Physical Sources of High‐Frequency Seismic Noise on Cascadia Initiative Ocean Bottom Seismometers

Abstract

Physical sources of high-frequency seismic noise in the ocean are investigated using data from the Cascadia Initiative (CI) ocean bottom seismometer (OBS) network, hindcasts of wind speed, waves, and the bottom currents predicted by a regional ocean circulation model and observed at sites on cabled observatories. Seismic data in the 5-12 Hz band are considered because it is best for detecting regional earthquakes and lies between the frequencies of local microseisms and the seasonal whale calls. Median noise levels in this range vary by~20 dB between sites at a given depth but on average decrease with increasing depth. On the continental shelf, the orbital motions of ocean waves are a major source of noise while at the quietest sites in the deep ocean, noise increases when wind speeds exceed~10 m/s. On the continental slope and abyssal plain within about 100 km of the slope, seismic noise is not predicted at specific sites by the bottom currents in the ocean circulation model. In these regions, ocean currents are inferred to be the primary source of noise, because noise varies on tidal periods, is low on buried seismometers, and has spatial variations broadly consistent with those of median absolute currents. Comparisons between OBSs suggest that high-frequency noise is reduced by low-profile hydrodynamic designs but not by shielding. Many OBSs also record numerous short duration events on and near the continental shelf that have been attributed elsewhere to animals bumping into the sensor or gas bubbles moving through sediments. Plain Language Summary Environmental noise generated by waves, wind, and ocean currents limits the ability of seafloor seismometers to detect the ground motions from earthquakes. This research uses data from a large deployment of seafloor seismometers off the coast of the Pacific Northwest of America to understand the seismic noise at frequencies of 5-12 Hz which coincides with the frequency band used to detect regional earthquakes. Instruments deployed in shallow water on the continental shelf record high noise levels caused by motions from ocean waves that reach the seafloor. The quietest sites are found in the deep ocean far from the coast, and here the noise is created by wind blowing spray from whitecaps. At intermediate sites on the continental slope and deep ocean just seaward of the continental slope, the noise results from ocean currents that are related to ocean circulation and tides. Some shallow instruments also record large numbers of short signals that are not earthquakes and may be either a result of animals bumping into the seismic sensor or gas bubbles moving through sediments. The best way to reduce seismic noise is to bury the seismic sensors, but when this is infeasible, the instruments should have a short hydrodynamic design.

Variability of Natural Methane Bubble Release at Southern Hydrate Ridge

“…Apart from these high amplitude events, which contain most of the seismic signal's power, the seismic frequency spectra for all three axes (Northward, Eastward, upward) are dominated by the main tidal frequency constituents. Higher frequency constituents can be found in the domain of the ambient seismic noise (Hilmo & Wilcock, 2020).…”

Tidal Triggering of Microseismicity at the Equatorial Mid‐Atlantic Ridge, Inferred From the PI‐LAB Experiment

“…There is no significant preference at particular hours ( p s = 0.20), something that is rather expected in an oceanic environment in the absence of anthropogenic noise, predominantly present during daytime in continental areas. Note, however, that the fact that detection level may be correlated with the tides cannot be ruled out (e.g., Hilmo & Wilcock, 2020). Nevertheless, we estimated that M C fluctuated between −0.34 to −0.05 during different tidal phase ranges.…”

A method for tracking blue whales (Balaenoptera musculus) with a widely spaced network of ocean bottom seismometers

“…After normalizing the seismic records to a uniform sensitivity, continuous spectrograms were constructed for each station using a 2-s Hanning window with 80% overlap ( Fig 2A–2C ). Because the 50-Hz sampling frequency of the majority of the OBSs was insufficient to record the high-amplitude third harmonic of the B call and many of the instruments sampling at higher sampling rates were noisy at high frequencies [ 28 ], detections were based on just the first harmonic. The detection kernel ( Fig 2D ) was constructed from representative calls and comprises a single tone that sweeps down linearly from 15.7 Hz to 14.4 Hz over 10 s with a bandwidth of 0.5 Hz.…”