Abstract: Data from passive acoustic monitoring (PAM) of Navy training ranges can be used to test predictions of marine mammal responsiveness to naval sonar from experimental behavioral response studies. Minke whale (Balaenoptera acutorostrata) vocalizations were detected, classified, localized, and associated into individual tracks from data collected on bottom-mounted hydrophones on the U.S. Navy’s Pacific Missile Range Facility located off Kauai, Hawaii. Data were analyzed from before, during, and after naval undersea warfare training activities in February 2014 and February 2017. The spatial distribution of acoustically derived whale tracks was modelled using “latitude,” “longitude,” “depth,” “year,” and “phase” (Before, During, and After) as covariates. It was found that, as well as inter-annual variability across all phases, the distribution of calling whales changed in both years, with fewer tracks in the vicinity of the center of ship activity in the During phase compared with the Before phase. This redistribution of calls extended for approximately five days after the end of the training activity in both years. As the data are from PAM, it is unknown whether this change in the probability of acoustic presence across the range is a result of whales moving away from the range or ceasing calling, or perhaps a combination of both. The expectation from experimental studies is that minke whales will exhibit avoidance in response to naval sonar activity. We conclude that this is a probable explanation for the results reported here, particularly given the nature of the redistribution (an increase in some areas distant from the center of ship activity coincident with a decrease near the center) and the time taken for the distribution observed in the Before phase to re-establish in the After phase of both years.
Key Words: naval sonar, passive acoustic monitoring, spatial distribution, behavioral response
DOI: https://doi.org/10.1578/AM.45.6.2019.661
Page Numbers: 661-674
Open access: Click here for PDF