01727nas a2200157 4500008004100000022001400041245010700055210006900162260001600231300001600247490000800263520120500271100002301476700001901499856005101518 2018 eng d a0001-496600aDetection and classification of whales calls using band-limited energy detection and transfer learning0 aDetection and classification of whales calls using bandlimited e cJan-09-2018 a1769 - 17690 v1443 a
The Monterey Bay Aquarium Research Institute has been recording since July 2015 almost continuously at the Monterey Accelerated Research System (MARS) cabled observatory in Monterey Bay, California, USA. This long-term recording contains thousands of whale calls to help further our understanding of interannual, seasonal, and diel patterns. Here we report on our highly accurate detection and classification method developed to classify blue whale A, B, and D calls and Fin whale 20 Hz pulses. The foundation of the method is a computationally efficient and tuned decimation filter to convert the broadband hydrophone 128 kHz signal to 2 kHz which preserves the low-frequency signal and avoids any high-frequency aliasing. Detection is done using a band-limited-energy-detection filter to find potential calls in the decimated data. Spectrograms are then generated for potential calls and enhanced with local image normalization followed by smoothing by convolving in either time or frequency. Classification is done using the Google Inception v3 model with a transfer learning method. Overall, false positive rates are very low despite variability in whale call shape and background noise.
1 aCline, Danelle, E.1 aRyan, John, P. uhttp://asa.scitation.org/doi/10.1121/1.506782502437nas a2200265 4500008004100000022001400041245014000055210006900195260001600264300001600280490000800296520158400304100001901888700002301907700002101930700002301951700002501974700001801999700001702017700002402034700001802058700002202076700002202098856005102120 2018 eng d a0001-496600aTemporal variations in humpback whale (Megaptera novaeangliae) song in Monterey Bay National Marine Sanctuary, northeast Pacific0 aTemporal variations in humpback whale iMegaptera novaeangliaei s cJan-09-2018 a1953 - 19530 v1443 aUsing two years of nearly continuous recordings from Monterey Bay National Marine Sanctuary, August 2015 through July 2017, variations in humpback whale song are examined on diel, seasonal, and interannual time scales. The cabled hydrophone is in humpback feeding and migratory habitat at 36.7128°N, 122.186°W. Diel analyses show 69% of song during night, 23% during day, and 8% during dusk or dawn. Seasonal analyses show song absence during summer (June–August), emergence during fall (September–October), peak during late fall/winter (November–January), and highly variable detection during spring (February–May). During both years >80% of song occurred during the November–January peak. Song detection within a month reached a maximum of 58% of the time during November 2016. Song length increased (p < 0.01) month-to-month from the start in fall through the end of the peak in January. The months of maximum song occurrence coincide with declining visual sighting of humpbacks within Monterey Bay, consistent with seasonal southward migration to breeding habitat and/or more offshore residence. Interannual variation in song was marked by a 51% increase between the first and second years. Habitat during the first year was strongly influenced by a prolonged regional oceanic warm anomaly and the largest toxic algal bloom ever recorded in the northeast Pacific. Alternative hypotheses for the interannual difference in song detection relate to differences in humpback regional abundance, behavior, and habitat occupancy.
1 aRyan, John, P.1 aCline, Danelle, E.1 aJoseph, John, E.1 aMargolina, Tetyana1 aStimpert, Alison, K.1 aForney, Karin1 aBlack, Nancy1 aDeVogelaere, Andrew1 aFischer, Mark1 aWahl, Christopher1 aChavez, Francisco uhttp://asa.scitation.org/doi/10.1121/1.506853702120nas a2200181 4500008004100000022001400041245010500055210006900160260001600229300001600245490000800261520152800269100002101797700002701818700001901845700002301864856005101887 2018 eng d a0001-496600aContinuous monitoring and classification of odontocete echolocation clicks from an ocean observatory0 aContinuous monitoring and classification of odontocete echolocat cJan-09-2018 a1922 - 19220 v1443 aPassive acoustic monitoring is a valuable technique for detecting the presence and inferring the activities of odontocetes, but is often limited by power and data-storage constraints of self-contained recorders. Since July 2015, a broadband (10 Hz–128 kHz) hydrophone has been recording continuously at the Monterey Accelerated Research System (MARS), a cabled observatory in Monterey Bay, California, USA. This acoustic record is notable for its combination of duration, bandwidth, and completeness. An automated detector identified more than 200 million unique clicks in this dataset, and extracted a variety of time-domain, spectral, and cepstral features from each click. Depending on the criteria used, clustering algorithms identified 4-8 click classes consistent with those of local odontocete species, including dolphins and beaked whales. Clicking rates were highly variable, with a median of 0.2 clicks minute-1 and a mean of 165 (± 130 standard error) clicks minute-1. Echolocation activity was 1-2 orders of magnitude higher at night than during daytime, and was seasonally higher in fall and winter. Prior work has shown that mesopelagic sound-scattering layers in Monterey Bay are densest at these times of year, suggesting higher availability of prey. Continuous passive monitoring has great potential to improve our understanding of these species’ foraging ecology, especially when integrated with environmental measurements and active acoustic measurements of their prey fields.
1 aUrmy, Samuel, S.1 aBenoit-Bird, Kelly, J.1 aRyan, John, P.1 aCline, Danelle, E. uhttp://asa.scitation.org/doi/10.1121/1.5068416