Distributed acoustic sensing (DAS) is a technology that turns a fiber-optic cable into an acoustic sensor by measuring the phase change of backscattered light caused by changes in strain from an acoustic field. In October 2022, 9 days of DAS and co-located hydrophone data were collected in the Puget Sound near Seattle, WA. Passive data were continuously recorded for the duration and a broadband source was fired from several locations and depths on the first and last days. This dataset provides comparisons between DAS and hydrophone measurements and demonstrates the ability of DAS to measure acoustics signals up to ∼700 Hz.

Acoustic monitoring is an important component of studying the wide variety of sounds and sound sources in the ocean. Applications of ocean acoustics range from general oceanography, studying of marine mammals, monitoring of natural and anthropogenic ocean noise, defense, ocean exploration, and more. Unfortunately, dense sampling of the acoustic field in large regions of the ocean can be impractical. Deployment of large hydrophone arrays is challenging—the hydrophones may be expensive and often require maintenance (which is particularly challenging for deeper regions of the ocean), and denser sampling is only achieved by deploying more hydrophones. Thus, alternatives that allow for increased coverage of the ocean at reduced costs are highly desirable.

Distributed fiber optic sensing (DFOS) is a class of techniques in which a fiber-optic cable, typically used for data transfer, acts as the sensor, capable of measuring temperature [distributed temperature sensing (DTS)], strain [distributed strain sensing (DSS)], or vibrations [distributed acoustic sensing (DAS)] (Bao and Chen, 2012). The use of fiber optic cables to measure acoustic waves (DAS) is a recent development in this class of measurement techniques. DAS utilizes Rayleigh backscattering of light from nano-scale defects in the fibers to measure changes to strain along the fiber, which can be indicative of acoustic waves (Hartog, 2017; Masoudi and Newson, 2016). An interrogator device attached to one end of the cable sends repeated laser pulses through the fiber, and as these waves interact with the imperfections in the fibers, phase changes in the scattered light over small sections of cable (the gauge length) allow spatially resolved measurement of strain or strain-rate. These strain and strain-rate measurements provide information about the average acoustic field over the chosen gauge length, sampled at regular intervals along the cable. The gauge length and sampling resolution are both parameters that can be varied depending on the application and limitations of the cable. The ranges over which the cables can be used to sense the acoustic field can be limited by several factors: the distance to the first fiber repeater (if applicable), the attenuation along the fiber resulting in a signal-to-noise ratio (SNR) that is too low, or the sampling rate (such that each light pulse has enough time to travel to the desired point along the cable, and the backscattered light to propagate back to the interrogator, prior to the next pulse). The range of acoustic frequencies that are detectable is still an active area of exploration, but is known to extend to at least several hundred Hertz (Taweesintananon , 2021; Lindsey and Martin, 2021).

DAS was first explored as a technique for seismic applications and has received significant attention in that community over the last decade. The first demonstration of DAS was in 2009, using the technology as a replacement for borehole geophones, with additional similar field trials in 2010 (Mestayer , 2011). These initial field trials demonstrated the ability to do seismic imaging with DAS and produce comparable results to geophone-produced images in terms of SNR and resolution. Additional demonstrations of DAS technology, capabilities, and applications followed over the next decade and it has become a significant area of research in seismology (Dou , 2017, Karrenbach , 2019; Zhan, 2020; Lindsey , 2019; Sladen , 2019; Daley ., 2013).

Recently, DAS has been explored as a means for measuring acoustic fields at frequencies above those typically relevant for seismic applications (>20 Hz). By extending DAS capabilities to higher frequencies, the technology can be utilized for measurements of acoustics in the water column. To date, only a few demonstrations of DAS for water column acoustics have been completed. DAS has shown the ability to produce seismic images comparable to those generated by a typical towed array method, particularly utilizing the lower spectral content of the seismic source and when the offset range did not exceed the channel depth (Taweesintananon , 2021; Matsumoto , 2021). Ship detection and tracking with signals up to 100 Hz has seen success in shallow and deep water channels, with deeper water performing better due to lower SNRs (Rivet , 2021). Finally, additional ship tracking and detection of Baleen whale calls below 100 Hz using a cable in the arctic (Bouffaut , 2022; Landrø , 2022) and ship noise and fin whale calls (<20 Hz) utilizing a cable from the Ocean Observatories Initiative (Wilcock , 2023) have been demonstrated.

The focus of this manuscript is on an active source experiment conducted in the Puget Sound near Seattle, WA. The goal for these data is twofold—to explore the capabilities of DAS at frequencies up to 1 kHz, and to provide hydrophone measurements taken close to the cable for ground truth acoustic field measurements and for future comparison and calibration of DAS measurements. The remainder of this manuscript provides a detailed overview of the experiment and a brief overview of some of the DAS measurements.

The DASCAL22 experiment took place in the Saratoga Passage region of the Puget Sound in Washington from October 19–28, 2022. In this experiment, three hydrophones were deployed adjacent to a fiber-optic DAS cable that lies on the seabed between Camano Island and Whidbey Island. The water depth varies from 0 m (at the entry points to the water), to a maximum of ∼100 m, with the majority of the cable lying between ∼80–90 m depth. The approximate location of the cable and the bathymetry along the cable are shown in Figs. 1(a) and 1(b), respectively. A mooring with the three hydrophones was deployed next to the cable at ∼93 m depth. Two hydrophones, an SQ26-H1B and a CR1A, were moored roughly 5 m from the sea floor, and a third hydrophone, another CR1A, was moored roughy 25 m from the sea floor (all hydrophones provided by Cetacean Research Technology, Seattle, WA). The hydrophones recorded ∼9 days of passive acoustic data with 44.1 kHz sampling rates. Hydrophone data discussed and presented in this manuscript will be in reference to the SQ26-H1B only. Figure 1(c) shows the noise floor obtained from 120 s of ambient noise at each channel along the submerged portion of the DAS cable, chosen during a period with no known boats and calm weather conditions. This figure demonstrates the variability in adjacent DAS channels that must be considered when examining DAS data. Some of the noise is likely due to equipment and much of the variability caused by unpredictable variations along the cable, as ambient acoustic noise would be expected to vary minimally between adjacent channels.

Fig. 1.

(a) The approximate locations of the DAS cable, mooring, and active source testing in the DASCAL22 experiment. (b) The bathymetry along the DAS cable with the mooring location indicated by a red “+,” approximate location of the DAS channels considered in this study indicated by a black “×,” and approximate location of the active source indicated by a blue circle (note that the source moved during the experiment, primarily in the out-of-page direction). (c) Noise floor of each DAS channel based on 2 min of acoustic data.

Fig. 1.

(a) The approximate locations of the DAS cable, mooring, and active source testing in the DASCAL22 experiment. (b) The bathymetry along the DAS cable with the mooring location indicated by a red “+,” approximate location of the DAS channels considered in this study indicated by a black “×,” and approximate location of the active source indicated by a blue circle (note that the source moved during the experiment, primarily in the out-of-page direction). (c) Noise floor of each DAS channel based on 2 min of acoustic data.

Close modal

The submerged portion of the DAS cable is just over 3.5 km long, the full length of which was sampled at a 2 kHz acoustic sampling rate. A gauge length of 6.38 m and spatial resolution of 6.38 m were used. Though the water entry and exit points are known, the positions on the seafloor are not perfectly known and are interpolated between these two points (there is no reason to suggest the cable was laid out in any way other than the direct path and thus far improved localization has not been possible for the majority of the cable). DAS channels are numbered with channel 0 beginning at the Whidbey Island entry point (recorded as channel 1140 in the dataset). DAS data were collected by Sintela Onyx v1.0 interrogator (Sintela Ltd., Pill, Bristol, Somerset, UK). Over the 9 days between the equipment deployment and recovery, the three moored hydrophones and DAS cable generated ∼14 TB of data and measured the acoustic field continuously, during which time there was significant boat traffic and a variety of weather, including windy and rainy conditions.

During the mooring deployment and recovery days, an acoustic source providing broadband impulsive signals was broadcast from three different depths (1, 5, and 10 m) at 5 s intervals from various locations near the mooring (with the boat's engine turned off during the broadcasts). For brevity, the results shown in this paper are from the data recorded on the recovery day (October 28, 2022). During this portion of the experiment, a weaker current and the crew's valuable experience from the first day of testing led to a cleaner set of data, improved logging, and better control of source positioning. The acoustic source used during the experiment was a bubble pulser designed for geophysical surveys. The source consists of two electromechanical plates that are drawn together by applying a voltage to the plates, reversing direction after impacting, thus producing short-duration impulsive signals. In all broadcasts, the endfire dimension (along the face of the plates) was aligned parallel to the boat. A small reference hydrophone, an HTI-96-min (High Tech, Inc., Long Beach, MS), was mounted 1 m from the center of the source to measure the signature from broadside and endfire orientations at a 2 kHz sampling rate, allowing for characterization of the source signal measured by the moored hydrophones and DAS cable.

The first consideration is whether the DAS cable is capable of detecting the active source and what the source signal looks like. Figure 2(a) provides a normalized spectral average of 10 shots (per curve) at 5 m depth, taken with the HTI hydrophone, at endfire and broadside orientations, as well as an ambient noise curve for reference, taken using windows of data just prior to the shot recordings. A 15 Hz highpass filter is applied to the measurements, as well as 1 Hz notch filters at 60, 180, 300, 420, 540, 660, 780, and 900 Hz, compensating for amplitude spikes caused by equipment noise at multiples of 60 Hz (this compensation is imperfect, and as a result, some sharp amplitude fluctuations are still visible at some of these frequencies). These plots show that the source provides a broadband signal with significant SNR across most of the frequency spectrum. Notably, the most significant amount of acoustic energy exists between ∼300–500 Hz in the endfire direction, and the broadside direction seems to provide a stronger signal at lower (<200 Hz) frequency ranges. Figure 2(b) shows a time domain measurement of three DAS channels with the source fired from three different depths and locations, confirming the capability of DAS to detect this source impulse. The Lloyd's mirror effect is noticeable across the three signals.

Fig. 2.

(a) Spectra of the bubble pulser measured by a hydrophone mounted approximately 1 m from the source center at endfire (blue) and broadside (orange) while at a 5 m depth. (b) Time domain measurements of the bubble pulser broadcasting from three depths (1, 5, and 10 m) at DAS channels 512, 516, and 519, respectively. (c) A waterfall plot for channels 455–550 (∼600 m of cable), highpass filtered above 1 Hz (time relative to 16:36:00 UTC), with each channel normalized by the maximum amplitude. (d) The same data as panel (c), zoomed in to channels 512–522 and filtered 300–500 Hz.

Fig. 2.

(a) Spectra of the bubble pulser measured by a hydrophone mounted approximately 1 m from the source center at endfire (blue) and broadside (orange) while at a 5 m depth. (b) Time domain measurements of the bubble pulser broadcasting from three depths (1, 5, and 10 m) at DAS channels 512, 516, and 519, respectively. (c) A waterfall plot for channels 455–550 (∼600 m of cable), highpass filtered above 1 Hz (time relative to 16:36:00 UTC), with each channel normalized by the maximum amplitude. (d) The same data as panel (c), zoomed in to channels 512–522 and filtered 300–500 Hz.

Close modal

Knowing that detection of this signal is possible, the usefulness of the extent of the cable is of interest. Figure 2(c) provides a waterfall plot of ∼600 m of the DAS cable on the Camano Island side, highpass filtered above 1 Hz and normalized by the maximum value on each channel, with several notable features. The vertical lines that appear intermittently are not due to the active source shots, but rather are low frequency events that are likely due to motion of the cable from coupling with waves and/or currents in the water. The lower and upper regions of the plot do not have such events (and thus, the normalization does not suppress the ambient noise; hence, the merging of lines). This difference most likely indicates that this portion of the cable on the sloped edge of the bathymetry is not buried under sediment, and thus, couples with waves and/or currents in the water column. The remaining ∼3 km of the cable is likely buried, as it behaves similarly (the full extent of the cable is not shown for brevity and clarity). In addition to the low frequency events, active source shots are present on the unburied channels, and also do not visibly extend beyond the narrow region of this cable, though they are difficult to see on this plot. Figure 2(d) shows a zoomed in version spanning channels 512–522 and filtered from 300–500 Hz, which clearly shows the shot patterns. Interestingly, as the source location changes, the effect of the arrival angle on the shot amplitude is evident, with shots appearing inconsistently across each channel. This is due to an effect causing the amplitude measured on each DAS channel to be directly proportional to co s 2  θ, with  θ =0° parallel to the cable (Rivet , 2021). This variability adds a challenge to processing the data effectively, as a single channel is only usable at specific times. Unfortunately, the buried regions of the cable are not showing evidence of recording the active source. Future analyses of other source types could provide valuable insights into the effects of a buried vs unburied cable.

While the DAS channels detecting the active source are limited, they can repeatedly detect the source firing in the right conditions. The longer duration acoustic field measurements for a hydrophone and several DAS channels (channel 516) are shown in Fig. 3. Simple BELLHOP (HLS Research, La Jolla, CA) simulations of the approximate source and receiver locations [panels (a) and (b)] demonstrate possible paths to expect, though this simulation is notably simplified—it assumes a constant sound speed profile and lacks sediment information—and is just intended to roughly validate the measured signal impulse responses. Figures 3(c) and 3(d) provide waterfall plots of 200 consecutive time windows for the DAS channel and hydrophone, spanning ∼15 min of shots, at source depths of 10 (shots 0–62), 5 (shots 68–122), and 1 m (shots 128–186), with the source turned off during depth changes (evident in the waterfall plots), and the boat slowly drifting with the current throughout the time window. During this time, the bathymetry at the source location varies from 66  (starting) to 43 m (ending). The DAS channel shown here clearly picks up a direct path signal corresponding to the bubble pulser. The source is expected to be very close to this channel; thus, a single strong arrival is expected. The DAS measurement shows a small time delay between the two arrivals, corresponding to the direct path and surface reflection (note the delay shrinking when the source depth is decreased), as well as a slightly changing arrival time as the boat drifts further from the cable channels. The hydrophone shows similar characteristics, with time delayed surface reflections and an impulse response slowly varying over time, as well as some multipath propagation. The first path corresponds to the direct path arrival, while the second is likely a path reflecting off of the sloped bathymetry, towards the hydrophone. The third path that arrives towards the top of the plot is likely due to a bathymetry change as the boat drifts, leading to a new, strong arrival.

Fig. 3.

BELLHOP simulations of possible propagation paths between (a) the source and a DAS channel and (b) the source and hydrophone. Waterfall plots of ∼200 shots of the bubble pulser for (c) DAS channel 512 and (d) hydrophone.

Fig. 3.

BELLHOP simulations of possible propagation paths between (a) the source and a DAS channel and (b) the source and hydrophone. Waterfall plots of ∼200 shots of the bubble pulser for (c) DAS channel 512 and (d) hydrophone.

Close modal

Figure 4 demonstrates the capability of DAS to record signals >100 Hz. Six consecutive shots from the active source, broadcast on the equipment recovery date (October 28, 2022), are plotted with data recorded by the hydrophone [Fig. 4(a)], showing a clear broadband signal with notable SNR. The output of a single DAS channel (channel 516) over the same period of time also clearly shows measurements of six pulses, with notably lower SNR despite being closer to the source, but still with a signal clearly visible between ∼300–450 Hz [Fig. 4(b)]. Because the channel output is a moving average with a period corresponding to the gauge length, the channel's response is proportional to sinc ( π f L g / c ), where f is frequency, L g is gauge length, and c is the speed of sound. This predicts frequency notches at ∼235 and 470 Hz (as well as 705 and 940 Hz)—corresponding very well to the frequency ranges seen in Fig. 4(b). The SNR for hydrophone (black circles) and DAS (gray triangles) data is calculated from a 60 ms shot sample and a 60 ms noise sample recorded 600 ms before the reception of the shot signal [Fig. 4(c)]. The SNR on the DAS output is significantly lower than that of the hydrophone, but the two other curves show that significant improvements to the DAS SNR can be made by combining outputs. Combining eight DAS channels (red diamonds) and combining eight shots on a single DAS channel (blue squares) both show significant SNR improvements, with the red curve performing better at higher frequencies and showing detection capabilities above 700 Hz. In general, channel stacking is more applicable, but shot stacking is included here due to the lack of available DAS channels as a means to emphasize potential capabilities of DAS (since these shots are repeatable, shot stacking should yield similar results to channel stacking in a case with improved and consistent channel outputs). The path length to the eight combined channels varies minimally, with the difference in arrival time from the source varying by a maximum of 7.5 ms.

Fig. 4.

(a) Six consecutive shots of the bubble pulser at 5 m below the surface measured by hydrophone A and (b) channel 516 of the DAS cable. (c) The SNR of the pulses as measured by hydrophone A (black circles), DAS channel 516 (gray triangles), DAS channel 516 with eight shots stacked (blue squares), and eight DAS channels stacked from the same shot (red diamonds).

Fig. 4.

(a) Six consecutive shots of the bubble pulser at 5 m below the surface measured by hydrophone A and (b) channel 516 of the DAS cable. (c) The SNR of the pulses as measured by hydrophone A (black circles), DAS channel 516 (gray triangles), DAS channel 516 with eight shots stacked (blue squares), and eight DAS channels stacked from the same shot (red diamonds).

Close modal

These results demonstrate the ability of DAS to record acoustic signals at frequencies >100 Hz, up to approximately 500 Hz without any signal combinations to improve SNR and up to 700 Hz with channel or shot combinations. While the hydrophone data clearly show superior SNR in this application, the purpose of these results is to understand DAS capabilities and the potential to utilize the advantages of DAS (i.e., its distributed nature and lower equipment costs per unit of data). These data provide multiple opportunities to explore and improve the understanding of DAS capabilities in ocean acoustics, such as improving SNR, applying standard array signal processing techniques to the data, extension of measurements up to 1 kHz, and consideration of other acoustic sources available in these measurements.

DAS technology is an exciting frontier in ocean acoustics, potentially providing the ability to continuously monitor large regions of the ocean with dense arrays. However, the technology is in its infancy and significant work exists to fully exploit its capabilities. This experimental dataset provides a significant step towards this goal, with higher frequency sampling rates than most existing DAS datasets, and providing co-located hydrophone and DAS measurements, allowing for ground truth data, direct comparisons of the two measurements, and calibration of signals recorded on DAS cables. An overview of the DASCAL22 experiment was provided and several conclusions from initial analysis of the data resulted.

First, it has been demonstrated that DAS technology is capable of detecting acoustic signals up to ∼700 Hz, and it is likely that the capabilities extend to higher frequencies if enough SNR is achieved. Second, it is shown that the SNR of DAS cables is significantly lower than that of traditional hydrophone recordings, which was expected, but that some SNR can be recovered with clever combinations of measurements in the densely sampled array. In these results, a difference of ∼5–15 dB SNR is seen, depending on whether the channels/shots are stacked or single measurements. However, the DAS channels used for this analysis are not co-located with the hydrophone, so these SNRs are both reported for reference but not meant to be a direct comparison of instrument capabilities. Third, there is evidence of some of the challenges related to DAS, particularly regarding gauge length dependent frequency responses and the effects of arrival angle on the cable. Further, the impact of what is hypothesized to be sediment cover has notable effects on the usability of the cable. Many questions still remain for future work within this dataset and DAS technology in general, including further investigation into the lack of signals measured in other regions of the cable and the effects of sediment cover and other environmental variables on DAS, a better understanding of DAS-specific noise sources, and analysis on other types of acoustic sources present during the experiment.

We thank the crew of the R/V Weelander in the School of Oceanography at the University of Washington. We also thank Dick Sylwester from the Northwest Geophysical Services for generously providing the bubble pulser used in this study. This work was funded through the Office of Naval Research Grant No. N00014-21-1-2727 and the lease of the DAS cable was in collaboration with Whidbey Telecommunications and was funded under National Science Foundation Grant No. OCE-2211274.

1.
Bao
,
X.
, and
Chen
,
L.
(
2012
). “
Recent progress in distributed fiber optic sensors
,”
Sensors
12
,
8601
8639
.
2.
Bouffaut
,
L.
,
Taweesintananon
,
K.
,
Kriesell
,
H. J.
,
Rørstadbotnen
,
R. A.
,
Potter
,
J. R.
,
Landrø
,
M.
,
Johansen
,
S. E.
,
Brenne
,
J. K.
,
Haukanes
,
A.
,
Schjelderup
,
O.
, and
Storvik
,
F.
(
2022
). “
Eavesdropping at the speed of light: Distributed acoustic sensing of baleen whales in the arctic
,”
Front. Mar. Sci.
9
,
901348
.
3.
Daley
,
T. M.
,
Freifeld
,
B. M.
,
Ajo-Franklin
,
J.
,
Dou
,
S.
,
Pevzner
,
R.
,
Shulakova
,
V.
,
Kashikar
,
S.
,
Miller
,
D. E.
,
Goetz
,
J.
,
Henninges
,
J.
, and
Lueth
,
S.
(
2013
). “
Field testing of fiber-optic distributed acoustic sensing (DAS) for subsurface seismic monitoring
,”
Leading Edge
32
,
699
706
.
4.
Dou
,
S.
,
Lindsey
,
N.
,
Wagner
,
A. M.
,
Daley
,
T. M.
,
Freifeld
,
B.
,
Robertson
,
M.
,
Peterson
,
J.
,
Ulrich
,
C.
,
Martin
,
E. R.
, and
Ajo-Franklin
,
J. B.
(
2017
). “
Distributed acoustic sensing for seismic monitoring of the near surface: A traffic-noise interferometry case study
,”
Sci. Rep.
7
,
11620
.
5.
Hartog
,
A. H.
(
2017
).
An Introduction to Distributed Optical Fibre Sensors
(
CRC Press
,
Boca Raton, FL
), p.
472
.
6.
Karrenbach
,
M.
,
Cole
,
S.
,
Ridge
,
A.
,
Boone
,
K.
,
Kahn
,
D.
,
Rich
,
J.
,
Silver
,
J.
, and
Langton
,
D.
(
2019
). “
Fiber-optic distributed acoustic sensing of microseismicity, strain and temperature during hydraulic fracturing
,”
Geophysics
84
,
D11
D23
.
7.
Landrø
,
M.
,
Bouffaut
,
L.
,
Kriesell
,
H. J.
,
Potter
,
J. R.
,
Rørstadbotnen
,
R. A.
,
Taweesintananon
,
K.
,
Johansen
,
S. E.
,
Brenne
,
J. K.
,
Haukanes
,
A.
,
Schjelderup
,
O.
, and
Storvik
,
F.
(
2022
). “
Sensing whales, storms, ships and earthquakes using an Arctic fibre optic cable
,”
Sci. Rep.
12
,
19226
.
8.
Lindsey
,
N. J.
,
Dawe
,
T. C.
, and
Ajo-Franklin
,
J. B.
(
2019
). “
Illuminating seafloor faults and ocean dynamics with dark fiber distributed acoustic sensing
,”
Science
366
(
6469
),
1103
1107
.
9.
Lindsey
,
N. J.
, and
Martin
,
E. R.
(
2021
). “
Fiber-optic seismology
,”
Annu. Rev. Earth Planet. Sci.
49
,
309
336
.
10.
Masoudi
,
A.
, and
Newson
,
T.
(
2016
). “
Contributed review: Distributed optical fibre dynamic strain sensing
,”
Rev. Sci. Instrum.
87
,
011501
.
11.
Matsumoto
,
H.
,
Araki
,
E.
,
Kimura
,
T.
,
Fujie
,
G.
,
Shiraishi
,
K.
,
Tonegawa
,
T.
,
Obana
,
K.
,
Arai
,
R.
,
Kaiho
,
Y.
,
Nakamura
,
Y.
,
Yokobiki
,
T.
,
Kodaira
,
S.
,
Takahashi
,
N.
,
Ellwood
,
R.
,
Yartsev
,
V.
, and
Karrenbach
,
M.
(
2021
). “
Detection of hydroacoustic signals on a fiber-optic submarine cable
,”
Sci. Rep.
11
,
2797
.
12.
Mestayer
,
J.
,
Cox
,
B.
,
Wills
,
P.
,
Kiyaschenko
,
D.
,
Lopez
,
J.
,
Costello
,
M.
,
Bourne
,
S.
,
Ugueto
,
G.
,
Lupton
,
R.
,
Solano
,
G.
,
Hill
,
D.
, and
Lewis
,
A.
(
2011
). “
Field trials of distributed acoustic sensing for geophysical monitoring
,” in
Proceedings for the SEG San Antonio 2011 Annual Meeting
, pp.
4253
4257
.
13.
Rivet
,
D.
,
de Cacqueray
,
B.
,
Sladen
,
A.
,
Roques
,
A.
, and
Calbris
,
G.
(
2021
). “
Preliminary assessment of ship detection and trajectory evaluation using distributed acoustic sensing on an optical fiber telecom cable
,”
J. Acoust. Soc. Am.
149
,
2615
2627
.
14.
Sladen
,
A.
,
Rivet
,
D.
,
Ampuero
,
J. P.
,
De Barros
,
L.
,
Hello
,
Y.
,
Calbris
,
G.
, and
Lamare
,
P.
(
2019
). “
Distributed sensing of earthquakes and ocean-solid Earth interactions on seafloor telecom cables
,”
Nat. Commun.
10
(
1
),
5777
.
15.
Taweesintananon
,
K.
,
Landrø
,
M.
,
Brenne
,
J. K.
, and
Haukanes
,
A.
(
2021
). “
Distributed acoustic sensing for near-surface imaging using submarine telecommunication cable: A case study in the Trondheimsfjord, Norway
,”
Geophysics
86
,
B303
B320
.
16.
Wilcock
,
W. S. D.
,
Abadi
,
S.
, and
Lipovsky
,
B.
(
2023
). “
Distributed acoustic sensing recordings of low-frequency whale calls and ship noise offshore Central Oregon
,”
JASA Express Lett.
3
(
2
),
026002
.
17.
Zhan
,
Z.
(
2020
). “
Distributed acoustic sensing turns fiber-optic cables into sensitive seismic antennas
,”
Seismol. Res. Lett.
91
,
1
15
.