Camera Traps

Camera traps are widely used in wildlife conservation to monitor animal presence, the number of sightings over time, and animal behaviour. They can be set to automatically take photos or videos when an animal is detected. Most camera traps use passive infrared (PIR) sensors that trigger the camera when an animal passes through "detection zones." PIR sensors do not actually detect movement but rather changes in temperature between the detection zones. The surface temperature of an animal is usually different from the background temperature. When an animal passes in between detection zones, the change in temperature is registered, and the camera is triggered (Figure 1).

What can camera traps tell us?

Camera traps are ideal for monitoring mammals, except for smaller mouse-sized species, as they can be difficult to distinguish. At their most basic level, cameras can confirm the presence of a species by focusing one or a few cameras in areas where signs of a particular species are suspected. Where  species have distinctive patterns, like quolls, numbats, and cats, individual identification is possible  and this can be used to provide an understanding  of  home range or population size. However, as it is not easy to distinguish individuals for most  species, camera traps provide data on occurrence or detection frequency, which helps to understand  aspects such as a species distribution or with enough information, habitat preferences. Although detection frequency does not directly estimate population size, changes in detection frequency can correlate with population trends when using well-designed camera trap setups. Long-term monitoring can reveal species trends influenced by management actions or natural variability. This information can be important  for identifying when actions, such as cat control, are needed to protect native species like bilbies or quolls. Coordinated programs using consistent methods across large areas can provide valuable data on a broad scale. Another analysis approach is  occupancy, which tracks the number of monitored sites where a species is found and how this changes over time.

Which camera traps should I use?

When deciding which camera traps to purchase for wildlife monitoring, it is important to understand the specifications and features that contribute to their effectiveness. Camera trap durability, image quality, PIR and flash range, trigger speed, and battery life directly influence their ability to capture accurate data. While high-resolution images and videos can assist in species identification it's important to note that claims of higher resolutions often refer to post-processing interpolation rather than true sensor quality, potentially resulting in larger file sizes without significant improvements in image clarity. For example, Reconyx cameras at 3 MP can produce sharper images than many other models claiming much higher resolutions, as can be seen in Figure 2. Shutter speed plays a critical role in capturing clear, identifiable images, especially in low-light conditions, ensuring minimal blur of wildlife activity. Additionally, a well balanced infrared flash helps to illuminate subjects evenly without either overexposing or underexposing the animal or overall image and improving ease of identification. Fast trigger speeds ensure images are captured quickly after detection and before they have moved from the field of view. Maintaining settings including date and time when batteries are flat or removed significantly reduces setup times and minimises the possibility of incorrect settings each time a camera is reset. A long warranty period and a repairable design further ensure prolonged functionality and cost-effectiveness.

Imperfect detection - camera traps sometimes miss animals

No matter what camera trap is used, sometimes they can fail to trigger due to several factors, including, placement, the speed of motion, the size of the animal, and the direction of movement relative to the PIR sensors. This "imperfect detection" is more likely to occur with reptiles than with birds and mammals because reptiles often have a surface temperature similar to the background temperature. While this does not mean reptiles are never detected, they may often go undetected. It is challenging to determine the extent and frequency of non-detections for any species or group without a carefully designed experiment. Detection also varies due to performance differences among various camera trap models, as well as the size of the species and their distance from the PIR sensors.

Choosing a camera trap height for maximum detection

Camera traps mounted high in trees pointing downwards have a limited detection zone and a relatively large minimum distance to detection, reducing the likelihood of detecting small animals and missing animals that would otherwise be seen by a camera lower to the ground facing outwards (Figure 3).

Similar variation can occur between cameras facing outwards but set at different distances from the ground (Figure 5). While a camera set high from the ground will be fine for detecting large animals such as kangaroos or larger, they will frequently miss many smaller species, including animals the size of cats or foxes. Conversely, cameras lower to the ground will detect both small and large species. As a general rule, the height of a camera should be close to the maximum body mass of the smallest species targeted. There are caveats on this though as too low and variation in the height of the terrain and other low obstacles obscure detection. In most circumstances, having the PIR sensor at 30 to 40cm (Figure 4) above ground level (~ maximum body mass of a cat or similar sized animal) has been found to be effective for capturing a range of species from the size of dunnarts through to cattle or camels.

The Challenge of False Triggers

The opposite issue, false detections, occurs when cameras take many "false" shots where no animals are present. This can happen for various reasons, including rapid changes in background temperature due to vegetation movement, camera movement, temperature changes from cloud cover, the sun directly hitting the front of the camera, and the sensitivity of the camera's PIR sensor. Most false detections are likely to occur on sunny, windy days but will be fewer at night when surface temperatures become more uniform. Careful positioning of cameras and the way they are mounted should be used to minimise false detections. Cameras should be mounted on solid object that don’t move in the wind such as stakes or star pickets using a solid mounting system such as a bolt where possible (Figure 6). When attaching cameras to a tree it is important to make sure the attachment point does not move in the breeze as this will invariably result in false detections. Unstable mounts such as cable ties should be avoided and only used where no other options are available, and never used in long term camera deployments. Turning down the PIR sensitivity can be used to reduce false detections but should be a last resort as this will also affect animal detections.

Getting the best results from your camera traps

Where arboreal or tree climbing species are targeted, a useful system is to attach a timber platform to the tree using a right-angle bracket and position the camera at the end of the beam facing towards the tree trunk (Figure 7). This system can include lures to improve detection of species for presence absence studies.

Obstructions, like spinifex, in the detection zone can significantly impact the effectiveness of camera traps. By choosing camera positions where such obstructions are minimized, detection rates improve (Figure 8). Additionally, positioning cameras to face in a southerly direction can reduce false triggers and lens flare caused by direct sunlight. Proper setup and planning are crucial for efficient data collection, avoiding the frustration of processing thousands of false detections due to poor camera placement.

Good planning and, high quality cameras, understanding how they work and the best settings and mounting for target species makes the difference between capturing lots of useful information efficiently, or putting a lot of work and resources in for little useful information and often many thousands of false detections. And, having to go through tens of thousands of false detections due to poor camera setup, all of which still all have to be processed in case there are important images among them, results in a lot of frustration and wasted effort.

Batteries are a critical component for camera trapping and using high quality batteries is essential to get the most reliable operation from cameras (Figure 9). The best batteries to use are rechargeable slow self drain NiMh batteries with a capacity of 1900-2000mah. Higher capacity batteries under harsh environmental conditions have often proved to not perform as well by being   less reliable and having a lower overall life at a significantly greater cost. Under extreme condition lithium batteries will perfor better than other chemistries however they are not rechargeable and are expensive, particularly if the project uses a large number of cameras.

In summary

The best general setup for cameras is:

• Solidly mounted via a bolt to a stake or peg.
• Facing outwards.
• No obstructions within the effective PIR range.
• PIR 30 to 40cm above ground level.

Ideal internal settings for the camera are:

• PIR set to high.
• Trigger events to take three images in rapid succession with no wait period between trigger events.
• Shutter speed and ISO should be balanced so that moving animals are captured with minimal blur.
• Programming can be used to schedule cameras to only come on late afternoon or at dusk and off again at dawn or early in the morning if target species are not active during the day- this will help reduce the number of false triggers.

Best practice camera management:

• Each camera should be individually labelled and have its own “a” and “b” SD cards labelled the same as the camera.

   

• The camera should be set so that the label information is transferred to the images to ensure information can always be traced back to a particular camera and its location.
• When setting the camera up in the field its important to record the date and time of where a camera was placed and turned on, the label of the camera, a site number, the site coordinates and level of battery charge.
• When cameras are serviced, e.g. have their batteries replaced and/or the SD card swapped over, this same information should be recorded so the status of the camera and its location at any point in time is always known.
• The coordinates are very important for any landscape scale assessments as these will often be used for mapping where cameras where or what species and how many were found at any particular location.