Contributors to the Arctic Acoustic Environment: Describing the Puzzle Pieces

Manuel Castellote1, Catherine Berchok2, Dimitri W Ponirakis3, Arial Brewer4, Yu Shiu5, Christopher W Clark5, Jessica M Kimber4 and Daniel Forrest Woodrich6, (1)NOAA Fisheries & U. of Washington - JISAO, Seattle, WA, United States, (2)NOAA Fisheries, Alaska Fisheries Science Center - MML, WA, United States, (3)Cornell Lab of Ornithology, Cornell University, K. Lisa Yang Center for Conservation Bioacoustics, United States, (4)NOAA Fisheries & U. of Washington - JISAO, United States, (5)Cornell Lab of Ornithology, United States, (6)NOAA Fisheries & Western Washington University, Bellingham, WA, United States
Abstract:
The Marine Mammal Laboratory of the Alaska Fisheries Science Center (NOAA Fisheries) has been collecting year-round acoustic recordings in 32 locations of the Alaskan waters since 2007. This represents a unique dataset to explore the US Arctic soundscape and its trends over the past decade. As a starting point, we focused our analysis on three locations for a period of 1 year (2016-2017) in the Bering and Chukchi seas as a representative sample of Arctic environments with low, middle and high anthropogenic influence. Data was analyzed manually to identify biotic (cetacean and pinniped), abiotic (ice) and anthropogenic (vessel and airgun) sources in the frequency range 10-8192 Hz. Pre-generated spectrograms converted to image files, indexed to allow for zoom/playback functioning during analysis, were manually scanned to detect and classify signals from different species, which included bowhead, right, humpback, gray, and minke whales, beluga, killer whale, bearded and ribbon seals, walrus, unidentified pinnipeds, as well as vessel noise, seismic airguns and ice noise. Sound file segments were extracted into several data sets based on the occurrence of these different sound sources. These data sets were then processed to obtain broadband and third octave band SPL, and spectral probability density (percentiles and spectral averages) to characterize 1) the baseline ambient noise (i.e., files without any evident sound source), and 2) the contribution of different biotic, abiotic and anthropogenic sources to the acoustic environment for each location. Sources seasonality is discussed and the dominant sources for each period of the year are described as a basis for a long-term, multiyear, evaluation of changes in the acoustic components of this Arctic environment.