Distributed Acoustic Sensing (DAS) is a rapidly developing technology providing spatially dense data of great value for seismic and acoustic monitoring. This paper presents NORFOX, a dedicated DAS installation in southeastern Norway. NORFOX comprises a dedicated fibre-optic array with a geometry designed for earthquake and explosion monitoring. A defining feature of the site is its co-location with the NORES seismic and infrasound arrays, enabling direct comparison of DAS observations with conventional seismic and acoustic measurements. This multi-sensor configuration provides a controlled environment for benchmarking DAS data, and it paves the way for advanced, multi-physics wavefield modelling. NORFOX records various natural and anthropogenic seismo-acoustic signals, including earthquakes, explosions, aircraft, and thunder. The ultra-dense spatial sampling provided by DAS can complement traditional monitoring networks, including those used for nuclear-test monitoring. NORFOX helps address key challenges in small-yield event detection, phase discrimination, and event classification. We describe the DAS array layout, the design rationale, the co-located instrumentation, and provide an initial characterisation of fibre performance and background noise behaviour. We also discuss practical aspects of operating the site and outline current limitations. The site is further complemented by an all-sky camera and weather station to record visual events such as meteoroids and lightning activity, and to measure environmental changes; both of which can support the interpretation of the DAS measurements. Using representative examples, we demonstrate the capability of NORFOX for DAS-based array analysis and benchmarking against conventional arrays. NORFOX therefore provides a valuable test-bed for developing, validating and comparing future DAS monitoring approaches in seismology and seismo-acoustic sensing.

Distributed Acoustic Sensing (DAS) has emerged as a promising tool for environmental and cryoseismological studies, yet its performance under the extreme conditions of the High Arctic remains poorly documented. Here we report on a multi-season DAS experiment conducted across tundra and glacier environments in Hornsund, Svalbard, using 9 km of fiber-optic cable. The study combines a description of the deployment strategy, instrumentation, and operational constraints with an exploratory analysis of the recorded data to assess the types of cryospheric processes that can be captured with DAS. We document logistical, environmental, and technical challenges and provides guidelines for future experiments, including issues related to coupling, noise sources, cable integrity, and seasonal accessibility. Furthermore, we demonstrate how the dataset can be used for detecting permafrost freezing using noise interferometry, locating icequakes and calving events, as well as monitoring runoff from river-induced seismic noise. The experiment provides a field-based reference for the design and interpretation of future DAS studies in Arctic environments and highlights considerations relevant for long-term cryoseismological monitoring.

Spectral analysis (SA) for image processing, utilizing the Fast Fourier Transform (FFT), computes the 2D power spectrum to capture the amplitude of each frequency component of an image. Recent studies have applied SA on digital elevation models (DEMs) to characterize repetitive and spatially homogeneous landforms in terms of their orientation, frequency, and amplitude. Here, we advance the application of SA by introducing a new preprocessing step and an appropriate windowing function, tailored to analyze heterogenous and complex topographies and derive lineament spatial distributions. The validation of our approach involved two phases: (a) testing on synthetic images, and (b) application to a case study. The synthetic image validation illustrated the length-weighted characteristics of SA-derived rose diagrams and the robustness of the method, evidenced by a 99% similarity across 1,000 synthetically generated lineament networks. The case study consisted of three areas characterized by different topographic patterns within the Oslo region of Norway. The SA-derived results were compared to lineaments automatically extracted using a conventional peak-and-valley seeking algorithm that mimics manual tracing of lineaments inside a 3D map domain. The comparison showed similarity better than 90%. Lastly, we addressed a key pitfall of SA by locating signatures observed in the power spectrum on the map through cross-correlation (CC) of profiles. Although CC results are not consistently perfect, they provide a promising avenue for further development.

On 26 September 2022 two seismic events near the Danish island of Bornholm in the Baltic Sea were detected. The first event with a magnitude Mw 2.3 occurred at 00:03 UTC 40 km east-southeast of Bornholm. The determined location and the origin time of the event are consistent with data of the pressure decrease on one of the Nord Stream 2 pipelines. Another sequence of events occurred 17 hours later at 17:03 UTC around 60 km north-east of Bornholm with a maximum magnitude of Mw 2.7. It consists of three closely successive, but separable, single events. Using relative localisation methods and the gas pressure inside the pipeline recorded at the landing site in Germany, we can assign the epicentres of the three events to the locations of the leaks in the pipelines of Nord Stream 1 and 2. Based on comparable events in the region, which include both tectonic earthquakes and explosions, the explosive character of the investigated Nord Stream events can be verified. Infrasound signals associated with the destruction of the Nord Stream pipelines were recorded at two stations (I26DE in the Bavarian Forest and IKUDE near Kühlungsborn) in Germany. Particularly after the event sequence at 17:03 UTC, distinctive signals were registered whose characteristics indicate an explosive event with subsequent gas leakage at the surface. Our modelling of the sources shows that the measured seismic signals can sufficiently be explained by the instantaneous gas release. Synthetic seismograms for such a source and a subsurface model adapted for the study area show high consistency with the measured signals. Based on the released energy and the characteristics of the recorded waveforms, we conclude that the impulsive gas release from the burst gas pipes constitutes the dominant part of the signal source. The model places an upper limit of approximately 50 kg TNT equivalent on the yield of the chemical explosive component of the events, but we note that smaller yields may also be consistent with the data. We also carried out an analysis of the seismic signals of the event on the Balticconnector pipeline between Finland and Estonia on 8 October 2023 and found that again the instantaneous gas release can sufficiently explain the observed data. This supports a possible mechanical cause of the damage.

Some of the more densely populated areas in Norway are in potential quick clay zones. When disturbed, the structure of quick clay can suddenly collapse, and behave and flow as a liquid, potentially having disastrous impact over large areas. One of the triggering factors for quick clay slides is heavy rainfall. Here, we focus on passive seismic data from two Raspberry shake sensors located in an urban area in Oslo, Norway with quick clay in the subsurface. Using coda wave interferometry, near-surface velocity variations are estimated during the extreme weather ”Hans” (August 2023).We compute auto-correlations and single station cross-correlations of anthropogenic seismic noise (> 1 Hz) over a two-year period leading up to ”Hans”. We observe environmental velocity fluctuations well correlated with air temperature, precipitation and the water level in a nearby river. In particular, freezing and thawing produces strong changes in seismic velocity (up to 4 %). Disregarding freezing, we see the largest change in seismic velocity following the heavy rainfall associated with ”Hans”. This extreme event is associated with a sharp velocity drop anti-correlated with pore pressure. The surface wave-coda is sensitive to changes in shear wave velocity, which in turn can be used to detect changes of the subsurface properties. Therefore, observed velocity variations at the site could have potential for monitoring and early warning of quick clay instabilities.

Sudden glacier acceleration and instability, e.g. surges, strongly influence glacier ice loss. However, lack of in-situ observations of the involved processes hampers our ability to understand, quantify and model such a role. We present an analysis of the initiation of a surge (Kongsvegen glacier, Svalbard), focusing on the interplay between climatic and glacier-specific drivers. We integrate two decades of in-situ observations (GNSS, borehole and surface seismometers) with runoff simulations, and remotely sensed surface-elevation changes. We show that initial glacier thinning led to localized acceleration and crevassing. Then, we show that stronger surface melt enabled meltwater to reach the glacier bed. This input promotes high basal water pressure and glacier sliding, and in turn further surface crevassing. Our observations suggest that this positive feedback leads to the expansion of the initially localized instability. Our findings highlight mechanisms that could trigger glacier instabilities under a warming atmosphere beyond the High Arctic.

Glacier flow instability often results from changes at the ice-bed interface. However, understanding these processes is challenging due to limited access to the glacier bed. Our study focuses on Kongsvegen glacier in Svalbard, which shows signs of an upcoming rapid flow event. To investigate the potential causes of such acceleration, we installed 20 seismometers along the glacier flowline, from the surface down to 350 m near the ice-bed interface. We combined our seismic monitoring with measurements of surface velocity, basal water pressure, and basal sediment deformation. First, we performed seismic noise interferometry between stations located along the glacier flowline with inter-station distances ranging from 1 to 12 km. We observed a multi-year decrease in seismic velocity, with a seasonal signal superimposed, showing a melt-season decrease in seismic velocity of 2 to 4%. We compared our observations with 1D models and concluded on the presence of damaged basal ice and/or a weakening of the subglacial sediments. This indicates a mechanical weakening of the ice-bed interface, promoting further glacier acceleration. Second, we conducted unsupervised clustering of seismic waveforms using a novel approach based on a deep scattering network. Doing so, we observed a yearly increase in surface crevasses concomitant with an increase in basal events, likely indicating stick-slip and/or basal crevasses. This increase is particularly visible during winter, where the number of events steadily increases from year to year. We suggest that, in response to an initial glacier acceleration, new crevasses have opened, providing access pathways for surface meltwater to the base of the glacier, affecting the ice-bed coupling. This mechanism represents a positive hydro-mechanical feedback that fuels further acceleration and crevassing, potentially having wider implications for triggering glacier-wide instabilities, increasing short-term sea-level rise, and local hazards.

Apart from classical earthquake monitoring, seismological data can also be used to detect explosions in near-real-time on both regional and global scales. We demonstrate how seismic and infrasound data can provide more comprehensive and objective information about conflict-related explosions or suspicious events that might be the result of targeted attacks. We can identify the underwater explosions at the Nord Stream pipeline infrastructure in the Baltic Sea in September 2022. Cross-correlation analysis allowed us to identify sub-events several seconds apart which can be associate with specific locations along the pipelines. Furthermore, we detect a signal at the Finish seismic array in October 2023 which may be associated with the damage along the Balticconnector. The other example is from Ukraine, where we present the ability to automatically identify and locate ground explosions related to the Russia-Ukraine conflict with data from the Malin array (AKASG). Between February and November 2022, we observe more than 1,200 explosions from the Kyiv, Zhytomyr, and Chernihiv provinces. Both seismic and infrasound detections can be used to verify and improve accurate reporting of military attacks and help to provide an unprecedented view of an active conflict zone. We analyze events with a variety of seismo-acoustic signatures and significant differences in explosive yield. These can be associated with various types of military attacks, including artillery shelling, cruise missile attacks, airstrikes, or the destruction of the Kakhovka dam NE of Cherson.

Stronger and more widespread surface melt may alter the flow of glaciers and ice sheets and trigger instability. However, observational deficiencies hamper our ability to better understand and thus predict such responses. We deployed surface and borehole seismometers along the centerline of a High Arctic glacier in Svalbard. The records span over six years and are analyzed in relation to the measured increase of surface velocity. We complement our seismic analysis (icequakes and seismic noise) with long-term measurements of glacier-surface velocity, surface-elevation changes, and runoff modeling. Since 2000, we observe glacier thinning and steepening, coinciding with acceleration of up to 1000%. In response, new crevasses have opened and provide access pathways for surface melt water to the base of the glacier, affecting the ice-bed coupling. This mechanism represents a positive hydro-mechanical feedback that fuels further acceleration and crevassing. This feedback may have wider implications for triggering of glacier-wide instabilities, increasing short-term sea-level rise and local hazards. Beyond the Arctic, we suggest that, under a warming atmosphere, glaciers may transition from stable to unstable flow through such a mechanism.

The aim of this study is to collect information about events in the city of Oslo, Norway, that produce a seismic signature. In particular, we focus on blasts from the ongoing construction of tunnels and under-ground water storage facilities under populated areas in Oslo. We use seismic data recorded simultaneously on up to 11 Raspberry Shake sensors deployed between 2021 and 2023 to quickly detect, locate, and classify urban seismic events. We present a deep learning approach to first identify rare events and then to build an automatic classifier from those templates. For the first step, we employ an outlier detection method using auto-encoders trained on continuous background noise. We detect events using an STA/LTA trigger and apply the auto-encoder to those. Badly reconstructed signals are identified as outliers and subsequently located using their surface wave (Rg) signatures on the seismic network. In a second step, we train a supervised classifier using a Convolutional Neural Network to detect events similar to the identified blast signals. Our results show that up to 87% of about 1,900 confirmed blasts are detected and locatable in challenging background noise conditions. We demonstrate that a city can be monitored automatically and continuously for explosion events, which allows implementing an alert system for future smart city solutions.

Seismic phase detection and classification using deep learning is so far poorly investigated for regional events since most studies focus on local events and short time windows as the input to the detection models. To evaluate deep learning on regional seismic records, we create a data set of events in Northern Europe and the European Arctic. This data set consists of about 151 000 three component event waveforms and corresponding phase arrival picks at stations in mainland Norway, Finland and Svalbard. We train several state-of-the-art and one newly developed deep learning model on this data set to pick P- and S-wave arrivals. The new method modifies the popular PhaseNet model with new convolutional blocks including transformers. This yields more accurate predictions on the long input time windows associated with regional events. Evaluated on event records not used for training, our new method improves the performance of the current state-of-the-art methods when it comes to recall, precision and pick time residuals. Finally, we test our new model for continuous mode processing on 4 d of single-station data from the ARCES array. Results show that our new method outperforms the existing array detector at ARCES. This opens up new opportunities to improve automatic array processing with deep learning detectors.

Abstract Seismometers are generally used by the research community to study local or distant earthquakes, but seismograms also contain critical observations from regional1,2 and global explosions3, which can be used to better understand conflicts and identify potential breaches of international law. The large overpressure generated by an explosion, shakes the Earth’s atmosphere and subsurface, and the resulting ground motion can be recorded by seismometers, while the infrasound signals that propagate through the atmosphere can be detected by microbarometers. While this technology is used by the International Monitoring System4 to monitor nuclear explosions as part of the Comprehensive Nuclear Test Ban Treaty, the detection and location of lower-yield military attacks requires a network of sensors much closer to the source of the explosions. Without dedicated sensor deployments, such networks are rarely available for conflict monitoring. Obtaining comprehensive and objective data that can be used to effectively monitor an active conflict zone therefore remains a significant challenge. We demonstrate how seismic waves generated by explosions in northern Ukraine can be recorded by a local network of seismometers and used to automatically identify individual attacks in close to real-time, providing an unprecedented view of an active conflict zone. Between February and November 2022, we observe over 1200 explosions from the Kyiv, Zhytomyr and Chernhiv provinces, providing accurate origin times, locations and magnitudes. We identify a range of seismo-acoustic signals associated with various types of military attacks, with the resulting catalogue of explosions far exceeding the number of publicly reported attacks5. Our results demonstrate that seismic data can be an effective tool for providing accurate and objective data in real-time from an ongoing conflict. We anticipate that the implementation of seismic-based monitoring techniques can provide invaluable information about potential breaches of international law.

Seismometers are generally used by the research community to study local or distant earthquakes, but seismograms also contain critical observations from regional1,2 and global explosions3, which can be used to better understand conflicts and identify potential breaches of international law. Although seismic, infrasound and hydroacoustic technology is used by the International Monitoring System4 to monitor nuclear explosions as part of the Comprehensive Nuclear-Test-Ban Treaty, the detection and location of lower-yield military attacks requires a network of sensors much closer to the source of the explosions. Obtaining comprehensive and objective data that can be used to effectively monitor an active conflict zone therefore remains a substantial challenge. Here we show how seismic waves generated by explosions in northern Ukraine and recorded by a local network of seismometers can be used to automatically identify individual attacks in close to real time, providing an unprecedented view of an active conflict zone. Between February and November 2022, we observed more than 1,200 explosions from the Kyiv, Zhytomyr and Chernihiv provinces, providing accurate origin times, locations and magnitudes. We identify a range of seismoacoustic signals associated with various types of military attack, with the resulting catalogue of explosions far exceeding the number of publicly reported attacks. Our results demonstrate that seismic data can be an effective tool for objective monitoring of a continuing conflict, providing invaluable information about potential breaches of international law.

<p>Soon after midnight on 26 September 2022 the Swedish National Seismic Network, using data from Sweden, Denmark and Germany, automatically detected a seismic event in the Baltic southeast of the Danish island of Bornholm. The event was followed 17 hours later by a second, more complex, event northeast of Bornholm. The automatic locations of the events were within 6-9 km of later reported gas leaks in the Nord Stream 1 and 2 pipelines. Using recently developed, machine learning based, classifiers both events were automatically classified as explosions. Subsequent analysis of the second event revealed that it was in fact two blasts, separated by about 7 seconds. As the events occurred in the transition zone between the Fennoscandian Shield and the younger terranes of Denmark and northern Germany, 3D tomographic P- and S-velocity models were developed to improve locations and assess uncertainties, bringing the locations closer to the pipelines. Spectral analysis of the blast data show clear reverberations consistent with underwater explosions and a blast depth of approximately 75 m. The conclusion that the events are underwater blasts are further supported by data on known underwater explosions and a few earthquakes in the area. The magnitude of the first event was estimated at ML 1.9 and the combined second and third event had ML 2.3. Estimating the equivalent yield in the explosions is, however, non-trivial. Comparison to ground truth underwater explosions suggests yields of one to a few hundred kilos of equivalent TNT. The contribution to the seismic energy from suddenly outflowing methane gas is under investigation and results will be included in the presentation.</p>

Autonomous algorithms can improve the processing of aftershock sequences, for example by reducing the analyst workload. We present a system for automatic detection and location of aftershocks in a specific region following a large earthquake. The system seeks to identify all signals generated by seismic events in the target region, while passing over signals generated by sources in all other regions. For a given station, we can generate a sensitive empirical matched field (EMF) detector for the target region using only an empirical template from the mainshock signal. These EMF detectors perform much better on seismic arrays than on 3-component stations. For each selected station in the network, a multivariate detector combines the EMF detector with an optimized continuous AR-AIC detector to generate a target-optimized detection list. For arrays, an additional continuous calibrated f-k process reliably screens out likely signals from other sources. A region-specific phase association algorithm takes the screened detection lists from each station and generates a preliminary aftershock bulletin. We have processed aftershock sequences from four major earthquakes: the Tohoku event in 2011 (Japan), the Illapel event in 2015 (Chile), the Papua New Guinea event in 2018 and the Gorkha event in 2015 (Nepal). We evaluate the results in detail by comparing the automatically generated origins and corresponding phase arrival times with matching events and associated arrivals in the analyst reviewed (REB) and automatic (SEL3) bulletins issued by the CTBTO Preparatory Commission. Between 40% and 65% of all events in the REB are found to closely match the locations and origin times of the events found by our EMF-based procedure. The resulting discrepancies are assessed with respect to signal-to-noise ratio, number of defining stations, and epicentral distance. Furthermore, the REB events not detected by the EMF method are analyzed and a few phase misidentifications (i.e., P vs. pP) are assessed to better understand the limitations of the autonomous procedure. In general, we find that our EMF solutions are closer to the matching REB events than the corresponding SEL3 events. The analyst is helped both by the improved location estimates and a lower number of qualitatively incorrect event hypotheses. A key factor in the performance is the number of contributing seismic arrays. Aftershock sequences in the southern hemisphere performed the worst given the poorer array coverage.

<p>Since the invasion of Ukraine in February 2022, daily media reports have shown the shocking effects of fighting and the inevitable devastation associated with war. However, getting a comprehensive and unbiased overview of the ongoing military attacks remains a challenge. The availability of geophysical data that can identify individual attacks provides much needed objectivity to this problem. The pressure waves generated by an explosion travel through the atmosphere and subsurface as sound and seismic waves, and their signature can be recorded by arrays of seismometers for ground motion or microbarometers for sound propagation. In this work, we demonstrate the first known case of using seismological data to detect conflict-related explosions in near-real-time. Using the Ukrainian primary station of the International Monitoring System (IMS), the Malin array (AKASG), we automatically locate explosions around the Kyiv and Zhytomyr provinces. We show how our resulting catalogue of explosions correlates with key events in the Ukraine conflict and how these data can be used to both verify and improve accurate reporting of military attacks. We analyze events with a variety of seismo-acoustic signatures and significant differences in explosive yield. These can be associated with various types of military attacks, including artillery shelling, cruise missile attacks and airstrikes. This work opens-up the possibility for future conflict monitoring using geophysical data.</p>

<p>The real-time seismo-acoustic monitoring of military conflicts can provide a unique alternative to conventional ground reports and sparse satellite coverage. The pressure waves generated by an explosion travel through the atmosphere and subsurface as sound and seismic waves, and their signature can be recorded by arrays of seismometers for ground motion or microbarometers for sound propagation. However, standard monitoring techniques can be both computationally expensive when localizing signals over large regions and/or prone to false detections when signals have low amplitudes. In this contribution we propose a Machine-Learning (ML) based solution to detect seismic and infrasound arrivals and locate sources close to real time. To validate our model we leverage the seismic data collected during the Russia-Ukraine conflict started in February 2022 using the Ukrainian primary station of the International Monitoring System (IMS), the Malin array (AKSAG). We test both the accuracy and computational efficiency of our approach against a threshold-based migration stacking model developed for near-real time monitoring in Ukraine. We hope that this first-ever ML detector of both seismic and acoustic phases could be employed for real-time monitoring of conflicts around the world across different network geometries and noise conditions.</p>

<p>Two clear seismic events were observed on 26<sup>th</sup> September 2022 associated with the reported leaks from the Nord Stream 1 (Event 1, NE of Bornholm) and Nord Stream 2 pipelines (Event 2, SE of Bornholm). Arrivals of both events were detected and associated using data from several arrays in Norway and Finland, including the IMS stations NOA, FINES and ARCES. Additional signal analysis with data from the Swedish National Seismic Network and the Danish station on Bornholm enabled a third event to be identified. Auto-correlation analysis of the Event 2 revealed the third event (Event 2B) about 7 seconds after the main amplitude of the P onset (Event 2A). In contrast, for Event 1 SE of Bornholm no additional events could be identified from auto-correlation analysis, which increases confidence that these additional arrivals are not caused by interaction with geological structures. We also observe an arrival 7 s after the Pn phase before the Pg arrival on the NORES array. However, we cannot exclude that this onset interferes with the arrival of the PnPn phase. We then use the time differences between Event 2A and 2B measured by auto-correlation analysis on the Swedish and Danish network stations to determine relative epicentre locations. The results suggest that the two overlapping events occurred just about 220 m apart from each other. The relative locations fit very well with the distance between both pipelines of Nord Stream 1 at the Westernmost gas plume location (NE of Bornholm). We also estimated preliminary full moment tensors for Event 1 and 2 using seismic waveform data and analysed them on a source-type diagram. The results show positive isotropic parameters consistent with explosion-type mechanisms.</p>

In order to estimate well-constrained seismic hazard and risk on local scales, the knowledge of site amplification factors is one of several important requirements. Seismic hazard studies on national or regional scales generally provide the level of earthquake shaking only at bedrock conditions, thereby avoiding the difficulties that are caused through local site effects. Oftentimes, local site conditions are not well understood or even non-existent. In this study we investigate an efficient and non-invasive methodology to derive the local average shear wave velocity in the uppermost 30 m of the ground (Vs30). The Vs30 value is a useful parameter to define soil classes and soil amplification used in seismic hazard assessment and to extend the knowledge of the site to include the depth to basement rock. At the level of the municipality of Oslo, there is currently no map available that describes the Vs30, and as such any seismic risk study is lacking potentially critical information on local site amplification. The new proposed methodology includes the use of existing well databases (with knowledge on minimum basement depth), topographic slope derived from Digital Elevation Models (as a proxy for both depth to basement and Vs30, integrated with geological maps) and near-surface Quaternary geological maps. The Horizontal to Vertical Spectral Ratio (HVSR) method and a statistics-based geological mapping tool (COHIBA) are used to integrate the various sources of data estimates. Finally, we demonstrate our new methodology and workflow with data from three different regions within the Oslo municipality and propose an approach to conduct cost-efficient mapping for seismic site amplification on a general municipality scale.

Prior to planned CO2 injection startup in the Horda platform offshore western Norway, in 2024, the Horda Network project has taken several measures to assess the potential of seismic hazard in the area. A study of the fault-plane solutions in the Horda platform region confirms that the direction of maximum horizontal stress is dominantly northwest–southeast to east–west over the entire area. The relative stress ratio is higher in the southeast near the Norwegian craton and lower in the northwest. Analysis of the catalog of seismicity (in the period of 2001–2021) in the Horda platform region suggests a moderate rate of seismicity with a b-value of ∼1. The magnitude of completeness is 1.5 (ML). One of the main challenges in monitoring offshore earthquakes in the Norwegian continental shelf (NCS) is the lack of azimuthal coverage when using the onshore permanent seismic stations from the Norwegian National Seismic Network (NNSN), located to the east of offshore events. To improve the azimuthal coverage, we integrated a limited number of offshore geophones from permanent reservoir monitoring systems of selected oil and gas fields (Grane and Oseberg on NCS) with the onshore NNSN seismic stations. This integration is challenging because of the level of ambient noise in the offshore geophones. To further improve the detection and location capability, we deployed a nine-element onshore array of broadband seismometers (HNAR) on Holnsnøy island to the east of the Horda platform. By incorporating array processing methods on HNAR, the signal-to-noise ratio is improved, and several previously uncataloged earthquakes could be detected. Offshore sensors are often subject to correlated noise from seismic interferences and platform or shipping noise sources, so we also incorporated array processing for selected geophones from offshore deployments, which greatly reduced such noise and hence improved the event detection.