Initially, remote cameras were only used to photograph animals for aesthetic reasons. The first camera traps were basic and utilised a baited tripwire system whereby a target animal would bite and pull on a piece of lure. This action would mechanically trigger a film camera set up nearby. In the 1880s, George Shiras III pioneered the use of night-time flash photography and tripwire photos, and his unique images won international acclaim. Over the next century, zoologists started realising the potential use of camera traps in research, but the available equipment was still very expensive and bulky and researchers had to build their own systems. In 1964, two biologists boasted that their camera system weighed ‘only 47pounds’ per unit (21.4 kg)!
During the 1980s there was a rapid expansion of camera trap technology and manufacturers in the US scrambled to meet the new demand. This was not driven by biologists, though, but primarily by deer hunters who wanted ‘game cameras’ to scout choice hunting sites. A population study of tigers by Indian zoologist Ullas Karanth in the 1990s was a watershed moment as it showed that camera trapping could be a tool for robust scientific monitoring. Since then, the field of camera trapping has grown exponentially, with analogue film cameras being replaced by digital; camera housings becoming ever smaller and lighter; the development of faster and more efficient triggers; and a booming commercial market. The data analysis side has seen an equally steep growth curve, with sampling methods and statistical models becoming increasingly more refined and robust, able to incorporate more variables and handle huge quantities of data with ease.
Camera traps are now used in a myriad different ways, including estimating species richness and the structure of biological communities; population estimates and dynamics; species distribution and animal behaviour; the monitoring of human-wildlife conflict and law enforcement in poaching hotspots; and citizen science initiatives and education drives.
These tools have become invaluable for pure scientific research as well as conservation and management. Like all tools, camera traps have both pros and cons. Some of latter include that they are subject to animal and environmental damage, theft and vandalism. If they malfunction a significant amount of data can be lost; mistakes in set-up can have expensive consequences; and they sample a relatively small area. However, these are far outweighed by the pros – camera traps are noninvasive; have a negligible effect on animal behaviour and no observer bias; they can function for weeks or months without requiring attention; they work for species that cannot be studied using ‘traditional’ means; and they provide robust data. They are like permanent fieldworkers, on duty day and night, rain or sunshine. They are non-selective, capturing everything that moves, thus providing ideal opportunities to gather data on other animals as well. An added bonus is the educational and promotional value of the resulting images which can be used to great effect by organisations and researchers.
Most camera traps have what is called a passive infrared (PIR) sensor – an electronic sensor triggered by body heat and movement as an animal passes in front of it. This means that pockets of hot air (sometimes generated by large rocks) or wind-blown vegetation moving in the detection zone can also trigger a camera – which can result in thousands of ‘empty’ photos. In one CLT survey, a single daisy blowing just in front of a camera was responsible for 10816 photos! It’s thus of utmost importance to choose the right site, and to properly prepare it before arming the camera. When optimising a survey for leopards, we want to be quite sure that these big cats will pass the camera at some stage. Fortunately, when given a choice, leopards will often take the path of least resistance, and they often patrol their territories via well-used game paths, hiking trails, quiet jeep tracks, dry watercourses, and similar byways. These areas normally prove to be good camera locations. We also look for places where animal movement is naturally channelled by geographic features such as ridge lines, saddles, dense vegetation or large boulders. Before deciding on a site, we also always scout the area for leopard field signs – spoor (tracks), scat (droppings), scratch marks and scrapes, and kill sites – to determine the likelihood of leopard presence.
Once a site has been chosen, the cameras are set up. To obtain good lateral flank shots of a passing leopard, the cameras need to be set around 25cm from ground level and perpendicular to the path, about two to three metres from where the animal will be walking. Each leopard has a distinctive rosette pattern, but the left and right flanks differ – it is not a mirror image – therefore two cameras are set opposite each other to get a simultaneous left and right photo to serve as a cat’s ‘identikit’. It’s of utmost importance to sync the time settings on these two cameras perfectly and to double-check all parameters, to ensure that when the team visits the cameras after a few weeks, there are no data issues or gaps.
Once a camera station has been set and reference photos taken, the ‘waiting game’ starts. It’s a game informed by a good understanding of your target species, but dictated by a lot of patience – and a little bit of good luck!
