Identification of western North Atlantic odontocete echolocation click types using machine learning and spatiotemporal correlates

Abstract

A combination of machine learning and expert analyst review was used to detect odontocete echolocation clicks, identify dominant click types, and classify clicks in 32 years of acoustic data collected at 11 autonomous monitoring sites in the western North Atlantic between 2016 and 2019. Previously-described click types for eight known odontocete species or genera were identified in this data set: Blainville's beaked whales (Mesoplodon densirostris), Cuvier's beaked whales (Ziphius cavirostris), Gervais' beaked whales (Mesoplodon europaeus), Sowerby's beaked whales (Mesoplodon bidens), and True's beaked whales (Mesoplodon mirus), Kogia spp., Risso's dolphin (Grampus griseus), and sperm whales (Physeter macrocephalus). Six novel delphinid echolocation click types were identified and named according to their median peak frequencies. Consideration of the spatiotemporal distribution of these unidentified click types, and comparison to historical sighting data, enabled assignment of the probable species identity to three of the six types, and group identity to a fourth type. UD36, UD26, and UD28 were attributed to Risso's dolphin (G. griseus), short-finned pilot whale (G. macrorhynchus), and short-beaked common dolphin (D. delphis), respectively, based on similar regional distributions and seasonal presence patterns. UD19 was attributed to one or more species in the subfamily Globicephalinae based on spectral content and signal timing. UD47 and UD38 represent distinct types for which no clear spatiotemporal match was apparent. This approach leveraged the power of big acoustic and big visual data to add to the catalog of known species-specific acoustic signals and yield new inferences about odontocete spatiotemporal distribution patterns. The tools and call types described here can be used for efficient analysis of other existing and future passive acoustic data sets from this region.

Fig 1. Western North Atlantic study area with long-term autonomous passive acoustic monitoring sites (red circles) and associated site name abbreviations.

HZ = Heezen Canyon; OC = Oceanographer’s Canyon; NC = Nantucket Canyon; BC = Babylon Canyon; WC = Wilmington Canyon; NFC = Norfolk Canyon; HAT = Hatteras; GS = Gulf Stream; BP = Blake Plateau; BS = Blake Spur; JAX = Jacksonville.

Fig 2. Results for Risso’s dolphin showing click type (a), acoustic presence (b), and historical sightings (c).

Click type plots, from left to right: median power spectrum (solid line) with 10^th and 90^th percentiles (dashed lines); distribution of modal ICI values from 1,000 5-minute bins; concatenation of normalized click spectra, sorted by received level; concatenation of normalized waveform envelopes, sorted by received level. For the concatenated spectra and waveform envelopes, the normalized magnitude of the frequency/pressure is represented by color such that warmer colors show greater magnitude. Acoustic presence shown as scaled circles depicting cumulative hours at each acoustic monitoring site per season, averaged across three years of data; classifier error given by color per legend in (b). Historical sightings per season (blue dots), shown relative to acoustic monitoring sites (red stars) and track lines of surveys undertaken in each season (grey lines). Inset within each sighting map shows number of sightings (N); total number of individuals summed across all sightings for which group size data was available is shown in parentheses.

Fig 3. Results for Sowerby’s beaked whale showing click type (a), acoustic presence (b), and historical sightings (c).

Subplots as in Fig 2.

Fig 4. Results for Blainville’s beaked whale showing click type (a), acoustic presence (b), and historical sightings (c).

Subplots as in Fig 2.

Fig 5. Results for Gervais’ beaked whale showing click type (a), acoustic presence (b), and historical sightings (c).

Subplots as in Fig 2.

Fig 6. Results for True’s beaked whale showing click type (a), acoustic presence (b), and historical sightings (c).

Subplots as in Fig 2.

Fig 7. Results for Cuvier’s beaked whale showing click type (a), acoustic presence (b), and historical sightings (c).

Subplots as in Fig 2.

Fig 8. Results for Kogia spp. showing click type (a), acoustic presence (b), and historical sightings (c).

Subplots as in Fig 2.

Fig 9. Results for sperm whale showing click type (a), acoustic presence (b), and historical sightings (c).

Subplots as in Fig 2.

Fig 10. Results for UD36 showing click type (a), acoustic presence (b), and historical sightings of probable species match, Risso’s dolphin (c).

Subplots as in Fig 2.

Fig 11. Results for UD26 showing click type (a), acoustic presence (b), and historical sightings of probable species match, short-finned pilot whales (c).

Subplots as in Fig 2.

Fig 12. Visually confirmed short-finned pilot whale bout.

Top panel: long-term spectrogram showing an acoustic encounter with visually-identified short-finned pilot whales at the JAX acoustic monitoring site. Bottom panel: concatenated spectra of clicks detected between 00:59 and 01:07. Most of these clicks exhibit spectral features consistent with UD26, though some natural variability is visible. In both plots the magnitude of the frequency is represented by color such that warmer colors show greater magnitude.

Fig 13. Results for UD28 showing click type (a), acoustic presence (b), and historical sightings of probable species match, short-beaked common dolphin (c).

Subplots as in Fig 2.

Fig 14. Results for UD19 showing click type (a), and acoustic presence (b).

Subplots as in Fig 2a and 2b.

Fig 15. Results for UD47 showing click type (a), and acoustic presence (b).

Subplots as in Fig 2a and 2b.

Fig 16. Results for UD38 showing click type (a), and acoustic presence (b).

Subplots as in Fig 2a and 2b.