SM2.1

Once a niche technology, Distributed Acoustic Sensing (DAS) has gained increasing popularity over the last decade, due to its versatility and ability to capture extremely dense seismic datasets in a wide range of challenging environments. While DAS has been utilised in some particularly remote locations, such as on glaciers and volcanoes, it also holds a great deal of potential closer to home; beneath our cities. As DAS is able to be used with existing telecommunication fibres, urban areas contain huge potential networks of strain or strain-rate sensors, right beneath our feet. This data enables us to monitor the local environment, recording events such as earthquakes, as well as characterising and monitoring the shallow subsurface. DAS experiments using dark fibres are unintrusive and highly repeatable, meaning that this method is ideal for long-term site monitoring.

In collaboration with the OTE Group (the largest telecommunications company in Greece), we have collected urban DAS data beneath North-East Athens, utilising existing, in-situ telecommunication fibres. This large dataset contains a wide range of anthropogenic signals, as well as many seismic events, ranging from small, local events, to an internationally reported Magnitude 6.4 earthquake in Crete.

We conduct a preliminary analysis of the dataset, identifying and assessing the earthquake signals recorded. This will be compared with the event catalogue of the local, regional network in Athens, to determine our sensitivity to events of different magnitudes, and in a range of locations. We hope to gain an understanding of how DAS could be combined with the existing network for seismic monitoring and earthquake detection.

Moving forward, we aim to also apply ambient noise methods to this dataset in order to extract dispersion measurements, and ultimately invert for a shallow velocity model of the suburbs of Athens.

How to cite: Smolinski, K. T., Bowden, D. C., Lentas, K., Melis, N. S., Simos, C., Bogris, A., Simos, I., Nikas, T., and Fichtner, A.: Distributed Acoustic Sensing in the Athens Metropolitan Area: Preliminary Results, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-11864, https://doi.org/10.5194/egusphere-egu22-11864, 2022.

Distributed Acoustic Sensing (DAS) is a rapidly evolving technology that can turn a fibre optic cable into thousands of acoustic sensors. In this study, we propose to present a seismic survey conducted as a business showcase relying on a collaborative work supported by five partners: FEBUS Optics, RealTimeSeismic (RTS), Gallego Technic Geophysics (GTG), Petro LS and Well-SENSE. The project was carried out at a deep solution mining site developed for salt production, operated by KEMONE, and located nearby Montpellier (South of France).

The seismic campaign was based on two different cable deployments.

On the first hand, a Vertical Seismic Profile survey was conducted on borehole seismic measurements using two different fibre optic cables deployed in a 1800m deep vertical well. The first set of tests was performed along a Petro LS wireline cable including optical fibres. This deployment corresponds to a conventional wireline operation. The second set of data has been acquired along a FibreLine Intervention system (FLI) developed by WellSENSE. The deployment of the FLI system relies on the unspooling a bare optical fibre using a probe along a wellbore. This solution is single-use and sacrificial and can be left in the well at the end of the survey.

On another hand, a short 400m-surface 2D profile has been achieved along both a fibre optic cable and a set of STRYDE nodes deployed by GTG.

Fibre optic cables have been connected to FEBUS DAS interrogator to collect distributed acoustic measurements.  The seismic tests, performed in collaboration with GTG, have been achieved with basic “weight drops” (1T falling from 4m) for the checkshot surveys and with an "IVI Mark 4" 44,000-pound seismic vibrator for VSP shots at offset from wellhead reaching 865m. Acquired data have been analysed by RTS.

This work will describe the survey, present the results, and discuss the learnings in two ways:  the optimisation of acquisition setups and processing parameters to obtain the best exploitable results and seismic surveys perspectives and challenges using DAS technology.

How to cite: Jestin, C., Maisons, C., Chérubini, A., Duboeuf, L., and Puech, J.-C.: A showcase pilot of seismic campaign using Distributed Acoustic Sensing solutions, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-5551, https://doi.org/10.5194/egusphere-egu22-5551, 2022.

Distributed Acoustic Sensing (DAS) systems have gained popularity in recent years due to the dense spatial coverage of strain observations; with one fiber and one interrogator researchers can have access to thousands of strain or strain-rate observations over a region. DAS systems have a limited range, however, with usual experiments being on the order of 10’s of kilometers, owing to their reliance on weakly backscattered light. In contrast, systems that transmit light through a fiber and measure signals on the other end (or looped back) can traverse significantly longer distances (e.g., Marra et. al 2018, Zhan et. al 2021, Bogris et. al 2021), and have the added advantages of being potentially cheaper and potentially operating in parallel with active telecommunications purposes. The disadvantage of such transmission systems is that only a single measurement of strain along the entire distance is given.

During September - October 2021, we operated examples of both systems side-by-side using telecommunications fibers underneath North Athens, Greece, in collaboration with the OTE telecommunications provider. Several earthquakes were detected by both systems, and we compare observations from both. The DAS system is a Silixa iDAS Interrogator measuring strain-rate. The newly designed transmission system relies on interferometric use of microwave frequency dissemination; signals sent along the fiber and back in a closed loop are compared to what was sent to measure phase differences (Bogris et. al 2021). We find that both systems are successful in sensing earthquakes and agree remarkably well when DAS signals are integrated over the length of the cable to properly mimic the transmission observations.

The direct transmission system, however, may not be as intuitive to interpret as an integral of displacement ground motions along the fiber. We discuss both theoretical and data-driven examples of how the observed phases depend on the curvature of a given length of fiber, and describe how asymmetries in the fiber’s index of refraction play a role in producing observed signals. Such an understanding is crucial if one is to properly interpret the signals from such a system (e.g., especially very long trans-oceanic cables). Given a full theoretical framework, we also discuss a strategy for seismic tomography given such a system: with a very long fiber, the spatial sensitivity should evolve over time as seismic signals reach different sections.

How to cite: Bowden, D., Fichtner, A., Nikas, T., Bogris, A., Lentas, K., Simos, C., Smolinski, K., Simos, I., and Melis, N.: Comparing two fiber-optic sensing systems: Distributed Acoustic Sensing and Direct Transmission, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-11599, https://doi.org/10.5194/egusphere-egu22-11599, 2022.

DAS (Distributed Acoustic Sensing) turns fibre optic cables used for telecommunications into multi-sensor antenna arrays. This technology makes it possible to detect an acoustic signal from a natural source such as cetacean or micro-earthquakes, or a man-made source by measuring the deformation of the cable. At sea, the coupling between the optical fibre and the ground on which it rests, as well as the calibration of the cable, is a critical point when the configuration of the cable is unknown. Is the fibre buried or suspended? What is the depth of the sensor being studied? What impact do these parameters have on the signal? The answers to these questions have an impact on the quality of the results obtained, the location of sources - seismic or acoustic - and the characterisation of the amplitude of signals are examples of this. Here, a first approach to study this calibration is proposed. Acoustic noise generated by passing ships in the vicinity of a 42km long optical fibre off Toulon, south-east France, is used to obtain signals for which the position and the signal of the source are known. Then, the synthetic and theoretical signal representing the ship's passage is modelled (3D model, AIS Long/Lat coordinates and depth, propagation speed in water c₀ = 1530m/s). This simulation allows us to compare the real and synthetic signals, in order to make assumptions about the actual cable configuration. We compare the signals through beamforming, f-k diagram and time-frequency diagram in particular. The grid search approach allowed us to determine the new position or orientation of a portion of the antenna. This new position is then evaluated from the signals of different vessels.

How to cite: Papotto, L., DeCacqueray, B., and Rivet, D.: Calibration and repositioning of an optical fibre cable from acoustic noise obtained by DAS technology, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-7311, https://doi.org/10.5194/egusphere-egu22-7311, 2022.

Distributed Fiber Optic Systems (DFOSs) refer to an ensemble of innovative technology that turns an optical fiber into a vast network of hundreds to thousands equally spaced sensors. According to the nature of the sensor, one can be sensitive to acoustic vibration (Distributed Acoustic Sensing, DAS) or to strain and temperature variation (Distributed Temperature and Strain Sensing, DTSS). DAS systems are well suited to detect short-term events in contrast to DTSS systems, which are intended to prevent long-term events. A combination of these two systems appears to be a good way to prevent against most possible events that can appear along an infrastructure. Furthermore, DFOS systems allow the interrogation of long profiles with very dense spatial and temporal sampling. Handling such amounts of data then appears as a challenge when trying to identify a threat along the structure. Machine learning solutions then proves their relevance for robust, fast and efficient acoustical event classification.

The goal of our study is to develop a method to handle, in real time, acquired data from these two DFOSs, classify them according to the nature of their origin and trigger an alarm if required. We mainly focus on major threats that jeopardize the integrity of pipelines. Our database contains leaks, landslides, and third-party intrusion (footsteps, excavations, drillings, etc.) simulated and measured at FEBUS Optics test bench in South-West France. Water and air leaks were simulated using electrovalves of several diameters (1mm, 3mm and 5mm), and landslides with a plate whose inclination was controlled by 4 cylinders. These data were acquired under controlled conditions in a small-scale model of pipeline (around 20m long) along different fiber optic cables installed along the structure.

Our method relies on several tools. A Machine Learning algorithm called Random Forest is used to pre-classify the DAS signal. Our implementation of this algorithm works in flow for a real time event identification. For DTSS signal, a simple threshold is used to detect if a strain or temperature variation occurs. Both results are then gathered and analyzed using a decisional table to return a classification result. According to the potential threat represented by its identified class, the event is considered as dangerous or not. Using this method, we obtain good results with a good classification rate (threat/non-threat) of 93%, compared to 87% if the DAS is used without the DTSS. We have noticed that the combination of both devices enables a better classification, especially for landslides hard to detect with the DAS. This combination enables to dramatically reduce the part of undetected threats from 16% to 4%.

How to cite: Huynh, C., Jestin, C., Hibert, C., Malet, J.-P., Lanticq, V., and Clément, P.: A real-time classification method for pipeline monitoring combining Distributed Acoustic Sensing and Distributed Temperature and Strain Sensing, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-4963, https://doi.org/10.5194/egusphere-egu22-4963, 2022.

In a surface-seismic setting, Distributed Acoustic Sensing (DAS) is still not a widely adopted method for near-surface characterisation, especially for reflection seismic imaging, despite the dense spatial sampling it provides over long distances. This is mainly due to the decreased broadside sensitivity that DAS suffers from when buried horizontally in the ground (that is when the upgoing wavefield (e.g. reflected wavefield) is perpendicular to the optical fibre). This is unlike borehole settings (e.g. zero-offset Vertical Seismic Profiling), where DAS has been widely adopted for many monitoring applications. Advancements in the field, like shaping the fibre to a helix, commonly known as helically wound fibre, allow better sensitivity for the reflections.

The promise of spatially dense seismic data over long distances is an attractive prospect for retrieving the local variations of near-surface properties. This is particularly valuable for areas impacted by induced seismicity, as is the case in the Groningen Province in the north of The Netherlands,  where near-surface properties, mostly composed of clays and peats, play an essential role on the amount of damage on the very near-surface and the structures built on it. Installing fibre-optic cables for passive and active measurements is valuable in this situation. We installed multiple cables containing different fibre configurations of straight and helically wound fibres, buried in a 2-m deep trench. The combination of the different fibre configurations allows us to obtain multi-component information. We observe differences in the amplitude and phase information, suggesting that these differences can be used for separating the different components of the wave motion. We also see that using enhanced backscatter fibres, reflection images can be obtained for the helically wound fibre as well as the straight fibre, despite the decreased broadside sensitivity for the latter.

How to cite: Al Hasani, M., Drijkoningen, G., and Wapenaar, K.: On the Multi-component Information of DAS for Near-Surface Seismic: A Pilot Field Experiment in the Groningen Area, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-2563, https://doi.org/10.5194/egusphere-egu22-2563, 2022.

On July 2, 2021, around 22:44 CET, a meteoroid was observed crossing the sky near Lake Thingvallavatn east of Reykjavik in Iceland. During this event, a large-N seismic network consisting of 500, 3-component geophones was monitoring local seismicity associated with the Hengill geothermal field southwest of the lake.  In addition to the large-N network, two fiber optic telecommunication cables, covering a total length of more than 40 km, were connected to distributed acoustic sensing (DAS) interrogation units. The systems were deployed under the frame of the DEEPEN collaboration project between the Eidgenössische Technische Hochschule Zürich (ETHZ), the German Research Centre for Geosciences (GFZ), NORSAR, and Iceland Geo-survey (ISOR). Both the large-N and the DAS recordings display multiple trains of infrasound arrivals from the meteoroid that coupled to the surface of the earth at the locations of the sensors. In particular, three strong phases and multiple other weaker arrivals can be identified in the DAS data.

Fitting each of the strong phases assuming point sources (i.e., fragmentations) generates travel time residuals on the order of several seconds, resulting in an unsatisfactory explanation of the observations. On the other hand, inverting the arrival times for three independent hypersonic-trajectories generating Mach cone waves reduces travel time residuals to well under 0.5 s for each arrival. However, whereas the 1st arrival is well constrained by more than 900 travel times from the large-N, DAS and additional seismic stations distributed over the Reykjanes peninsula, the 2nd and 3rd arrivals are mainly constrained by DAS observations near Lake Thingvallavatn. The less well-constrained, latter trajectories show a weak agreement with the trajectory of the first arrival. Currently, neither the multi-Mach-cone model nor the multi-fragmentation model explain all our observations satisfactorily. Thus, traditional models for interpreting meteoroid observations appear unsuitable to explain the combination of phase arrivals in the large-N network and DAS data consistently.

How to cite: Vera Rodriguez, I., Dahm, T., Isken, M. P., Kraft, T., Lamb, O. D., Wu, S.-M., Heimann, S., Sanchez-Pastor, P., Wollin, C., Baird, A. F., Wüstefeld, A., Kristjánsdóttir, S., Jónsdóttir, K., Eibl, E. P. S., Goertz-Allmann, B. P., Jousset, P., Oye, V., and Obermann, A.: Multiphase observations of a meteoroid in Iceland recorded over 40 km of telecommunications cables and a large-N network, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-7203, https://doi.org/10.5194/egusphere-egu22-7203, 2022.

We present and evaluate array processing techniques and algorithms for the characterization of snow avalanche signals recorded with Distributed Acoustic Sensing (DAS).

Avalanche observations rely on comprehensive measurements of sudden and rapid snow mass movement that is hard to predict. Conventional avalanche sensors are limited to observations on or above the surface. Recently, seismic sensors have increased in their popularity for avalanche monitoring and characterization due to their avalanche detection and characterization capabilities. To date, however, seismic instrumentation in avalanche terrain is sparse, thereby limiting the spatial resolution significantly.

As an addition to conventional seismic instrumentation, we propose DAS to measure avalanche-induced ground motion. DAS is a technology using backscattered light along a fiber-optic cable to measure strain (-rate) along the fiber with unprecedented spatial and temporal resolution - in our example with 2 m spatial sampling and a sampling rate of 1kHz.

We analyze DAS data recorded during winter 2020/2021 at the Valleé de la Sionne avalanche test site in the Swiss Alps, utilizing an existing 700 m long fiber-optic cable. Our observations include avalanches propagating on top of the buried cable, delivering near-field observations of avalanche-ground interactions. After analyzing the properties of near-field avalanche DAS recordings, we discuss and evaluate algorithms for (1) automatic avalanche detection, (2) avalanche surge propagation speed evaluation and (3) subsurface property estimation.

Our analysis highlights the complexity of near-field DAS data, as well as the suitability of DAS-based monitoring of avalanches and other hazardous granular flows. Moreover, the clear detectability of avalanche signals using existing fiber-optic infrastructure of telecommunication networks opens the opportunity for unrivalled warning capabilities in Alpine environments.

How to cite: Paitz, P., Edme, P., Fichtner, A., Lindner, N., Sovilla, B., and Walter, F.: Near-field observations of snow-avalanches propagating over a fiber-optic array, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-4014, https://doi.org/10.5194/egusphere-egu22-4014, 2022.

Volcanic explosions produce energy that propagates both in the subsurface as seismic waves and in the atmosphere as acoustic waves. We analyse thousands of explosions which occurred at different craters at Etna volcano (Italy) in 2018 and 2019. We recorded signals from infrasound sensors, geophones (GPH), broadband seismometers (BB) and Distributed Acoustic Sensing (DAS) with fibre optic cable. The instruments were deployed at Piano delle Concazze at about 2 to 2.5 km from the active craters, within (or onto) a ~300,000 m² scoria layer deposited by recent volcanic eruptions. The DAS interrogator was setup inside the Pizzi Deneri Volcanic Observatory (~2800 m elevation). Infrasonic explosion records span over a large range of pressure amplitudes with the largest one reaching 130 Pa (peak to peak), with an energy of ca. 2.5x10¹¹ J. In the DAS and the BB records, we find a 4-s long seismic “low frequency” signal (1-2 Hz) corresponding to the seismic waves, followed by a 2-s long “high-frequency” signal (16-21 Hz), induced by the infrasound pressure pulse. The infrasound sensors contain a 1-2 Hz infrasound pulse, but surprisingly no high frequency signal. At locations where the scoria layer is very thin or even non-existent, this high frequency signal is absent from both DAS strain-rate records and BB/GPH velocity seismograms. These observations suggest that the scoria layer is excited by the infrasound pressure pulse, leading to the resonance of lose material above more competent substratum. We relate the high frequency resonance to the layer thickness. Multichannel Analysis of Surface Wave from jumps performed along the fibre optic cable provide the structure of the subsurface, and confirm thicknesses derived from the explosion analysis. As not all captured explosions led to the observation of these high frequency resonance, we systematically analyze the amplitudes of the incident pressure wave versus the recorded strain and find a non-linear relationship between the two. This non-linear behaviour is likely to be found at other explosive volcanoes. Furthermore, our observations suggest it might also be triggered by other atmospheric pressure sources, like thunderstorms. This analysis can lead to a better understanding of acoustic-to-seismic ground coupling and near-surface rock response from natural, but also anthropogenic sources, such as fireworks and gas explosions.

How to cite: Jousset, P., Costes, L., Currenti, G., Schwarz, B., Napoli, R., Diaz, S., and Krawczyk, C.: Non-linear ground response triggered by volcanic explosions at Etna Volcano, Italy, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-4478, https://doi.org/10.5194/egusphere-egu22-4478, 2022.

Stromboli is an open-conduit volcano characterized by mild intermittent explosive activity that produces jets of gas and incandescent blocks. Explosions occur at a typical rate of 3-10 events per hour, VLP signals have dominant periods between 2 and 30 seconds. Seismic activity is also characterized by less energy short-period volcanic tremor related to the continuous out-bursting of small gas bubbles in the upper part of the magmatic column. The high rate of activity as well as the broadband frequency contents of emitted signals make Stromboli volcano an ideal site for testing new techniques of fibre-optic sensing.

In September 2020, approximately 1 km of fiber-optic cable was deployed on the Northeast flank of Stromboli volcano, together with several seismometers, to record the seismic signals radiated by the persistent Strombolian activity via both DAS and inertial-seismometers, and to compare their records.

The cable was buried manually about 30 cm deep over a relatively linear path at first and in a triangle-shaped array with 30-meters-long sides in the highest part of the deployment. The strain rate was recorded using a DAS interrogator Febus A1-R with a sampling frequency of 2000 Hz, a spatial interval of 2.4 m and a gauge length of 5m. Data were re-sampled at 200 Hz. A network of 22 nodes SmartSolo IGU-16HR 3C geophones (5 Hz) has been distributed over the fibre path. A Guralp digitizer equipped with a CMG CMG-40T 30 sec seismometer and an infrasound sensor were placed in the upper part of the path. The geolocation of the cable was obtained by performing kinematic GPS measurements with 2 Leica GR25 receivers. All equipment recorded simultaneously several hundreds of explosion quakes between September 20 and 23.

Data analysis provided the following main results:

• DAS interrogator clearly recorded the numerous explosion-quakes which occurred during the experiment, as well as lower amplitude tremor and LP events.
• DAS spectrum exhibits a lower resolution at long periods with a cut-off frequency of approximately 3 Hz.
• VLP seismic events generated by Strombolian activity are identified only at a few DAS channels belonging to a specific portion of the path, which seems affected by local amplification. At these channels, they display waveforms similar to those sensed by the Güralp CMG-40T.
• Comparison of DAS strain waveform to particle velocity recorded by co-located seismometers shows a perfect match in phase and a good agreement in amplitude.
• Beamforming methods have been applied to nodes data located on the upper triangle and to strain rate data, both in the 3-5 Hz frequency band. Slightly different back-azimuths were obtained, values estimated via DAS point more to the southwest with respect to the crater area. Apparent velocities obtained with DAS recordings have lower values compared to those obtained with nodes.

How to cite: Metaxian, J.-P., Biagioli, F., Ripepe, M., Stutzmann, E., Bernard, P., Longo, R., Bouin, M.-P., and Caudron, C.: Strombolian seismic activity characterisation using fibre-optic cable and distributed acoustic sensing, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-5952, https://doi.org/10.5194/egusphere-egu22-5952, 2022.

Seismic monitoring refers to the measurement of time-lapse changes of seismic wave velocities and is a frequently used technique to detect dynamic changes in the Earth‘s crust. Its applications include a broad range of topics, such as natural hazard assessment and structural health monitoring. To obtain reliable measurements, results are usually stacked over time. Thereby, temporal resolution is lost, which makes the measurement less sensitive to short-term environmental processes. Another problem is that conventional datasets often lack spatial density and velocity changes can only be attributed to large areas. Recently, distributed acoustic sensing (DAS) has gained a lot of attention as a way to achieve high spatial resolution at low cost. DAS is based on Rayleigh-scattering of photons within an optical fibre. Because measurements can be taken every few meters along the cable, the fibre is turned into a large seismic array that provides information about the Earth’s crust at unprecedented resolution.

In our study, we explore the potential of DAS for monitoring studies. Specifically, we investigate how spatial stacking of DAS traces affects the measurements of velocity variations. We use data recorded by a 21-km-long dark fibre located on Reykjanes Pensinsula, Iceland. The cable is sampled with a channel spacing of 4 meters. We analyze the energy of the oceans microseism continuously recorded between March and September 2020. At first, we stack adjacent traces on the fibre in space. We then cross correlate the stacks to obtain approximations of the Green’s functions between different DAS-channels. By measuring changes in the coda waveform of the extracted seismograms, velocity variations can be inferred. Our analysis shows that spatial stacking improves the reliability of our measurements considerably. Because of that, less temporal stacking is required and the time resolution of our measurements can be increased. In addition, the enhancement of the data quality helps resolve velocity variations in space, allowing us to observe variations propagating along the cable over time. These velocity changes are likely linked to magmatic intrusions associated with a series of repeated uplifts on the Peninsula. Our results highlight the potential of DAS for improving the localization capabilities and accuracy of seismic monitoring studies.

How to cite: Maaß, R., Schippkus, S., Hadziioannou, C., Schwarz, B., Krawczyk, C., and Jousset, P.: Overcoming limitations of seismic monitoring using fibre-optic distributed acoustic sensing, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-11311, https://doi.org/10.5194/egusphere-egu22-11311, 2022.

Slow slip plays an important role in accommodating plate motion along plate boundaries throughout the world. Further understanding of the interplay between aseismic and seismic slip has gained particular attention as it is crucial for the assessment of seismic risk. A wide range of instruments and acquisition techniques exist to quantify tectonic deformation which spans multiple orders of magnitude in duration as well as spatial extend. For example, seismometers acquire dense temporal data, however are sparsely deployed, leading to spatial aliasing. As opposite, remote sensing techniques have wide aperture but rather crude temporal resolution and accuracy (mm-range). In selected areas, strain is continuously measured with laser or borehole strainmeters.
In this contribution, we investigate the distribution of permanent strain along a telecommunication optic fibre on the Reykjanes Peninsula, South West Iceland. Continuous strain-rate was recorded via DAS (Distributed Acoustic Sensing) over a period of six months during the recent unrest of the Svartsengi volcano which began in January 2020. The interrogated fibre connects the town of Gridavik with the Svartsengi geothermal power plant and was patched to a second fibre leading to the western most tip of the Reykjanes Peninsula. It is approximately between 10 and 20km west of the active volcanic area which produced abundant local seismicity as well as surface uplift and subsidence in areas crossed with the optical fiber. The fibre was installed in a trench at less than one meter depth and consists of two roughly straight segments of 7 and 14km length. Whereas the longer segment trends WSW parallel to the strike of the Mid-Atlantic Ridge at this geographic height, the shorter segment trends NEN and thus almost coincides with the maximum compressive stress axis of the region.
Inspection of the spatio-temporal strain-rate records after the occurrence of local earthquakes indicates the accumulation of compressive as well as extensive strain in short fibre sections of a few dozen meters which could correlate with local geologic features like faults or dykes. This holds for events of M~2.5 and fibre segments in epicentral distances of more than 20km. Preliminary results regarding the total deformation of the fibre as response to an individual seismic event show a distinct behaviour for differently oriented fibre segments correlating with the overall stress regime, i.e. shortening in the order of some dozen nanometers in the direction of SHmax. Unfortunately, recordings of the two largest intermediate M>=4.8 events indicate saturation of the recording system or loss of ground coupling thus preventing a meaningful interpretation of their effect on permanent surface motion. 
Perspectively, our efforts aim at investigating the feasibility of distributed optical strain-rate measurements along telecommunication infrastructure to track locally accumulated strain.

How to cite: Wollin, C., Jousset, P., Reinsch, T., Lipus, M., and Krawczyk, C.: Strain accumulation along a 21km long optic fibre during a seismic crisis in Iceland, 2020, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-10322, https://doi.org/10.5194/egusphere-egu22-10322, 2022.

In most subduction zones, a great portion of seismicity is located offshore, away from permanent onland seismic networks. Chile is not the exception; since the upgraded seismic observation system began operating in 2013, 35% of the ~7000 earthquakes with M≥3 recorded yearly were located offshore. Most importantly, the epicenters of the largest earthquakes (M>7.5) from 2014 to 2016 were located offshore as well.

The Chilean national seismic network is mainly composed of coastal and inland stations, except for two stations located on oceanic islands, Rapa Nui (Easter Island) and Juan Fernandez archipelago. This station configuration makes it difficult to observe in sufficient detail the lower-magnitude seismicity at the nucleation points of large events. Moreover, the lack of seafloor stations limits the efficiency of earthquake early warning systems during offshore events. These challenges could be overcome by permanently instrumenting existing submarine telecom cables with Distributed Acoustic Sensing (DAS).

Thanks to GTD, a private telecommunications company that owns a 3500-km-long network of marine fiber optic cables with twelve landing points in Chile (Prat project), from Arica (~ 18⁰S) to Puerto Montt (~ 41⁰S), we conducted the POST (Submarine Earthquake Observation Project in Spanish) DAS experiment on the northern leg of the Concón landing site of the Prat cable. This experiment, one of the first to be conducted on a commercial undersea infrastructure in a very seismically active region, was carried out from October 28 to December 3, 2021. Based on the longitudinal strain-rate data measured along 150 km of cable with a spatial resolution of 4 meters and a temporal sampling of 125 Hz, we present preliminary results of analyses to assess the possibility of building a new type of permanent, real-time and distributed seafloor observatory for continuous monitoring of active faults and earthquake early warning systems.

How to cite: Rivet, D., Barrientos, S., Sánchez-Olavarría, R., Ampuero, J.-P., Lior, I., Bustamente Prado, J.-A., and Villarroel Opazo, G.-A.: Building a new type of seafloor observatory on submarine telecom fiber optic cables in Chile, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-11508, https://doi.org/10.5194/egusphere-egu22-11508, 2022.

Distributed Acoustic Sensing (DAS) enables the use of existing underwater telecommunication cables as multi-sensor arrays. The great majority of underwater telecommunication cables are deployed from the water surface and the coupling between the cable and the seafloor is not fully controlled. This implies that there exists many poorly coupled cable segments less useful for seismological research. In particular, underwater cables include segments that are suspended in the water column across seafloor valleys or other bathymetry irregularities. However, it might be possible to use DAS along the suspended sections of underwater telecommunication cables for other purposes. A first one investigated here is the ability to monitor deep-ocean currents. It is common to observe that some particular sections of a cable oscillate with great amplitudes. These oscillations are commonly interpreted as due to vortex shedding induced by the currents. We investigate this hypothesis by estimating the oceanic current speeds from vortex frequencies measured in two underwater fiber optic cables located at Methoni, Greece, and another in Toulon, France. Our results in Greece are in agreement with in-situ historical measurements of seafloor currents while our estimations in Toulon are compatible with synchronous measurements of a nearby current meter. These different measurements therefore point to the possibility to exploit DAS measurements as a tool to monitor the activity of seafloor currents. A second possible application of DAS is to estimate how the cable is coupled to the seafloor, even in the absence of the strong oscillations associated to vortex shedding. For that, we have analyzed the spectral signature of the different cables. Some sections feature fundamental frequencies as expected from a theoretical model of in-plane vibration of hanging cables. By analyzing how the fundamental frequencies change along the cable, we are potentially inferring the contact points of the cable with the seafloor, which will promote fatigue of the cable and potential failure. This mapping of the coupling characteristics of the cable with the seafloor could also be useful to better interpret other DAS signals.

How to cite: Mata, D., Ampuero, J.-P., Mercerat, D., Rivet, D., and Sladen, A.: The Potential of DAS on Underwater Suspended Cables for Oceanic Current Monitoring and Failure Assessment of Fiber Optic Cables, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-3729, https://doi.org/10.5194/egusphere-egu22-3729, 2022.

In June 2022, in the frame of the PREST interreg Caraïbe project, we installed an optical OBS offshore the Les Saintes archipelago (Guadeloupe, Lesser Antilles), at the termination of a 5.5 km long optic cable buried in the sea floor and landing in Terre-de-Bas island (FIBROSAINTES campaign: Antea vessel from the FOF, plow from GEOAZUR). This innovative seismometer, developped in the last decade by ESEO, is based on Fabry-Perot (FP) interferometry, tracking at high resolution (rms 30 pm) the displacement of the mobile mass of a 10 Hz, 3 component, purely mechanical geophone (no electronics nor feed-back). This optically cabled OBS is the marine version of the optical seismometer installed at the top of La Soufrière volcano of Guadeloupe, in 2019, at the termination of a 1.5 km long fiber (HIPERSIS ANR project). Both seismometers are telemetered in real-time to the Guadeloupe Observatory (IPGP/OVSG). The optical seismometer, located at a water depth of 43 m near the edge of the immersed reef, is aimed at improving the location of the swarm-like seismicity which still persists after the Les Saintes 2004, M6.3 normal fault earthquake. The considerable advantage of such a purely optical submarine sensor over commercial, electric ones is that its robustness, due to the absence of electrical component, guarantees a very low probability of failure, and thus significantly reduces the costs of maintenance. In May 2022, an optical pressiometer and an optical hydrostatic tiltmeter designed and constructed by ENS shoud be installed offshore and connected to the long fiber, next to the optical OBS.

Based on the same FP interrogator, ESEO and IPGP recently developped a high resolution fiber strainmeter, the sensing part being a 5 m long fiber, to be buried or cemented to the ground. A prototype has been installed mid-September 2021 on the Stromboli volcano, in the frame of the MONIDAS (ANR) and LOFIGH (Labex Univearth, Univ. Paris) projects. The interrogator was located in the old volcanological observatory, downslope, and the optical sensors, at 500 m altitude, were plugged at the end of a 3 km optic cable. They consist of three fibers, 5 m long each, buried 50 cm into the ground. Their different orientation allowed to retrieve the complete local strain field. The four weeks of continuous operation clearly recorded the dynamic strain from the frequent ordinary summital explosion ( several per hour), and, most importantly, the major explosion of the 6th of October (only a few per year). The records show a clear precursory signal, starting 120s before this explosion, corresponding to a transient compression, oriented in the crater azimuth, peaking at 0.9 microstrain  10 s before the explosion.

These two successfull installations of optical instruments open promising perspectives for the seismic and strain real-time monitoring in many sites, offshore, on volcanoes, and more generally in any site, natural or industrial, presenting harsh environmental conditions, where commercial, electrical sensors are difficult and/or costly to install and to maintain, or simply cannot be operated.

How to cite: Bernard, P., Plantier, G., Ménard, P., Hello, Y., Savaton, G., Metaxian, J.-P., Ripepe, M., Bouin, M.-P., Boudin, F., Feron, R., Deroussi, S., and Moretti, R. and the optic-OBS-strain-2022 team: Innovative high resolution optical geophysical instruments at the termination of long fibers: first results from the Les Saintes optical ocean bottom seismometer, and from the Stromboli optical strainmeter, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-10574, https://doi.org/10.5194/egusphere-egu22-10574, 2022.

On the 13th December 2020, a Strombolian eruption occurred on Mount Etna. We present a study of the temporal and spatial variation of strain measured at the underwater base of volcano during this event. 

As part of the FOCUS project, a BOTDR (Brillouin Optical Time Domain Reflectometry) interrogator has been connected to the INFN-LNS ( Istituto Nazionale di Fisica Nucleare - Laboratori Nazionali del Sud) fibre optic cable that extends from the port of Catania 25km offshore to TTS (Test Site South) in a water depth of 2km. This interrogator has been continuously recording the relative strain changes at 2m spacing along the length of the cable every 2 hrs since May 2020. 

On preliminary analysis, a change in strain is observed at the around the time of the eruption, however this variation occurs close to the shore where seasonal variations in water temperatures are in the order of 5°C. As Brillouin frequency shifts are caused by both temperature and strain variations, it is necessary to remove this effect. To do so, numerical simulations of seasonal sea temperature specific to offshore Catania have used to estimate the change in temperature along the cable. This temperature change is then converted to a Brillouin frequency shift and removed from the frequency shift recorded by the interrogator before being converted to relative strain measurements. This processing produces a strain signature that is consistent with deformation observed by nearby geodetic stations on land.

How to cite: Murphy, S., Garreau, P., Palano, M., Ker, S., Quetel, L., Jousset, P., Riccobene, G., Aurnia, S., Currenti, G., and Gutscher, M.-A.: Strain evolution on a submarine cable during the 2020-2021 Etna eruption, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-7742, https://doi.org/10.5194/egusphere-egu22-7742, 2022.

Distributed Acoustic Sensing (DAS) is ideally suited for the challenges of Earthquake Early Warning (EEW). These distributed measurements allow for robust discrimination between earthquakes and noise, and remote recordings at hard to reach places, such as offshore, close to the hypocenters of most of the largest earthquakes on Earth. In this study, we propose the first application of DAS for EEW. We present a framework for real-time strain-rate to ground accelerations conversion, magnitude estimation and ground shaking prediction. The conversion is applied using the local slant-stack transform, adapted for real-time applications. Since currently, DAS earthquake datasets are limited to low-to-medium magnitudes, an empirical magnitude estimation approach is not feasible. To estimate the magnitude, we derive an Omega-squared-model based theoretical description for acceleration root-mean-squares (rms), a measure that can be calculated in the time-domain. Finally, peak ground motions are predicted via ground motion prediction equation that are derived using the same theoretical model, thus constituting a self-consistent EEW scheme. The method is validated using a composite dataset of earthquakes from different tectonic settings up to a magnitude of 5.7. Being theoretical, the presented approach is readily applicable to any DAS array in any seismic region and allows for continuous updating of magnitude and ground shaking predictions with time. Applying this method to optical fibers deployed near on-land and underwater faults could be decisive in the performance of EEW systems, significantly improving earthquake warning times and allowing for better preparedness for intense shaking.

How to cite: Lior, I., Rivet, D., Sladen, A., Mercerat, D., and Ampuero, J.-P.: Real-Time Magnitude Determination and Ground Motion Prediction using Optical Fiber Distributed Acoustic Sensing for Earthquake Early Warning, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-8294, https://doi.org/10.5194/egusphere-egu22-8294, 2022.

Ocean Bottom Distributed Acoustic Sensing (OBDAS) is emerging as a new measurement method providing dense, high-fidelity, and broadband seismic observations from fiber-optic cables. Here, we use ~40 km of a telecommunication cable located offshore the Sanriku region, Japan, and apply ambient seismic field interferometry to obtain an extended 2-D high-resolution shear-wave velocity model. In some regions of the array, we observe and invert more than 20 higher modes and show that the accuracy of the retrieval of some modes strongly depends on the processing steps applied to the data. In addition, numerical simulations suggest that the number of modes that can be retrieved is proportional to the local velocity gradient under the cable. Regions with shallow low-velocity layers tend to contain more modes than those located in steep bathymetry areas, where sediments accumulate less. Finally, we can resolve sharp horizontal velocity contrasts under the cable suggesting the presence of faults and other sedimentary features. Our results provide new constraints on the shallow submarine structure in the area and further demonstrate the potential of OBDAS for offshore geophysical prospecting.

How to cite: Spica, Z., Viens, L., Perton, M., Nishida, K., Akuhara, T., Shinohara, M., and Yamada, T.: Understanding surface-wave modal content for high-resolution imaging with ocean-bottom distributed acoustic sensing, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-2455, https://doi.org/10.5194/egusphere-egu22-2455, 2022.

Recent advances in range and performance of distributed acoustic sensing (DAS) enable new geophysical applications by measuring fiber strain in existing telecom cables and subsea power cables that incorporate optical fibers. We will  present new field data showing the usability of DAS for environmental and geophysical applications, focusing especially on seabed surface waves and the sub-Hz domain. These examples show that highly sensitive DAS technology can be a valuable tool within seismology and oceanography.

The sensitive range along the fiber for DAS was previously limited to about 50 km. We will demonstrate a newly developed system (named OptoDAS) that allows for launching several orders of more optical power into the fiber, and thereby significantly improving the range beyond 150 km.

This new interrogation approach allows for high degree of flexibility optimizing the interrogation parameters to optimize the noise floor, spatial and temporal resolution according to the application. The gauge length (spatial resolution) can be set from 2 to 40 m. For interrogation of 10 km fiber, we achieve a record low noise floor of 1.4 pε/√Hz with 10 m spatial resolution. For interrogation of fibers beyond 150 km, we achieve a noise floor below 50 pε/√Hz up to 100 km. Above 100 km, the noise is limited by the level of reflected optical power, and the noise increases by ~0.3-0.4 dB/km, corresponding to the dual path optical loss in the fiber.

A modern instrument control interface allows for automatic optimalization of interrogation parameters based on application parameters in a few minutes. The instrument computer provides a flexible platform for different applications. The high-capacity storage system can store recorded time-series of several weeks to support e.g., geophysical investigations where extensive post-processing is required. The computational capacity can also be used for real-time visualization and advanced signal processing, for example for event detection and direct reporting of estimated parameters.

The OptoDAS system can convert a submarine cable into a 100 km+ densely sampled array.  From the recordings on a telecom cable in the North Sea, we will show examples of propagating Rayleigh and Love acoustical modes bounded to the seafloor surface. These modes can be excited by acoustic sources on or above the seafloor, such as trawls and anchors. The dense spatial sampling allows for accurate estimates of the location of these sources. The system also allows for applications in seismology and earthquake monitoring. When attached to a cable with non-straight geometry, the measurements have substantial information to determine the location of seismic events. This will be demonstrated using field data from the North Sea telecom cable.

From recordings on a submarine cable between Norway and Denmark, we present the DAS response in the frequency range 0.1 mHz-10Hz across a cable span of 120 km. The response in this frequency range will be a combination of temperature changes, ocean swells and tides. We show that increasing the gauge length in post-processing allows for improving the sensitivity for detecting ultra-low frequency signals.

How to cite: Rønnekleiv, E., Waagaard, O. H., Morten, J. P., and Brenne, J. K.: Long range distributed acoustic sensing technology for subsea geophysical applications, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-11869, https://doi.org/10.5194/egusphere-egu22-11869, 2022.