Distance estimation of howling golden jackals (Canis aureus) using relative sound level

Abstract

Golden jackals (Canis aureus) display a complex repertoire of calls, utilized in different communication types (e.g., marking territories, attraction of mating partners). Resident golden jackal groups can successfully be detected by active bioacoustic stimulation, as well as with passive recording devices. For monitoring, basic knowledge of the calls of the focal species and potential restrictions and strengths of the monitoring devices should be considered. We therefore tested possible applications of a low-budget autonomous recording unit for bioacoustic golden jackal monitoring and examined the following research questions: How far can group calls be detected? Can the distance to the recording device be estimated? To answer these questions, we placed 11 AudioMoth recording devices in a linear transect to record live imitated and replayed howls. For the estimation of the number of responding animals, the number of howling individuals was determined based on the maximum number of simultaneously visible fundamental frequencies in a spectrogram. To predict the distance of the playback howls to the recording devices, the relative sound level (RSL) of each call was measured and fitted in linear models. Reliable distance estimations using RSL were possible up to 400 m. Estimated number of responding animals showed a negative relationship with distance. Our results present a baseline for future studies and show that AudioMoths can be a helpful asset in distance estimation of golden jackal packs—both in passive but also active monitoring.