Thu, Aug 18, 2022: 5:00 PM-6:30 PM
ESA Exhibit Hall
Background/Question/Methods: Locating and monitoring wildlife is essential to a variety of scientific disciplines particularly the fields of ecology, conservation biology, and animal behavior. Acoustic triangulation is a technique that uses automated recorders placed at known locations to locate the source of a sound, which can be a valuable tool to study species that are vocal. Despite the proven utility of acoustic triangulation there are several logistical challenges that limit the widespread use of this technique. In this study we validated the use of acoustic triangulation to locate the Yucatán Black Howler Monkeys (Alouatta pigra) in Belize. The goal of this study was twofold: 1) to determine the effective range out acoustic recorders to detect howler monkeys calls and 2) to quantify the accuracy of an acoustic triangulation array to locate the source of howler monkey calls coming from known locations. In Jan 2022 we deployed 10 Wildlife Acoustic SM4TS units in a grid pattern on a 466-ha private reserve in southern Belize. Recording units were set to record continuously over seven days and the location of actively calling monkeys was noted by direct observation of the researchers.
Results/Conclusions: We were able to record a total of 15 independent howling bout in which we able to determine a known location of the calls from the direct visual observation of the calling bout. Of calling bouts originating from known locations, 10 were of insufficient recording quality or were detected by two or fewer recorders making them unusable for acoustic triangulation. We calculated the distance between the true location of the calls and its estimated location (as determined by acoustic triangulation) to be 556 ± 491m (mean ± 1SD). This relatively high error is likely attributed the environmental heterogeneity as well as poor audio quality associated with large recording distances. The distance that the calling troops were located from the microphone had a significant effect on detectability (P< 0.001), resulting in an approximate 55% call detection probability based on our array spacing. Overall, we found that microphone spacing utilized in our pilot study was insufficient for triangulation accuracy, which resulted in poor recording quality and relatively low call detectability. In future studies we plan to decrease recorder separation to resolve these issues.
Results/Conclusions: We were able to record a total of 15 independent howling bout in which we able to determine a known location of the calls from the direct visual observation of the calling bout. Of calling bouts originating from known locations, 10 were of insufficient recording quality or were detected by two or fewer recorders making them unusable for acoustic triangulation. We calculated the distance between the true location of the calls and its estimated location (as determined by acoustic triangulation) to be 556 ± 491m (mean ± 1SD). This relatively high error is likely attributed the environmental heterogeneity as well as poor audio quality associated with large recording distances. The distance that the calling troops were located from the microphone had a significant effect on detectability (P< 0.001), resulting in an approximate 55% call detection probability based on our array spacing. Overall, we found that microphone spacing utilized in our pilot study was insufficient for triangulation accuracy, which resulted in poor recording quality and relatively low call detectability. In future studies we plan to decrease recorder separation to resolve these issues.