Fig 1.
Map of the study area, drawing of the FAAE platform, and field of view for both HD camera and DIDSON sonar.
A) Map of the study area with precise location of seafloor instruments and the FAAE platform. Map reprinted from ONC under a CC BY license, with permission from ONC, original copyright 2024. B. Scaled drawing of the FAAE platform with placements of video, sonar, scaling lasers, LED lights, Mini Junction-Box (MJB) housing all device controlling electronics; and the connector plug where fiber-optic and power are exchanged with the main instrument platform. C. Field of view of the camera showing the bait dispenser (b) and hydrophone (h) in frame. D. DIDSON sonar field of view cone (14° deg. horizontal) showing the backscatter signal and the positions of the bait dispenser and hydrophone.
Table 1.
Elephant seal presence at Barkley Canyon Node.
Details provided for each period (dates range, total events as hourly bins, total events on the HD videos and total events on the sonar videos) and for each identified elephant seal (age class, dates range of presence, timespan, total events and count of different days for which each elephant seal has been observed).
Fig 2.
Elephant seal presence duration and first appearance.
A) Box plot of elephant seal presence duration (in seconds) between the first and last appearances on HD camera (red) and sonar (blue) videos. Sonar videos (n = 210) are approximately 15 min long each. Camera videos (n = 113) are approximately 4 min 40 sec long each. B) Box plot of first appearance (in seconds) of the elephant seals on the HD videos for periods when the sonar had failed (green, n = 24) or was functional (orange, n = 89). Middle line shows the median, while lower and upper lines of the boxes show 25th and 75th percentiles, respectively. The ends of the vertical line indicate minimum and maximum values and the dots indicate outliers.
Fig 3.
Elephant seal behavior for each period.
Stacked bar chart illustrating for the 210 sonar videos with elephant seal presence, the count (# for each behavior) of each behavior relative to each period (Before: Before lights turn on; During: When lights are turned on and HD camera is recording; After: After lights are turned off). Each period is approximately 5 min long. Behaviors included: Resting (lying down or staying immobile), Passing through (swimming) or Absent (if a resting animal was suddenly swimming away when lights turned on and did not come back later it was counted as absent in the period corresponding to the HD camera recording).
Fig 4.
Occurrence of individual elephant seals at Barkley Canyon Node.
A) Diel and hourly occurrence of elephant seals at Barkley Canyon Node in 1-h bins. Light grey areas indicate that either the DIDSON sonar or the video camera were recording, or both. Dark grey shaded areas indicate elephant seal presence only in the sonar (example at the top). Black areas indicate unidentified seals present on video recordings only or on both video and sonar recordings. Colored areas indicate identified individuals (Brian, Dennis, Carl, Mike, Al, David, Blondie and Bruce) on the video camera. White areas indicate a lack of any recordings. Photos reprinted from ONC under a CC BY license, with permission from ONC, original copyright 2024. B) Diel pattern of elephant seal presence as individuals (colored bars), unidentified (black bars) on the video camera or found only on the sonar (dark gray bars), expressed in local time (UTC– 8hr).
Fig 5.
Elephant seal behavior types and possible prey types observed in HD videos.
Percent occurrence of elephant seal behavior types within 113 HD videos where elephant seals were observed A) overall behavior in the video; B) behavior observed at the start of the video recording; C) possible prey types observed in the same video as the elephant seal (not necessarily at the same time).
Fig 6.
Foraging behavior and diet of elephant seal.
A) On July 27, 2022 one individual (Dennis) elephant seal (ES) actively hunted the two swimming sablefishes (SF, on the left) but ignored the immobile sablefish (SF, in the center). B) On August 2, 2022 one individual (Dennis) elephant seal (ES) successfully caught one sablefish (SF). On both May 16 (C) and May 17 (D), 2023 one individual (Bruce) elephant seal (ES) successfully caught one sablefish (SF). E) On August 5, 2022, in the background on the video, one individual elephant seal (ES) pursued and apparently successfully caught that seemed a squid (SQ, based on the propulsive swim). F) On July 23, 2022 one individual (Dennis) elephant seal (ES) mistakenly caught a snailfish (SN) in an attempt to capture a sablefish. The snailfish was rapidly released. [Videos for events A–F are in Supplemental materials]. Photos reprinted from ONC under a CC BY license, with permission from ONC, original copyright 2024.
Fig 7.
Sound produced by the sonar and northern elephant seal hearing range.
A) Waveform (top) and spectrogram (bottom) of sound production by the DIDSON acoustic camera as it transitioned from continuous pings approximately every second produced in the standby mode, to the more rapid ping rate of about 4 pings/s while in recording mode. Spectrogram parameters: 1024 point Hanning windowed FFT with 50% overlap. B) Spectrum density level for periods when the Didson was turned ON (scanning mode; 15 min every hour) on 10 June 2022 (light blue line), when the Didson was turned OFF (standby mode; 45 min every hour) on 10 June 2022 (dark blue line), and without the Didson on 10 June 2023 (black line). The dashed line is the estimated northern elephant seal audiogram based on behavioral threshold data from Kastak and Schusterman [60].
Fig 8.
Six examples of the clearest low frequency pulses corresponding to head bobbing motions of seals in close pursuit of sablefish.
Each panel contains the waveform (relative dB) on top and spectrogram on the bottom (1024 point Hann windowed FFT with 50% overlap, 2 kHz filter). Time is in UTC. A) June 28, 2022 at 17:11 (UTC) 2 head bobs, sablefish out of view but dust cloud suggests startle or strike; B) July 4, 2022 at 02:11 (UTC) four pulses, last two correspond to 2 head bobs as seal enters view chasing sablefish, sablefish startles; C) July 4, 2022 at 12:14 (UTC) six pulses corresponding to six head bobs as seal chases sablefish, last head bob occurs as seal move off screen, sablefish out of view but evasive; D) July 24, 2022 07:12 (UTC) two pulses corresponding to 2 head bobs as seal chases sablefish, sablefish strong startle. E) December 19, 2022 at 04:14 (UTC) 3 pulses corresponding to 3 head bobs as seal chases sablefish close to hydrophone, sablefish offscreen but evasive; F) May 15, 2023 at 05:14 (UTC) 10 pulses corresponding to 10 head bobs as seal approaches sablefish which does not startle but glides slowly (no fin movements) to shelter under instrument platform. G) and H) Comparison of the normalized relative power spectra of sounds produced during head bobbing (Hanning FFT 32768 samples, spectral resolution 3.9 Hz) on two dates: G) July 4, 2022 at 2:14 (UTC, spectrogram shown in C), H) December 19, 2022 at 04:14 (UTC, spectrogram shown in E).