Deep-sea soundscapes of Japan

This dataset is an archive of acoustic data associated with the deep-sea soundscapes of Japan. Python codes to visualize the audio data are also provided in a notebook based on Google Colab.

Recording Locations

Four deep-sea ecosystems around Japan were recorded. This includes the benthic habitats off Sanriku, the twilight zone of Suruga Bay, the hydrothermal vents of Suiyo Seamount, and the abyssal plains off Minamitorishima Island.

  • Off Sanriku: multiple recording locations were surveyed between July 21 and August 5, 2019 within an area enclosed by 39°02.3486’N, 39°24.1055’N, 142°08.2092’E, and 142°27.0315’E. The depth range was between 250-1011 m.
  • Suruga Bay: one recording location was surveyed November 4-5, 2019 at 34°58.3952’N, 138°43.2089’E. The depth was 1686 m.
  • Suiyo Seamount: three recording locations were surveyed between August 21-25, 2019 at 28°34.2715’N, 140°38.6216’E, 1384 m (an active hydrothermal vent without visible bubbling), 28°34.1438’N, 140°38.9641’E, 994 m (caldera rim), and 28°34.3288’N, 140°38.5758’E, depth 1381 m (an active vent orifice).
  • Minamitorishima Island: one recording location was surveyed on March 14, 2020 at 23°05’N, 153°50’E, depth 5500 m.

Acoustic Recorders

Two recording systems were used. The first system, designed to a depth rating of 2000 m, was deployed in Sanriku waters, Suruga Bay, and Suiyo Seamount. It was composed of a DS-850 digital recorder (Olympus, Japan) and a AQH-020D hydrophone (AquaSound Inc., Japan). The AQH-020D hydrophone has a sensitivity of -195 dB re 1V/μPa and a flat frequency response between 20 Hz and 20 kHz. The second system was designed to 6000 m depth rating (off Minamitorishima Island). It was based on an LS-P4 digital recorder (Olympus, Japan) and a HTI-94-SSQ hydrophone (High Tech Inc., U.S.A.). The HTI-94-SSQ hydrophone has a sensitivity of -198 dB re 1V/μPa and a flat frequency response between 2 Hz and 30 kHz.

Configuration of Audio Recording

(1) Duty Cycle: continuous. (2) Sampling Rate: 44.1 kHz (off Sanriku, Suruga Bay, and Suiyo Seamount) and 88.2 kHz (off Minamitorishima Island). (3) Channels: 1 (single input but stereo output). (4) File Format: WAV. (5) Audio Gain: High. (6) High Pass Filter: Off.

Field Deployment

At each deployment, one recorder was fixed at the seafloor. Each deployment lasted for from a few hours to a maximum of two days. Deployment and recovery of recorders were conducted by free fall camera systems (off Sanriku), Edokko Mark 1 (Suruga Bay and off Minamitorishima Island), DSV Shinkai 6500 (Suiyo Seamount). Operational noises produced from these deep-sea observation platforms and research vessels were also recorded.

Data Processing

We used the LTSA_gui to generate long-term spectrograms (LTS) and save the LTS in mat files. Each mat file contains median-based LTS and mean-based LTS. Median-based LTS was obtained by measuring median power spectral densities within each 1-min segment. Mean-based LTS was obtained by measuring mean power spectral densities within each 1-min segment. Both types of LTS have a frequency resolution of 10 Hz.

Associated Publication

Chong Chen, Tzu-Hao Lin, Hiromi Kayama Watanabe, Tomonari Akamatsu, Shinsuke Kawagucci. Baseline soundscapes of deep-sea habitats in Japan reveal heterogeneity among ecosystems and sensitivity to anthropogenic impacts, under review.

Data and Resources

Tags

Wikidata Keywords

  • Q1358257
  • Q686881
  • Q5250646

Basic Information

Data Type
  • Source code
  • Audiovisual data
  • Scientific and statistical data formats
Language English (eng)

Spatio-temporal Information

Temporal Resolution Daily
Start Time 2019-07
End Time 2020-03

Management Information

Author Tzu-Hao Lin, Chong Chen, Hiromi Kayama Watanabe, Tomonari Akamatsu, Shinsuke Kawagucci
Contact Person Tzu-Hao Lin
Contact Person Email schonkopf@gmail.com