Estimating the effectiveness of using wildlife cameras versus visual-encounter surveys to detect herpetofauna

Abstract

Context Efficient monitoring of herpetofauna can prove challenging to agencies and NGOs responsible for their management. Wildlife cameras have been proposed as a method to monitor herpetofauna; however, estimates of detection rates and factors affecting detection by cameras are generally lacking and therefore limiting their application. Aims We determined the effect of body size and temperature differential on the detection of snakes and frogs by passive infrared (PIR) wildlife cameras. We hypothesised that detection would differ among models and be positively correlated with body size and surface-temperature differential between the animal and substrate. We then conducted a field study to compare the detection of herpetofauna by a traditional method with PIR cameras. Methods We tested 10 cameras of seven models on five snakes and one camera on six frogs. Photographs were downloaded to determine the detection rate of each species by each camera. We then chose a camera model to compare two herpetofauna survey methods, namely, drift fences equipped with cameras and visual-encounter surveys. Surveys were conducted monthly over 12 months in Irwin County, GA, USA. Key results The highest mean detection rates of snakes were 0.65 (s.e. = 0.33), 0.50 (s.e. = 0.34), and 0.49 (s.e. = 0.34) for the Browning Dark Ops, Reconyx Hyperfire 2, and Mossy Oak Covert Scouting Camera respectively. The detectability of larger snakes was greater than that of smaller snakes and increased as the absolute temperature differences between the snake and the substrate increased. The detectability of frogs was influenced by absolute temperature differential alone. PIR cameras generated five times more observations, documented more herpetofauna species, and were seven times more efficient than traditional surveys. Conclusions The effectiveness of PIR cameras to detect herpetofauna varies among models and depends on the likelihood that the animal will have a body temperature significantly different from the temperature of the substrate. PIR cameras generated observations far more efficiently than traditional sampling methods. Implications PIR wildlife cameras may be most effective at detecting larger, diurnal herpetofauna and least effective at detecting smaller, nocturnal species. Wildlife cameras have the potential to efficiently monitor some herpetofauna, providing a means to better evaluate management objectives.