as amusing as that image is, it is untrue in regards to camera trapping. I wouldn’t be in the field if it was true. In the case of camera trapping, we use remotely triggered cameras, placed on trees, posts or anything that is upright and able to hold a camera, to take pictures that we use. For Science!
These cameras are usually housed inside of plastic or some other material to protect the sensitive innards from the elements; there are places one can put a bike bungie cord or rope—or in the case of Madagascar, a vine—through the camera housing, so that you can then wrap it around the tree and keep it steady and upright. Most camera traps in scientific studies are placed on trails because wildlife is not interested in self-punishment and will take the path of least resistance (i.e., a trail) instead of blundering through the bush. An infrared beam—invisible to the eye—is shot out of the camera and hovers above the trail. Think of all those movies where a thief is attempting to steal some jewel out of a museum; one touch of the beams glowing around the jewel’s casing triggers an alarm. When an animal walks past the camera, it breaks the infrared beam, this triggers the camera, and the camera takes a picture, sometimes with flash and sometimes not, depending on the type of camera you have. One photo event of an animal is called a ‘capture’; thus, comes the name camera trapping.
Scientific use of remote camera trapping began in the early 1900s with a ‘census of the living’ on Barro Colorado island in Panama by Frank M. Chapman, the Curator of Ornithology at the American Museum of Natural History, and Tappan Gregory, who worked with the several national and federal organizations in scientific expeditions, despite being an attorney by profession. But even before scientific use of camera traps became popular, people were using remote cameras—triggered by trip wires—for recreational photography; people still use it today for that express purpose. For more on camera trapping history and use in scientific studies, see Camera Traps in Animal Ecology by O’Connell,Nichols, Karanth and Ullas.
As the years went on and as camera trapping grew less expensive and more portable, it has exploded into popularity. Nowadays, digital camera traps—that use SD cards instead of film—seems to be more common, although some think that the photos aren’t as nice. Camera trapping has been used to look at a variety of species, including mice, American martens, snow leopards, owls, songbirds, coyotes, and deer. It can be used to confirm the presence of animals at a site, determine when animals are most active and estimate the rate of nest predation—and who the predators are. When placed in front of a kill that has been found by ranchers or rangers, it can shed light on who killed the animal or who is scavenging from it. If animals are individually marked—a tiger’s stripes, a leopard’s spots or even a fossa’s ear nicks—you can identify individuals through photos, estimate density or even survival rates if you have long-term data. That’s a lot of versatility for something that initially started out as a mere hobby.
How are we using camera trapping in Makira? Our camera stations are spaced at 500m apart (500m in any direction from another station); about the size of a fanaloka’s home range. We would love—and really should—have them the same size as a fossa’s home range. But seeing as a fossa’s home range can be enormous (20-70 km2) and just hiking the 500m between camera stations can be a muddy, thorny, vine-and-root ridden trial to make one think that you have entered the very bowels of hiking hell…Well, it is not very realistic.
We have twenty-four or so camera stations in a grid—think a giant square made of tinier squares, with each corner of the tinier squares equaling a camera station. Stations are located in areas that have prominent man-made trails or game trails–we want to get as many pictures as possible. Cameras are placed around 30 inches up from the ground, so that we are at the body level of most of the animals in the forest. We have two cameras at each station, placed opposite each other on the trail, so that we can get pictures of both sides of an animal (hopefully). Of course, not all pictures are clear, crisp and worthy of publication in National Geographic magazine. There are many times you get a tail tip, or a blur, or an inquisitive nose that takes up 80% of the photo. You still have to identify the animal, if at all possible.
And, as with anything mechanical, sometimes (a lot of the time) the cameras malfunction. People steal cameras. People forget to put in SD cards, or forget to take them out and pictures are overwritten. Batteries die. The flash mechanism breaks or the infrared beam breaks or it’s just TOO WET or the camera just won’t trigger because you have forgotten to sacrifice a chicken to the camera trapping gods and they are not pleased. But sometimes, you get great photos, either in quality or just because they are pictures of things that no one has seen, or could expect.
We try and check cameras every week in Makira, mostly because of the rain. When we check them, we trigger them with a placard that gives the date, time, camera station and camera ID so that we can figure out where pictures are from once they are in the computer. We then check battery level, how many pictures have been taken and switch out SD cards. Once in the computer, pictures are entered into a database. We include information such as: the number of animals in the photo, how many photos have been taken of the animal, the species, the time, the date, the station, and anything else pertinent. The camera grid at each site runs for around 60 days, because if you go shorter than that you run the risk of not getting enough pictures, and if you go longer, you run the risk of violating the assumption of demographic closure, which is essential to a variety of analyses.
There are many things that you can estimate from camera trap data, which I intend to explain in later posts. I will explain one thing that seems de rigueur to calculate: trapping success. Trapping success—is a measure—not an actual estimate!—of relative abundance. You can compare it between sites to tell you how “active” the site is in regards to a species.
For example, you have a fossa trapping success of two at one site and a fossa trapping success of one at another…the first site is more “active” in regards to fossa. This DOES NOT mean that there are more fossa at the first site as compared to the second; to figure that out, you have to estimate the true abundance (N) of both sites, which takes into account detection probability and is a bit complicated. Suffice to say, trapping success is a useful thing to calculate, but it is to be used with caution.
Welcome to the brotherhood of camera trappers. We get to carry boxes and boxes of C and D batteries! And we do not have cookies! But, certainly, now that you are well-versed in the art of camera-trapping, you want to know how we study the lemurs present at our sites. It’s a lot more hands-on (as in, hands gripping onto tree trunks to keep yourself from slipping) than camera-trapping. Read the next installment, Walk the Line!
And if you happen to be one of those people who like reading the manuals of any electronic equipment that you buy, just for the thrills, feel free to click here!