PS 37-202 - A quantitative evaluation of the recording performance of the open source AudioMoth autonomous recording unit

Tuesday, August 16, 2022

5:00 PM – 6:30 PM EDT

Located in: ESA Exhibit Hall

Presenting Author(s)

JK

Justin Kitzes

Assistant Professor
University of Pittsburgh
Pittsburgh, Pennsylvania, United States

Coauthor(s)

SL

Sam Lapp

University of Pittsburgh
Pittsburgh, PA, United States

Background/Question/Methods

Bioacoustic field surveys are becoming an increasingly popular method for estimating the occupancy and abundance of sound-producing taxa, including birds, bats, anurans, and insects. The scope and scale of such field surveys have historically been limited by the cost of commercial autonomous recording units (ARUs). In recent years, several new inexpensive, open source ARU designs have been released. One of the most popular of these is the AudioMoth recorder designed by Open Acoustic Devices, which can be purchased for less than US$80. In support of our group’s field deployments involving several thousand AudioMoths, we conducted extensive tests of the audio recording performance of the AudioMoth v1.1 and v1.2. Here, we report our results from three sets of tests. First, we quantify the AudioMoth’s on-axis (i.e., a sound directly in front of the recorder) and polar frequency response curves from 100 Hz to 17 kHz, approximately the audible frequency range for humans. Second, we measure the effects of several protective housings on on-axis and off-axis frequency response. Third, we examine the effects of mounting an AudioMoth on a tree.

Results/Conclusions

Without a case or other housing, we find the on-axis frequency response to be relatively flat below 2 kHz, with an increase in sensitivity of approximately 7 dB at higher frequencies, performance that we consider acceptable and roughly comparable to other commercial microphones. The polar response showed that the AudioMoth is relatively omnidirectional, with highest losses directly behind the device and in frequencies above 10 kHz. We found that placing the AudioMoth inside a Ziploc or vacuum seal bag resulted in surprisingly little degradation in recording quality below 10 kHz, with a drop of 10-20 dB above this frequency. An injection molded case designed by Open Acoustic Devices was found to significantly increase directionality, increasing the amplitude of sounds arriving from the front of the device while reducing sensitivity in other directions. Mounting the AudioMoth on a medium-sized tree results in a loss of approximately 25 dB directly behind the device, equivalent to the recorder capturing sounds approximately four times farther away in front of the recorder than behind. We conclude that the AudioMoth has acceptable performance for field bioacoustics surveys, with an overall qualitative impression that the recorder has slightly but not significantly poorer performance than commercially available ARUs.