Speakers and Posters

This presentation highlights our technical developments on fiber strain measurements and deep seafloor borehole observatories in the Nankai Trough aimed to monitor slow slip events and slow earthquakes, with several examples during the past 10 years.
　Slow earthquakes have been known to occur episodically in the offshore area of the Nankai Trough, but long-term continuous observations in a borehole drilled in the Kumano-nada as part of the IODP NantroSEIZE project have revealed the first episodic slow slip that occurs repeatedly over a period of about one year. These slow slip events vary in whether or not they are accompanied by slow earthquakes and are thought to depend on factors such as the structure and stress conditions of the plate boundary. We believe that wide-area observations are needed to resolve such features. We have been deploying a submarine fiber-optic strainmeter off the Kii Channel since 2019, and we were able to capture the slow slip and VLFE that occurred in 2022. This shows that observations close enough to the fault can directly capture the slow motion of the fault. Such submarine fiber-optic strainmeter observations were deployed to several components in 2022 and to the Kumano-nada C0002 site in 2024, and offline observations are planned to be deployed to the Hyuga-nada in 2025.
　In fiber optic sensing using submarine fiber optic cables, we are targeting the observation of slow earthquakes and SSE. We have developed fiber optic sensing technologies such as broadband DAS using ultra-stable lasers and TW-COTDR, which enables stable measurement of crustal deformation, and are conducting observations using submarine cables extended up to 120 km off Muroto. Although the detection of seafloor crustal deformation requires correction for seafloor temperature changes, we have been able to observe and analyze low-frequency tremors, and have observed tremors off Muroto that are thought to have been induced by the Noto Peninsula earthquake of January 1, 2024.
　The seafloor borehole observation capable of SSE observation also incorporates these fiber-optic strain observation techniques, and a new seafloor borehole observation system was installed off the Kii Channel in November 2023, and long-term observations have begun (C9038B). In the future, similar borehole observation systems will be installed off Shikoku and off Hyuga-nada to understand plate boundary phenomena in the Nankai Trough.

We introduce the time-series data of pore pressure in three boreholes off Kumano and new one off Kii Channel, in order to detect crustal deformation.

In the Nankai subduction zone, megathrust earthquakes with magnitudes of ~8 have
occurred repeatedly, and slow earthquakes have been observed by offshore seismic and geodetic
data. To investigate the temporal change in the stress state around the plate boundary where
megathrust and slow earthquakes occur, monitoring of slow and regular earthquake activities is
important. In the Nankai Trough, distributed acoustic sensing (DAS) measurement has been
conducted continuously using a submarine cable with a 120 km length extending from the Cape
Muroto in Shikoku. In DAS measurement, temporal change in strain rate along a fiber-optic cable is
measured and high-density observation with a meter-scale is realized. In this area, an offshore
seismic network DONET was also installed. We worked on slow and regular earthquake analyses
using the DAS and DONET data.
On January 1st in 2024, 13 shallow tremors, a type of slow earthquakes at the frequency
range of 2–8 Hz, were observed by DAS and DONET data. This tremor activity started
approximately three hours after the M 7.6 Noto Peninsula earthquake and continued for about nine
hours. We located these tremors by envelope correlation and amplitude attenuation with distance in the DAS and DONET data. The tremors were located on the western edge of the slow earthquake area reported in previous studies. Although broadband seismometers of many stations have not worked correctly during the tremor activity due to the strong motions of the Noto Peninsula earthquake, DAS has not been affected and has worked continuously. The hypocenter location was achieved by the DAS observation. DAS can be an effective tool when a large earthquake occurs.
To monitor the change in seismic activity promptly, we are constructing an automatic hypocenter location system using the DAS and DONET data. In this system, P- and S-wave arrivals of earthquakes are picked automatically using a deep neural network model, PhaseNet (Zhu and Beroza, 2019) trained with DAS and DONET seismographs. After picking, the hypocenter is located by a hypocenter location program, hypomh (Hirata and Matsu’ura 1987) and a physics-informed neural network model, HypoNet Nankai (Agata et al., 2024). In addition, to consider the focal mechanisms of regular earthquakes, we calculated cross-correlations between an earthquake pair near the DAS cable following Li et al. (2023). In this presentation, we present the latest results of these analyses.

“Seismic gaps” refer to segments along active geologic faults that have not ruptured in a time period comparable with the recurrence intervals of past earthquakes, and are therefore expected to rupture at any time. This concept usually describes spatial variations in earthquake recurrence in the strike dimension. However, modern seismological observations show that rupture can occur at a variety of depths, so that the lack of a seismic gap in the strike direction may not indicate the lack of a seismic gap in the dip direction.This recognition has important hazard implications. For example, within a given strike segment of a subduction megathrust, an earthquake may rupture only one depth section, with another section still accumulating energy for a future large earthquake. Here, we present unambiguous historical evidence for the presence of such an unruptured depth section in metropolitan Chile. In this area, four great earthquakes (1730, 1822, 1906 and 1985) were documented in the written history, but only the first (1730) of this sequence was accompanied by a large tsunami. By combining newly found first-hand accounts of coastal uplift and tsunamis with coupled deformation-tsunami models, we show that the three post-1730 earthquakes failed to produce large tsunamis because their rupture zones were confined to relatively large depths beneath land. In contrast, the large tsunami in 1730 was the consequence of large slip of the shallow section of the megathrust beneath the sea. Because the shallow section has not ruptured in a time period comparable with the recurrence intervals of prehistoric tsunamis, it creates a “tsunami gap” today along Chile’s most populated coast that could be filled by a large tsunami at any time. Similar tsunami gaps are likely present in other subduction zones and should be addressed in hazard analysis and risk mitigation.

Understanding the complex coupling behaviors in subduction zones remains a crucial challenge in geodesy. Land-based geodetic stations often fail to capture offshore deformation beyond approximately 40 km, making them unable to resolve subduction coupling in the regions of largest slip and tsunami potential in large earthquakes. Because of this limitation, the physical process of coupling, especially in the shallower near-trench regions of the subduction zone, is still poorly understood. This includes seismic behavior from both giant earthquakes (Mw > 8.5), and a class of events called tsunami earthquakes, which produce tsunamis around 10-30 times larger than expected from its seismic magnitude. Alaska has experienced both such events over the past century, and its rapid convergence allows for high-quality geodetic case studies that can be applied to subduction zones across the globe. Because of this, the region was recently selected for two NSF-funded near-trench studies of geodetic behavior: The Near-Trench Community Geodetic Experiment, and The Mesh GNSS-Acoustic Array design experiment allowing for a dense field of study within just a few km of the trench around the rupture area of a tsunami earthquake that occurred in 1946. All sites were deployed and first measured in the Summer of 2024. Subsequent observations will be made annually over the next few years. Another project was previously funded in 2017.
Here, we evaluate the effectiveness of the distribution of these seafloor geodetic sites for answering questions of large-to-small scale coupling of the megathrust by evaluating the resolvability of individual model patches, using a detailed 3D slab interface (Slab 2.0) modified to account for bathymetry. The modeling efforts here will help to guide future studies (perhaps to densify the coverage) and to improve our understanding of subduction zone earthquakes and tsunamis.

We present 2-D velocity (Vp) tomographic models and multichannel seismic reflection images from offshore Chile. Additionally, we introduce a new slab depth grid constraining the top of the oceanic Nazca Plate offshore South America, which will be compared with previousslab models.

The megathrust fault between the subducting Cocos plate and the overriding Caribbean plate has experienced two recognized tsunami earthquakes: the September 2, 1992, Nicaragua event (MW 7.6) and the August 27, 2012, El Salvador event (MW 7.3). The Nicaraguan event is a tsunami earthquake, generating enormous tsunamis relative to its earthquake size. It has been concluded that the Nicaraguan earthquake occurred with a slow rupture, and the associated crustal deformation hosted a long-time constant (Ide, Imamura, and Abe, 1993). El Salvador event has a relatively long source duration, low seismic moment-scaled radiated energy, and a depleted short-period seismic source spectrum. These characteristics are similar to those of the 1992 Nicaragua tsunami earthquake, which ruptured the adjacent shallow portion of the plate boundary (Ye, Lay, and Kanamori, 2013). Borrero et al. (2014) show that this event generated a surprisingly large local tsunami. Ye, Lay, and Kanamori (2013) suggest that other tsunami earthquakes could occur along the shallow megathrust extending from Guatemala to Nicaragua. We used the work of Newman and Okal (1998), which demonstrated that scaled energy calculated from observed waveforms is one of the most potent discriminants for identifying tsunami earthquakes. Tsunami earthquakes typically show scaled energy values ranging from 7x10^-7 to 3X10^-6 (Venkataraman and Kanamori, 2004). We estimate the radiated seismic energy from teleseismic P-waves using the methodology proposed by Perez-Campos and Beroza (2003). To identify other earthquakes with features similar to tsunami events, we calculated the scaled energy of events near the subduction zones in Central America. The lowest scaled energy value corresponds to an MW 6.7 event on March 3, 1996, near the 1992 Nicaragua event in Central America. In contrast, many strike-slip earthquakes exhibit higher energy.

Tectonic tremor is the primary seismic signature of slow fault slip, and its spatiotemporal behavior is key to understanding the underlying physical mechanisms driving slow deformation. Tremor is thought to emerge from the superposition of many low-frequency earthquake sources, each with distinct P- and S-phases. However, tremor lacks clear seismic phases, requiring envelope-based detection and location methods that discard complex phase information and retain only coarse measures of signal energy. The utility of envelope-based tremor locations depends strongly on the density and aperture of the seismic array used to backproject the seismic energy toward a grid of potential sources. Evolving network coverage and intermittent deployment of dense networks results in slow slip events resolved with differing quality, particularly when analyzing historical datasets or regions with sparse instrumentation.
We address the evolving tremor detection capacity of seismic arrays by training a convolutional neural network to map continuous waveforms from a single three-component seismic station to the tremor times and locations from the Cascadia tremor catalogs of Ide (2012) and Wech (2021). Implicitly, this approach uses the detection capacity of the full seismic network of present-day Cascadia to identify patterns in tectonic tremor not visible when viewed in isolation. Tectonic tremor is composed of repeating earthquakes whose waveforms encode a stationary seismic source and location, but the complex temporal activation of multiple such sources makes it difficult to explicitly use this information for tremor location. By directly mapping from single-station continuous seismic data to the network-constrained tremor locations, we produce a representation of the tremor that is invariant to the particular temporal activation of LFEs, encoding the Green’s function between the tremor source and the seismic station. We apply this model to over 20 years of continuous seismic data in Cascadia to detect tremor sources with uniform detection capacity.

We present data from a first-of-its-kind deployment of A-0-A "drift-free" pressure instruments on the Hikurangi margin. These data are sensitive to the vertical secular strain signal from ongoing convergence. We will discuss the challenges of dealing with oceanographic signals and the promising outcomes from additional data that is currently being collected in the region.

The nature of incoming oceanic plates likely plays a significant role in a range of subduction-zone processes that are central to the SZ4D and SZNet initiatives. Key lithosphere characteristics include variations in sediment distribution, abyssal-hill fabric, crustal faulting, oceanic crustal thickness and associated topography, mantle-lithosphere composition, and crust and mantle hydration. Here, we present new seismic imaging analyses using broadband ocean-bottom arrays that suggest strong velocity heterogeneity in oceanic lithosphere across a wide range of ocean-plate environments. These results suggest that the full range of lithosphere characteristics vary in oceanic plates over length scales that are comparable to along-strike variations in subduction behavior.

The southernmost part of the Cascadia region is where the Gorda, North American, and Pacific plates meet each other, forming the Mendocino Triple Junction (MTJ). The interaction of these three plates leads to intense seismic activity near the MTJ and creates complex fault networks that extend from the overriding plate down to the mantle of the subducting plate. Existing earthquake catalogs provide limited constraints on earthquake locations offshore. Utilizing data from the amphibious seismic array deployed during the Cascadia Initiative experiment, we applied machine-learning detection algorithms to build an enhanced earthquake catalog for this region. Combined with focal mechanism solutions, our catalog reveals several northeast-southwest-oriented strike-slip faults in the incoming and downgoing Gorda plate, clusters of earthquakes near the subduction plate interface, and both normal and thrust faults in the overriding plate. This diverse faulting reflects rapidly varying stress conditions near the MTJ and highlights the multifaceted earthquake hazards in this area.

I will present results from the deployment of two seafloor optical fiber strainmeters off the coast of Oregon in the Cascadia subduction zone, with the goal of detecting shallow slow slip at the base of the locked zone. We detect a signal on one strainmeter that is a close match to the predicted signal from offshore slow slip. I will also discuss the seafloor optical fiber strainmeter instruments in general, including their design, installation, data corrections, and detection capabilities over a wide range of frequencies, spanning both the seismic and geodetic bands.

Seismic ambient noise has revolutionized regional crust and upper mantle imaging, with ambient-noise tomography becoming a critical tool for broadband ocean bottom seismometer (BBOBS) studies. Marine environments offer unique advantages, such as strong microseism sources and thin, simple crust, enabling clear seismic observations. However, the oceanic setting also poses challenges, including the presence of interface waves and noise sources such as tilt and compliance, which degrade seismic signal quality. Advances in noise mitigation strategies, including pressure-to-vertical transfer functions, have enabled the separation of fundamental Scholte waves from overtone Rayleigh waves, improving ambient noise cross-correlation functions (CCFs). Here, we leverage data from recent U.S.-led BBOBS deployments to evaluate the variability in ambient noise CCFs across different oceanic environments and instrumentation. Our results provide new insights into the environmental conditions and instrumental designs that shape ambient noise tomography performance in BBOBS deployments, offering insights for optimizing future studies not only using traditional broadband seismic array deployments, but novel seafloor instrumentation technologies.

Cabled ocean bottom seismometer (OBS) solutions are financially and logistically challenging and autonomous OBS systems do not provide operators with seismic data until recovery. To address this issue Güralp has developed Aquarius, an ultra-low-power, free-fall OBS system, operational at any angle and with the ability to transmit seismic data in near-real-time from the seafloor without the use of cables.
The Güralp Aquarius incorporates seafloor-to-surface acoustic communication technology that allows state-of-health and noise performance interrogation during installation followed by retrieval of seismic data throughout the deployment period.
Omnidirectional broadband seismometer components allow the Aquarius to land and operate on steep slopes without requiring a gimbal mechanism that inherently introduces noise and failure modes. Raw data is recorded uncorrected for orientation to allow users to correct during post-processing.
These unique features allow the sensor to function on uneven seafloor as well as transmitting seismic data to the surface where the operator can use noise characteristics, location, and orientation data to determine if the landing site is suitable.
Intelligent battery design allows for typical 18-month deployments, with charging being possible alongside data transfer. This allows recharging, download and configuration simultaneously on the ship in between deployments.
Ease of configuration, deployment and recovery followed by simple data processing are all central themes to the Aquarius. The capital investment required to purchase OBS systems often means that the OBS instrument must be adaptable to a range of use-cases.

Real-time monitoring of seafloor vertical displacements just above subduction zones would contribute to estimate coupling strength state. However, it has been considered difficult to extract vertical displacements from pressure records due to inherent drifts of pressure gauges. We have developed a calibration method of the real-time seafloor pressure gauges at the centimeter scales using the mobile pressure calibrator. The in-situ calibrations have been repeatedly carried out since 2018. Here we present long-term pressure records without the inherent drift components in the Nankai trough seismogenic zone. Furthermore, we are developing a technology to measure high-precision seafloor topography using a high-precision pressure meter and a laser scanner. This technology is envisioned to automatically calibrate seafloor hydrometers using an autonomous underwater vehicle in the future. This presentation will also introduce this effort.

Pressurization experiment using a pressure standard (a dead weight tester) is implemented prior to the installation of pressure sensors into the seafloor or the borehole to evaluate the sensor drift.

TBD

We aim to provide an in-depth understanding of the slip behavior in the shallower segment of the seismogenic zone in the Atacama seismic gap, Chile, by constraining the updip extension of the locked zone, its structure, and mechanical control. This region last experienced large earthquakes in 1819 (Mw8.0) and 1922 (Mw8.4) and is therefore considered a mature seismic gap. In a collaborative framework between the Integrated Deep-Ocean Observing System (IDOOS) project and GEOMAR (Kiel, Germany), we installed five Absolute Pressure Sensors (APG) in March-April 2023. The APG sensors are Fetch Ambient-Zero-Ambient (AZA) transponders, capable of removing instrumental drift and measuring vertical displacements of the seafloor with millimeter to centimeter resolution. We select the APGs location using a Bayesian methodology that maximizes the resolution of shallow megathrust slip. Two oceanographic moorings near the pressure sensors will be installed in September 2023 to provide valuable in situ measurements of conductivity, temperature, pressure (CTD), CO2, and oxygen at various depths. These data will be used to remove oceanographic noise from the pressure time series observed on the seafloor. The IDOOS platform is planned to operate for five years, providing a unique opportunity to integrate long-term multi-parametric oceanographic observations with a dense land-based geodetic-seismological network to comprehend the mechanisms controlling deformation and seismicity in a mature seismic gap in Chile.

The use of remotely operated surface vehicles has substantially reduced the costs of some seafloor geodetic observations in subduction zones. However only optimal submarine conditions have been thus far been significantly explored, limiting the environments deemed feasible for application. Here, we report on a design experiment that applies an efficient mesh geometry to directly address assumed depth and gradient limitations to enable wide utilization of these tools.
We are in the 1st year of a 3yr design experiment where we deployed a network of transponders between 4 and 6 km water depth (deeper and steeper than previously tested) along a segment of the overriding plate near the 1946 tsunami earthquake rupture along the Aleutian trench. To improve observations at greater depths, we reduce signal attenuation using lower frequency acoustics are used than for shallower studies. The mesh-network design allows for a substantial increase in observation points over a traditional geometry that requires three transponders, as having the ability to reuse existing transponders from multiple centroids demonstrates a substantial reduction in instrumentation and cost. While the mesh design geometry will dictate the ultimate transponder efficiency for observation, an example network may have between 4 and 10 times the number of possible observations, substantially reducing instrument costs, and facilitating high-density observations of seafloor deformation that can be used to discriminate shallowly seated sources including submarine magmatism, near-trench deformation, hydrocarbon withdrawal, and submarine landslides. The experiment provides progress in methodologies regarding deployment strategies for dense observations, as well as observations in steep and deep environments.

We present a compilation of the systematic computation of Energy-to-Moment ratios using the formalism of Newman and Okal, which allows the identification of source slowness, notably in the context of the so-called "tsunami earthquakes", and conversely of snappy, usually intraplate events leading to excessive accelerations and hence destruction. Its extension to intermediate-depth and deep earthquakes reveals much more homogeneous ratios in each of the depth bands considered, with only two clearly slow events in the 100-km band, out of more than 700 events studied. Remarkably, these two earthquakes occurred at the same location under the island of Bougainville (Papua New Guinea). The origin of this intriguing observation remains speculative.

Imaging the type and distribution of slip behavior in subduction megathrusts is crucial to understand the physics of earthquakes and their impact on human settlements. Particularly, a robust characterization of the updip seismogenic limit is key for improving tsunami hazard assessments, as shallow slip commonly results in larger seafloor deformation. In addition to slip inversion being a highly ill-posed problem, it is particularly difficult to estimate subduction megathrust near-trench slip unless offshore geodetic observations are available. Here, we take advantage of recent improvements in slip inversion and uncertainty quantification techniques to develop a novel methodology to determine the optimum location of Absolute Pressure Gauge (APG) sensors to improve fault slip estimation. We select the best location of 4 APG sensors, whose deployment was coordinated by the Integrated Deep-Ocean Observing System for geoscience research (IDOOS) in northern Chile, offshore of the Taltal town (~25ºS) Chile, complementing a densification of the GNSS network deployed in the region by PRECURSOR ACT 192169 ANID Chilean project.

GNSS/Acoustic technology enables the measurement of absolute horizontal seafloor displacements through repeated surveys of acoustic beacons over several years. However, a key limitation of this approach is that the beacons must remain underwater for extended periods.

In this study, we explore an alternative method that combines photogrammetry and GNSS/Acoustic techniques to determine the absolute positions of distinct seafloor features, such as rocks, outcrops, or shipwrecks. Our approach utilizes an Autonomous Underwater Vehicle (AUV) equipped with a high-resolution digital camera capable of capturing overlapping, near-vertical images with sub-centimetre pixel resolution. The AUV's trajectory is managed using multiple Uncrewed Surface Vessels (USVs), which are geolocated with GNSS systems.

This method produces orthorectified images of the seafloor that may still exhibit internal deformations and limited georeferencing accuracy. To address this, we incorporate a temporary GNSS/Acoustic system deployed during the photogrammetric survey. Acoustic beacons are used as ground control points to correct internal distortions within the photogrammetric model and to accurately anchor it to an external reference frame with centimetre-level precision.

In this poster, we present the current status of our project and highlight several key challenges that need to be addressed to operationalize this innovative approach.

Similar to efforts at subduction zone forearcs, studying the transform plate boundary that links the Cascadia and Alaska convergent margins in SE Alaska and British Columbia requires obtaining geophysical data that extends along the continental slope. Along this margin, where the Queen Charlotte fault accommodates > 5 cm/year of slip, significant deformation extend across the shelf and slope, offshore where water depths span ~100m to >3000m. In this presentation, I will share successes and challenges associated with imaging the plate boundary faults and deformation structures across the Queen Charlotte margin and continental slope, and recording micro-seismic earthquake activity there using a dense array of broadband ocean bottom seismometers. Although a different tectonic regime, may lessons learned from the Queen Charlotte experiment have relevance to future plans for obtaining frontier datasets crossing subduciton zone forearcs.

This talk will provide a review of deformation and rheology both along the shallow megathrust and in wall rocks, with a focus on: (1) distribution and rates of deformation relevant to offshore monitoring and the interpretation of geodetic data; (2) quantification of in situ stress and fluid states that provide context for models of fault locking and slip; and (3) measurements of frictional rheology that offer mechanistic insights into shallow megathrust behavior. Examples from the Nankai and Hikurangi margins will highlight the integration of observations to illuminate processes that may govern fault slip and locking behavior.

We present GNSS-A observations on the Cascadia Subduction Zone, which indicate that the shallow megathrust is mostly locked near the deformation front offshore Oregon. We also describe our operational protocols and lessens learned for collecting GNSS-A data using an unmanned surface vehicle.

A seismic observation using Distributed Acoustic Sensing measurement (DAS) is attractive because spatially high-density data along a fiber can be obtained for a long distance. DAS measurement using seafloor cables is effective to increase the number of seismic stations in marine areas. The ERI, the University of Tokyo deployed a seafloor seismic tsunami observation system using an optical fiber cable off Sanriku in 1996. We have performed temporary DAS measurements since February 2019 using spare fibers of the Sanriku system. In 2010, the ERI also installed a seafloor cable seismic observation system near Awashima Island in the Japan Sea. From April 2023, temporary DAS measurements were sporadically performed using a fiber with a length of about 22 km. Especially, the aftershock observation of the 2024 Noto-Hanto earthquake was conducted from February to May, 2024. As a result of these observations, many earthquakes including micro-earthquakes and teleseismic events could be observed. The noise levels of the DAS measurement were comparable to those of conventional servo-type accelerometers. The Nankai Trough Seafloor Observation Network for Earthquakes and Tsunamis (N-net) is being constructed in the western Nankai trough area. A part of N-net was deployed and a pilot DAS observation was perform from August to October, 2024. During the pilot DAS observation, a large earthquake with a magnitude of 7.1 occurred near the cable system and was recorded. Following large aftershocks were also recorded. Using the records of large events, performance of DAS measurement under a large deformation was recognized. In this presentation, observations using DAS technology on actual seafloor cables around Japan and scientific results from the DAS data will be introduced.

Our standard offline broadband ocean bottom seismometer (BBOBS) has been established around 2005, and we have made several large array observations in these 20 years. Besides, we also try to improve the data quality and to extend the region to cover by the BBOBS. One aspect is the BBOBS-NX with a penetrator BB sensor that is operated by a submersible vessel , and its autonomous version, NX-2G. Another is the ultra-deep BBOBS (UDBBOBS) which is designed to deploy around the trench deeper than 6000m. The NX-2G and the UDBBOBS are now in the functionality evaluation observation.

We developed a novel GNSS-Acoustic instrument for seafloor geodesy that em- ploys a solar-powered motored autonomous surface vessel equipped with a trans- ducer emitting acoustic pulses. The positions of seafloor geodetic reference stations over time help constrain seafloor strain rates, e.g., to determine plate- tectonic motion, and infer seismic and volcanic hazards. We minimize the least- squares travel-time misfit to estimate receiver position, with an additional step to debias the results from the uncertainty associated with the submerged position of the transducer, applicable to the case where that is unknown and may be large, such as when traditional crewed vessels are used. The question of interest is how instrument noise propagates into receiver position uncertainty. By sta- tistical analysis of a suite of synthetic experiments that replicate the real-world scenario of an experiment conducted in 5 km deep-water offshore Bermuda, we determined that 2 cm of GNSS and 20 µs of receiver noise transform into approximately 1 cm in receiver position uncertainty. Thus, our method holds promise to measure geodetic positions and geologic strain rates in the deep-sea over repeated measurement campaigns.

Like earthquakes, slow slip events release elastic energy stored on faults. Yet, the mechanisms behind slow slip instability and its relationship with seismicity are debated.
Here, we use a seismo-geodetic deployment to document a shallow slow slip event in 2023 on the Chile subduction. The slow slip initiation is driven by structurally-confined fluid overpressure, as evidenced by seismic swarms next to a subducted seamount, migrating along the interface and activating splay fractures. The seismic rupture of a permeability seal allows slow slip to accelerate and expand. Historical earthquake swarms highlight the persistent structural control and recurrent nature of such slow slip events. Our observations show that interactions between slow slip and seismicity are controlled by creep on a fluid-infiltrated fault with fractally distributed asperities.
Such observation would be facilitated by offshore instrumentation.

Nucleation and propagation of seismic rupture can occur inside a restricted range of temperatures (100-400ºC) where the frictional properties of seismogenic faults allow unstable sliding. As temperature of the megathrust is the main factor influencing the amount of heat flowing through the Earth surface, in-situ measurements of heat flow (HF) are essential to constraint models of earthquake generation. Here we show that the distribution of HF data offshore Chile is very heterogeneous, with a rather small number of good-quality measurements north of 33ºS compared with the southern segment. This heterogeneity impacts the uncertainties of megathrust thermal models and therefore degrades our capacity to predict the extend of the seismogenic zone, mostly in the area proposed for the SZ4D Multi-Array. We will discuss possible strategies to advance in the solution to this problem in the mid-term.

In the last years, fibre optic sensing methods, among them Distributed Acoustic Sensing (DAS) and State-of-Polarisation (SoP) have been demonstrated to be suitable for monitoring Earth System parameters in submarine cables through several pilot experiments, but full integration with telecommunication infrastructure has not yet been achieved. A new development concerns sensor pods containing classical sensors placed along telecom Cables (making them SMART-cables). Here, we introduce the EU HORIZON funded SUBMERSE project (SUBMarinE cables for ReSearch and Exploration) and the Helmholtz Large Infrastructure proposal SAFAtor (SMART Cables And Fiber-optic Sensing Amphibious Demonstrator).
The SUBMERSE project links Research and Education Networks (RENs), universities, research institutes and industry to establish multi-method monitoring along submarine optical telecommunication cables at several key oceanic cable routes branching off from Sines in Portugal, Madeira, Svalbard and in the Ionian Sea. Those pilot sites should serve as a blueprint for establishing continuous monitoring services along many more cables. The project comprises technical developments for integrating DAS and SoP measurements, for establishing differential SoP measurements and for operating DAS in in fibres carrying telecom traffic. A range of geoscientific and marine biology use cases are included, which seek to establish code/services for monitoring earthquakes and tsunamis, tracking whales, etc. It is planned to distribute the data collected by SUBMERSE through EIDA but one of the lessons from this project is that striking a balance between national and cable security concerns and data openness can be challenging, and suitable protocols are still being worked on.
SAFAtor aims at closing the observational gap in the continental shelf, slope, and deep oceans, with three main objectives 1) establish DAS permanent offshore monitoring at three existing Plate Boundary Observatories in coastal areas (N Chile, Marmara Sea and Sicily; 2) drive innovation in SMART sensor technologies by equipping a submarine telecom cable with robust sensor packages to measure temperature, absolute pressure and ground acceleration on the sea floor. 3) develop a FAIR data infrastructure to process, archive and distribute these new DAS and cable data, and enable the global user community to select and process the data services in a user-friendly and interoperable way.

With over 26,000 stations in the International Registry of Seismograph Stations, Earthquake-prone regions around the globe are densely instrumented and monitored. The scientific community is increasingly looking to the ~71% of Earth's surface covered by oceans for a deeper understanding of the Earth's structure, tectonic processes, and potential hazards through Ocean-bottom seismic (OBS) data acquisition. However, the unique challenges of deep-sea environments require innovative engineering solutions and robust manufacturing techniques to safeguard data quality, data completeness and system reliability, without compromising ease of deployment. These challenges have largely restricted the scope of studies until the recent wider market availability of OBS platforms. Currently, a range of cabled and autonomous ocean bottom solutions represent the state-of-the-art deployments that support the global community’s study of the oceans’ environment, its dynamic properties, and natural or triggered events on the seabed

This presentation provides a comprehensive overview of the engineering challenges in the domain of OBS platforms, as well as the technology and capability solutions that the Nanometrics team developed to address those challenges for differing configurations and use cases. With proven technologies such as integrated kinematic gimbals to level at all landing tilt-angles and designs certified for deployment depths to 6000m, Nanometrics has enabled seamless multidisciplinary data collection across diverse marine environments, supported by a wide range of application-driven options for sensing instruments and dataloggers.

We present preliminary results of the analysis of the interseismic coupling and the relationship with the spatial distribution of the seismicity at the southern Peru subduction zone including the Nazca Ridge. We used and analyzed GPS data from over 120 permanent and campaign GPS sites that were collected during the last decade. The obtained GPS velocity field shows the current state of interseismic deformation of the Peruvian subduction margin reflecting high rates of horizontal displacement decreasing inland. The inversion of the geodetic displacements allows us to estimate the interseismic coupling at the plate interface. Our results show that the interseismic coupling is heterogeneous, with two areas of low to weak coupling, one located over the Nazca ridge and the other near the Nazca fracture zone. These results are compared with the spatial distribution of the seismicity recorded by the Peruvian national seismic network and a temporary seismic network for the period 2022-2024, which accounts for more than 100,000 events. The analysis reveals a remarkable correlation of the areas where high interseismic coupling is observed with lack of seismicity, whereas in the areas with low interseismic coupling an intense seismic activity is observed. These results confirm the hypothesis that the Nazca Ridge is low to weakly locked so can act as a persistent barrier against the propagation of earthquake rupture, also these results suggest that the interseismic coupling patterns could be associated with the seismic activity. This ongoing work sheds light on a formidable subduction setting and provides valuable information for understanding the tectonic processes in the region and their implications for the earthquake potential.

Resolving the distribution of transient slow slip event (SSE) and interseismic coupling processes at offshore subduction zones is hampered by the difficulty of measuring offshore crustal deformation. Four methods are overcoming these challenges: (1) GNSS-Acoustic techniques to track the horizontal motion of the seafloor; (2) ocean bottom pressure sensors to discern cm-level seafloor pressure changes due to vertical deformation, (3) formation pore pressure changes in subseafloor IODP observatories as a proxy for formation volumetric strain, and (4) direct-path acoustic measurements to resolve mm-level changes in baseline length over short (a few km) baselines. All of these approaches have been employed (concurrently with deployment of ocean bottom seismometers) in several experiments over the last decade along the Hikurangi subduction zone offshore New Zealand, with the goal of capturing near-trench slip behavior, and its relationship to seismicity. I will present new insights into shallow megathrust slip behavior (transient deformation and interseismic coupling) from the diverse array of seafloor geodetic deployments undertaken over the last decade. I will discuss lessons learned from the various techniques, and the suitability of each type of seafloor geodetic technique for detecting a range of subduction margin processes, and insights on network designs capable of capturing a range of deformation processes

The Chilean margin is one of the Earth’s tectonically most active plate boundaries and is shaped by the subduction of the oceanic Nazca Plate beneath the continental South American Plate. Although tsunamigenic megathrust earthquakes repeatedly occurred along the margin, some parts are still largely underexplored. The South Atacama region (25°-27.8°S) is suggested to be a mature seismic gap with the last >M W 8 earthquake in 1922. Precise information about the velocity structure of the marine forearc, the crustal thickness, the slab geometry of the down-going plate, and the morphological structures of the seafloor is missing for that area. We use wide-angle seismic refraction and local seismicity data to image the marine and continental forearc as well as the outer trench region. For the seismic refraction data, we utilize seismic signals from offshore airgun shots and onshore mining blasts that were recorded on ocean bottom seismometers and seismic land stations to image two amphibious 2D profiles. The local seismicity is derived from a densely spaced network of 53 temporary land seismometers. The P-wave velocity model of the incoming Nazca Plate reveals typical oceanic crust characteristics with slightly lower velocities and increased thicknesses at the Copiapó Ridge seamounts. In addition, we observe elevated seismic velocities at the base of several seamounts, which we relate to local magmatic underplating. Compared to an abrupt dip angle change from 10° to 22° at ~22°S, the down-going slab in our study area subducts with an angle between ~12° and ~18° below the marine forearc, smoothly increasing to ~22° at greater depths (40 – 60 km). The local seismicity shows three separated bands of earthquakes parallel to the down-going plate, which we relate to the subduction interface, the oceanic Moho and the Double Seismic Zone ~20 km below the plate interface. The largest event (M W 5.9) and aftershocks during our observation period (February – June 2023) occurred close to the coast beneath a densified land station network. The event occurred in the deepest band at ~54 km depth, ~20 km below the subduction interface in the upper oceanic mantle. We will discuss our findings in the context of other seismicity and refraction studies in the area and analyze variations in seismicity and subduction geometry along the northern Chilean margin.

Japan Coast Guard conducted the bathymetric surveys just after the occurrence of the 2024 Noto Peninsula earthquake. By comparing the bathymetry before and after the earthquake, we clarified the change of seafloor topography near the source region.

The Alaska Peninsula has a long history of plate subduction with along-arc variations in volcanic eruption styles and geochemistry. The Alaska slab morphology below 200 km depth remains debated due to limited seismic data and thus low tomography resolution in this region. Here we utilize the newly available regional and teleseismic data to build 3-D high-resolution VP and VS models to 660 km depth. We find that the high-velocity Pacific Plate subducts to the bottom of the mantle transition zone (MTZ) with complex deformation and gaps. In the southwest, we observe a wide gap in the high-velocity slab at 200–500 km depths. Towards the northeast, the slab becomes more continuous extending to the MTZ with a few holes below 200 km. We interpret these gaps as a slab tear that coincides with the subducted ancient Kula-Pacific Ridge. We also invert for 3-D VP, VP/VS, and QP models to 200 km depth with higher resolutions and find strong along-strike changes in slab dehydration and sub-arc melting, indicated by low VP, high VP/VS, and high 1/QP anomalies. Slab dehydration and sub-arc melting are most extensive below the Pavlof and Shumagin segments in the southwest, becoming limited below the Chignik and Chirikof segments in the northeast, and extensive again beneath the Kodiak segment further to the northeast. We propose that the variations of slab hydration at the outer rise significantly influence slab dehydration at greater depths and further control sub-arc melting beneath the Alaska Peninsula.

Detecting shallow slow slip events offshore is important but very challenging due to ocean noise. Here we suggest that including seafloor current data could help reduce oceanic noise in pressure data therefore better detecting shallow slow slip events. We will present some data from Cascadia and New Zealand.

A strain gauge using optical interferometers has been developed and is installed in the deep sea to measure crustal deformation. The strain gauge uses two interferometers with optical fiber with different temperature characteristics for the sensing optical fiber to eliminate the effects of temperature from strain measurement. The improved measurement method proposes an optical system that reduces the temperature effect of the fiber referenced in the strain measurement with additional interferometer.

Advances in GNSS-A analysis technology, observation method technology, and sea surface platform technology have made it possible to perform horizontal and vertical measurements with centimeter accuracy on an annual scale. This paper introduces the author's research and GNSS-A history, and introduces how GNSS-A contributes to slow & fast earthquake measurements.