Power of faecal pellet count and camera trapping indices to monitor mammalian herbivore activity

Abstract

Context Monitoring spatial and temporal change in relative abundance using statistically powerful designs is a critical aspect of wildlife management. Many indices of relative abundance are available, but information regarding their influence on statistical power is limited. Aims We compared the statistical power associated with occurrence-based and frequency-based indices derived from faecal pellet counts and camera trapping to detect changes in the activity of five mammalian herbivores. Methods We deployed camera traps and counted faecal pellets in native vegetation subjected to four management treatments in south-eastern Australia. We used simulation coupled with generalised linear mixed models to investigate the statistical power associated with a range of effect sizes for each combination of species, survey method and data type. Key results The index derived from camera frequency data provided the greatest statistical power to detect species’ responses and was the only index capable of detecting small effect sizes with high power. The occurrence index from camera trapping did not provide the same level of statistical power. Indices derived from faecal pellet frequency data also detected spatial and temporal changes in activity levels for some species, but large numbers of plots were required to detect medium to large effect sizes. High power to detect medium to large effects could be achieved using occurrence indices derived from pellet presence–absence data, but required larger sample sizes compared to the camera frequency index. Conclusions Both camera trapping and pellet counts can be applied to simultaneously monitor the activity of multiple mammalian herbivore species with differing activity patterns, behaviour, body size and densities, in open and closed habitat. However, using frequency indices derived from camera trapping may improve management outcomes by maximising the statistical power of monitoring programs to detect changes in abundance and habitat use. Implications Frequency indices derived from camera trapping are expected to provide the most efficient method to detect changes in abundance. Where the use of cameras is cost prohibitive, occurrence indices derived from pellet presence–absence data can be used to detect medium to large effect sizes with high power. Nonetheless, the cost-effectiveness of camera trapping will improve as equipment costs are reduced and advances in automated image recognition and processing software are made.