In Japan, where the forest coverage is high and the terrain is steep, there was no established methods for estimating density of wild boar (Sus scrofa), which are difficult to identify by appearance. In recent years, density estimation methods using camera traps without individual identification, the Random Encounter Model (REM) and the Random Encounter and Staying Time Model (REST), have been developed and are beginning to be used in category 2 specified wildlife control plan for wild boar. However, there have been no comparative studies of REM and REST targeting wild boar populations in Japan, and it is necessary to verify the practicality of these methods. Therefore, this study aimed to compare the estimated population size of REM and REST in an isolated wild boar populations in Gifu Prefecture. We installed 20 infrared triggered cameras from October to December in 2020 and 2021, and estimated population densities by both methods. The median population estimates in 2020 and 2021 using conventional REM, REM using REST’s effective detection zone, and REST were 21 and 9, 122 and 41, and 101 and 53, respectively. Considering the number of captures in this study area, the density estimated by conventional REM is an underestimate, and it is important to define the effective detection zone. On the other hand, it was suggested that REM and REST, which define the effective detection zone, are capable of equivalent density estimation.
View full abstract