Feasibility assessment of active and passive acoustic monitoring of sika deer populations

Abstract

Regardless of whether populations of sika deer (Cervus nippon) are native or introduced, their distribution continues to expand, presenting new ecological threats in several regions of the world, especially Japan. Consequently, there is growing demand for precautionary measures against such expansion, which is primarily associated with males during the initial stage of invasion. Therefore, the present study introduces two novel approaches—passive acoustic monitoring (PAM) and active acoustic monitoring (AAM)—to detect males with high sensitivity, using their howls during the rut. This study verifies the feasibility of these approaches by comparing them with existing methods based on spotlights and camera traps at five sites that support different deer densities. To validate PAM, we set unattended sound recorders to evaluate the time and variation in howl frequency at different sites. In addition, we measured the detection range of the howl. To verify AAM, we assessed whether sika deer males are vocally responsive to audio playbacks, and if so, the extent to which the frequency of the howl-back against different sound sources could explain male abundance. Both the PAM and AAM successfully detected males, even at sites with extremely low deer density where detection using existing methods failed. PAM had a detection zone of around 6ha in defoliated forests, which was >200-times greater than that of camera traps. The frequency of howl-back immediately after playbacks was higher than that of spontaneous howls; thus, AAM proved effective at explaining male abundance when using playbacks of real howls. In conclusion, our findings suggested that bioacoustic approaches could enhance the monitoring of low-density deer populations that might be in the initial stage of deer invasion or at the stage where the target density has been reached following population control.