Abstract We use Columbia Glacier as a case study to demonstrate how seismically cataloged calving events in Alaska can track glacier evolution over two decades. By combining this catalog—a serendipitous byproduct of earthquake monitoring—with terminus positions, bathymetry, glacier thickness, surface velocity, and environmental records, we show how this seismicity responds to changes in the glacier. We find that fjord bathymetry at the terminus is a strong control on seismicity. As the glacier recedes into shallow water, the number of glacier quakes increases due to the transition to grounded conditions, which favor serac failure. We find evidence that glacier speed, precipitation, and warm ocean temperatures raise the rate of cataloged glacier quakes. We also observe that the energy of individual quakes increases with the elevation of the terminus. Our results highlight potential future uses of the cataloged calving record in Alaska that spans more than a dozen other glacier systems.

Large, rapid landslides are a global hazard that can occur in remote, mountainous areas. Eyewitness reports of landslides and satellite imagery can often be limited or delayed, particularly during inclement weather. However, landslide-generated seismic and infrasound (low-frequency atmospheric sound) waves can be remotely detected in near real-time. This information can significantly expedite characterization and possible landslide response activities. Here, we highlight these capabilities using a 6.1 million m3 ice-rock avalanche in Denali National Park and Preserve (Alaska). This event was detected via a landslide-specific seismic location and volume estimation algorithm deployed in Alaska, and-notably-by standard earthquake monitoring systems. Following rapid detection of this event, we combined its seismic and infrasound data set with optical, synthetic aperture radar, and oblique aerial imagery, multitemporal digital elevation models, and a numerical flow model to reconstruct its failure timeline and dynamics. We apply array processing to infrasound signals traveling >250 km and find that two precursory events occurred minutes prior to the main failure. We use long-period seismic signals to infer the force exerted by the landslide on the Earth and constrain the rheological parameters of our numerical flow simulation with this result and deposit morphology. The main failure produced a steeply dipping impulsive initial downward force and reached speeds exceeding 60 m/s. This impulsive force generated relatively strong seismic body waves, which contributed to the earthquake system detection. This large, remote Alaska landslide underscores the key value of seismic and infrasound analysis for rapid landslide assessment and motivates efforts to further operationalize these approaches.

The Fairbanks region of central Alaska is part of a broad zone of intraplate crustal deformation, situated north of the Denali fault and north of the ongoing collision and flatslab subduction of the Yakutat oceanic plateau. Seismicity in the Fairbanks region occurs both in diffuse areas as well as in well-defined lineaments, such as the left-lateral Salcha fault, which hosted the 1937 Ms 7.3 earthquake. Starting with the regional seismicity catalog, we perform waveform cross-correlation, network-matched filtering, and relative relocation to obtain an enhanced seismicity catalog over the time period 2014- 2024. Based on the relocated catalog, we interpret a set of 15 fault segments, including two conjugate faults and two new faults east of the previously documented fault system. Considering the combined seismicity in the Minto and Fairbanks regions, the median depth of seismicity decreases from east (6 km) to west (20 km). Our interpreted faults provide guidance for future tectonic modeling and assessment of seismic hazards in this region.

Early in the morning of 10 August 2025, a >64 × 106-cubic meter landslide struck Tracy Arm fjord in Alaska. The landslide was preconditioned by glacial retreat caused by climate change. The resulting 481-meter runup megatsunami followed an initial 100-meter-high breaking wave traveling at >70 meters per second. The landslide was preceded by several days of microseismicity, which increased in rate and magnitude until ~1 hour before failure. The landslide produced globally observed long-period seismic waves equivalent in size to a moment magnitude 5.4 earthquake. A long-period (~66 second) global seismic signal, produced by a landslide-induced seiche trapped within the fjord, persisted for up to 36 hours, the second time a days-long seiche had thus been observed. With fjord regions increasingly visited by cruise ships, and climate change making similar events more likely, this unanticipated, near-miss event highlights the growing risk from landslides and tsunamis in coastal environments.

The Barry Landslide, located in Barry Arm of Prince William Sound, Alaska, poses a major hazard due to its steep, unstable slopes, and the potential for a massive landslide-generated tsunami. With an estimated volume of 500–700 million cubic meters, the Barry Landslide could trigger highly destructive waves. In this study, we focus on seismic signals from the Barry Landslide, which are critical for providing timely tsunami warnings. Since the summer of 2020, the region has been instrumented to monitor the landslide, but the seismic record is complicated by the presence of nearby glaciers and frequent regional earthquakes. Among these signals, we analyze a specific class of short-duration, high-frequency seismic events that exhibit strong seasonal variability, increasing in rate from late summer to midwinter before ceasing abruptly in late winter or early spring. Our analysis suggests that the source of these signals is likely near or beneath Cascade Glacier, adjacent to the landslide, rather than within the landslide mass itself. We apply detection algorithms to construct a time history for this signal type, which we then compare with environmental factors like precipitation, temperature, and slope displacement data from ground-based radar and remote sensing. Correlations indicate that these seismic events may be driven by seasonal hydrological changes, particularly the freeze-up of subglacial water pathways. Although these events are not directly linked to landslide motion, they serve as indirect markers of subsurface hydrological conditions that influence slope stability. Our findings highlight the complex interplay between glaciers, groundwater, and landslide dynamics, emphasizing the need for multiparameter monitoring to assess evolving geohazards in the region.

The Minto Flats fault zone (MFFZ) in central Alaska is a left-lateral strike-slip fault system situated between the continental-scale right-lateral Denali and Kaltag-Tintina faults. The MFFZ has the potential to generate magnitude 7 earthquakes, and it hosted a magnitude 6 earthquake in 1995. It has also produced exotic events, such as very-low-frequency earthquakes and nucleation signals. We use network-matched filtering and relative earthquake relocation techniques to derive a detailed catalog of earthquake locations for the MFFZ. The catalog spans from August 2014 to December 2019, a time period including 13 temporary seismic stations in the region. Our results provide the most complete catalog for the MFFZ and include deeper events, clusters of shallow seismicity, and a complex and segmented fault structure not observed in the original regional catalog. We document right-lateral strike-slip faulting, conjugate to the main northeast-striking left-lateral faults of the MFFZ. Below Nenana basin, the relocated seismicity reveals northwest-dipping left-lateral faults, supporting the inference that deep crustal active faulting is associated with recent basin deformation.

Alaska's coastal communities face growing landslide hazards owing to glacier retreat and extreme weather intensified by the warming climate, yet hazard monitoring remains challenging. As part of ongoing experimental monitoring in Prince William Sound, we detected three large landslides (0.5-2.3 M m3) at Surprise Inlet on 20 September 2024, within the span of an hour. These events were identified in near real-time through seismic data and later confirmed using satellite imagery, tidal records, and infrasound. The landslides generated a modest tsunami, and a 4 cm wave was recorded by a tide gauge 18 km away, marking the first recorded landslide to reach water since monitoring began in this region in 2021. Here, we examine the detection and interpretation of these landslides using multiple data sources and modeling. We demonstrate the effectiveness of this regional seismic monitoring system and show how complementary instrumentation, where available, can enhance detection capabilities.

Using 155 distributed seismic stations spanning Alaska and western Canada, we document how environmental factors like storms and sea ice influence microseismic noise. We examine power spectral densities of continuous seismic data and focus on secondary microseisms (5–10 s) and short period secondary microseisms (1–2 s) from 2018 to 2021. We cross‐correlate the height of ocean waves across the region with the power spectral density time series. We find that the Gulf of Alaska is the dominant source of secondary microseisms in Alaska. The eastern Gulf, in particular, produces more energetic secondary microseisms despite, at times, lower overall wave amplitudes. We find that the short period secondary microseismic noise is produced in the coastal waters and attenuates quickly moving inland. We show that this band is heavily modulated by the influence of sea ice in the coastal ocean by comparing it with sea ice concentrations. We also document how these two microseismic bands vary seasonally and spatially as they respond to different environmental phenomena. We find that this seismic energy closely tracks the seasonal arrival and departure of sea ice in the coastal waters. We also compare the inter‐annual variability of short period secondary microseisms in the northern Arctic from 2009 to 2023 with shorefast ice data. The findings of this study are crucial for monitoring global climate change through seismology.

Machine learning (ML) earthquake phase detection algorithms continue to gain popularity and are routinely used to generate research catalogs with thousands of previously uncatalogued events. Many ML algorithms are available pretrained on large, global data sets. These pretrained models promise regional transferability and applications in real-time monitoring organizations. However, the adoption of these ML algorithms by monitoring agencies requires trusted performance across a wide range of seismic monitoring challenges. We apply a pretrained algorithm to four characteristic studies representing a range of network, tectonic, and environmental challenges. We establish a three-catalog comparison framework between our ML catalog, a real-time catalog, and an analyst reviewed catalog. We visually assess and label all ML and real-time events not found in our analyst reviewed catalog. Finally, we subset all additional events that match our catalog standards, establishing a one-to-one performance comparison between our ML and real-time algorithmic catalogs. For each study, we find our ML catalog provides a consistently higher match to our analyst reviewed catalog than the real-time catalog. However, the ML catalog from each study introduces additional complexities ranging from the large addition of poorly constrained, smaller magnitude events; misidentification of nonearthquake signals; and missed detection of large magnitude, felt earthquakes. These discrepancies warrant further training data set scrutiny and suggest that the establishment of a location-based training data set is necessary for consistent and reliable ML performance.

Alaska presents several challenges for earthquake early warning (EEW) systems. These include the presence of offshore earthquakes, transform boundaries, and crustal faults extending hundreds of kilometers, deep earthquakes, and a complicated coastline. This variety, combined with population centers spread far apart, makes it challenging to anticipate early warning performance and design systems accordingly. As Alaska begins to plan for early warning, we present here a set of scenarios intended to inform these activities. Our objective is to envision how, and how well, early warning will function in Alaska. We present warning time estimates for groups of deterministic earthquake scenarios along known faults. These scenarios demonstrate how changes in source characteristics, such as magnitude, depth, location, and fault style, impact the timeliness of warnings and associated ground motions. We combine source time models and travel times for the current seismic network to model detection and alert times. We compare the resulting warning times and peak ground motions to determine the warning effectiveness. Our results demonstrate that even the current network geometry is theoretically capable of providing 0-20 s of warning for intensity 6 for many shallow crustal earthquakes. Increasing the station density can add another 5-15 s to these warning times and provide positive warning times for shaking of intensity 8. Deep and offshore earthquakes benefit less directly from increased station density. For these earthquakes, it is theoretically possible to provide positive warning times for most shaking of intensity 8. Informed by these results, we discuss where we expect an EEW system to excel and what challenges should be tackled to improve other areas.

SUMMARY This study examines the feature space of seismic waveforms often used in machine learning applications for seismic event detection and classification problems. Our investigation centres on the southern Alaska region, where the seismic record captures diverse seismic activity, notably from the calving of marine-terminating glaciers and tectonic earthquakes along active plate boundaries. While the automated discrimination of earthquakes and glacier quakes is our nominal goal, this data set provides an outstanding opportunity to explore the general feature space of regional seismic phases. That objective has applicability beyond ice quakes and our geographic region of study. We make a noteworthy discovery that features rooted in the spectral content of seismic waveforms consistently outperform statistical and temporal features. Spectral features demonstrate robust performance, exhibiting resilience to class imbalance while being minimally impacted by factors such as epicentral distance and signal-to-noise ratio. We also conduct experiments on the transferability of the model and find that transferability primarily depends on the appearance of the waveforms. Finally, we analyse misclassified events and find examples that are identified incorrectly in the original regional catalogue.

Seismic data contains a continuous record of wind influenced by different factors across the frequency spectrum. To assess the influences of wind on ground motion, we use colocated wind and seismic data from 110 stations in the Alaska component of the EarthScope Transportable Array. We compare seismic probability power spectral densities and wind speed and direction during 2018 to develop a quantitative measure of the seismic sensitivity to wind. We observe a pronounced increase in seismic energy as a function of wind speed for almost all stations. At frequencies below the microseism band, our observations agree with previous authors in finding that sensor emplacement and ground materials are important, and that much of the wind influence likely comes from associated changes in barometric pressure. Wind has the least influence in the microseism band, but that is only because its contribution to noise is much smaller than the ubiquitous microseism background. At frequencies above the microseism band, we find that wind sensitivity is correlated with land cover type, increasing with vegetation height. This sensitivity varies seasonally, which we attribute to snow insulation, the burial of vegetation and objects around the station, and potentially the role of frozen ground. Wind direction also manifests in seismic data, which we attribute to turbulent air on the lee side of station huts coupling with the ground and the seismometer borehole cap. We find some dependence on bedrock type, with a greater seismic response in unconsolidated sediment. These results provide guidance on site selection and construction, and make it possible to forecast seismic network performance under different wind conditions. When we examine the factors at work in a warming climate, we find reason to anticipate increasing seismic noise from wind in the Arctic over the decades to come.

As glaciers retreat, landslide-driven tsunamis pose mounting threats across the high latitudes. The recent landslide tsunamis in Alaska and Greenland have spurred efforts to forecast and monitor these events. We use nine large landslides spanning southern Alaska to test an approach for rapid detection and characterization. We use long-period seismograms recorded within three minutes of the start of a landslide to estimate the location and approximate volume. In the presence of good seismic network coverage, location errors are no more than a few kilometers, and detection limits are well below 1 Mm3. The combination of detection time, location, and size provides the ability to rapidly determine whether a landslide occurred close to open water and, if so, its tsunamigenic potential. Our approach is rapid enough to support National Oceanic and Atmospheric Administration (NOAA)’s five-minute tsunami warning goal. The historical analysis we present provides the foundation and parameter tuning for a prototype system that is now providing real-time detections.

Our warming climate is adversely affecting cryospheric landscapes via glacial retreat, permafrost degradation, and associated slope destabilization. In Prince William Sound, Alaska, the rapid retreat of Barry Glacier has destabilized the slopes flanking the glacier, resulting in numerous landslides. The largest of these landslides (∼500 Mm 3 in volume) is more than 2 km wide and has the potential to generate a tsunami that could affect nearby recreationists, marine traffic, infrastructure, natural and cultural resources, and the community of Whittier, located 60 km from the landslide. Here, we combine landslide structural and kinematic element mapping with data acquired from bi‐yearly airborne lidar, multi‐week satellite‐based synthetic aperture radar (SAR), sub‐hourly ground‐based SAR, and seismic monitoring from 2020 to 2022 to characterize this landslide and examine its evolution. While some methods serve as a snapshot in time that is a culmination of events, others emphasize the ever‐evolving nature of the landslide and associated hazards. Four major kinematic elements define the overall structure of the landslide, which vary in deformation type and rate, from creep (5 mm per day over several months) to episodic movement (2 m in 30 days) and landslide‐wide to localized events. In some areas of the landslide, short‐term deformation deviates from structures formed by cumulative movement, implying structural and kinematic evolution associated with glacier retreat. These insights are important for assessing landslide hazards and hazard evolution for large, slow‐moving bedrock landslides in actively deglaciating environments.

The nearest term pathway to the deployment of a seismometer on Venus is an instrument that can operate under ambient surface conditions on battery power. We conduct a series of studies on combined hardware and software approaches to maximize the quality of data returned under the likely restrictions of minimal on-board data storage and only being able to transmit in real time during a small fraction of a multimonth deployment. We assess likely Venus seismicity by examining different terrestrial analog settings; we find that likely Venus analog settings all fall within about an order of magnitude of mean Earth in terms of seismicity level. We use the seismic record from a station in central Alaska as a Venus surrogate for algorithm development. We tested various transmission triggers and developed a simple low-memory algorithm that mimics the common terrestrial long-term average/short-term average trigger. If the seismometer can operate in coordination with an orbiter that can remotely turn off data transmission, then the frequency content of a few seconds of data can be used to distinguish small, nearby earthquakes from large, distal ones, and total data transmission can be tuned to favor the latter. If an orbiter can also turn on transmission for other nearby seismometers, it would further enhance the ability to distinguish small- and large-magnitude earthquakes autonomously and increase the chances of capturing the initial onset of significant events.

Earthquake magnitude estimation using peak ground velocities (PGVs) derived from Global Navigation Satellite Systems (GNSS) data has shown promise for rapid characteri-zation of damaging earthquakes. Here, we examine the feasibility of using GNSS-derived velocity waveforms as interchangeable data for rapid magnitude and ground motion esti-mation that typically rely on strong-motion seismic records. Our study compares PGVs derived from high-rate GNSS to those computed from high-rate seismic records (strong -motion and velocity) at collocated and closely located stations. The recent 2021 M-w 8.2 Chignik earthquake in Alaska that was recorded on collocated GNSS and strong-motion sensors provides the perfect opportunity to compare the two data streams and their appli-cation in rapid response. The Chignik velocity records appear almost identical at collocated GNSS and strong-motion stations when observed at frequencies <0.25 Hz. GNSS and strong-motion derived velocity data are further employed to generate rapid estimates of PGV-derived moment magnitudes for the earthquake. The moment magnitude esti-mates from GNSS and joint GNSS and joint (GNSS and seismic) data are within similar to +/- 0.4 mag-nitude units (Fang et al., 2020) of the final magnitude (M-w 8.2). ShakeMaps generated for the 2021 Chignik earthquake using GNSS and seismic PGVs show notable agreement between them, and show negligible shifts in PGV contours when collocated and closely located GNSS and seismic stations are substituted for one another. Therefore, we posit that GNSS is a powerful alternative or addition to seismic data and vice versa.

The addition of 108 infrasound sensors—a legacy of the temporary USArray Transportable Array (TA) deployment—to the Alaska regional network provides an unprecedented opportunity to quantify the effects of a diverse set of site conditions on ambient infrasound noise levels. TA station locations were not chosen to optimize infrasound performance, and consequently span a dramatic range of land cover types, from temperate rain forest to exposed tundra. In this study, we compute power spectral densities for 2020 data and compile new ambient infrasound low- and high-noise models for the region. In addition, we compare time series of root-mean-squared (rms) amplitudes with wind data and high-resolution land cover data to derive noise–wind speed relationships for several land cover categories. We observe that noise levels for the network are dominated by wind, and that network noise is generally higher in the winter months when storms are more frequent and the microbarom is more pronounced. Wind direction also exerts control on noise levels, likely as a result of infrasound ports being systematically located on the east side of the station huts. We find that rms amplitudes correlate with site land cover type, and that knowledge of both land cover type and wind speed can help predict infrasound noise levels. Our results show that land cover data can be used to inform infrasound station site selection, and that wind–noise models that incorporate station land cover type are useful tools for understanding general station noise performance.

The Lazufre Volcanic System (LVS), on the border of northern Chile and Argentina, is an active complex of two volcanoes, Lastarria to the north and Cordón del Azufre to the south. The LVS is not regularly monitored with any scientific equipment despite being recognized as a top ten volcanic hazard in Argentina by the Observatorio Argentino de Vigilancia Volcánica of the Servicio Geológico y Minero Argentino. The system has shown unusual inflation signatures observed in InSAR but the level of seismic activity and its spatial and temporal distribution were unknown due to the lack of a permanent local seismic network. The PLUTONS Project deployed eight broadband seismic stations throughout the LVS between November 2011 and March 2013. This study shows event locations and types from November 2011 through March 2012. We analyze 591 seismic events within 20 km of Lastarria. Most events cluster tightly beneath Lastarria and almost no activity is observed beneath Cordón del Azufre or the primary inflation center. All events are reviewed manually, and located using a velocity model that assimilates prior studies and accounts for hypocenters within the edifice up to 5 km above sea level. More than 90% of the resulting hypocenters are shallower than 10 km below sea level. The waveforms have characteristics similar to those observed at many other volcanoes, suggesting five classes of events: volcano-tectonic (VT), long-period 1 (LP1), long period 2 (LP2), hybrid (HY), and unknown (X). Frequency-magnitude analysis reveals distinct b-values ranging from 1.2 for VT events to 2.5 for LP1 events. Based on the spatial distribution of events and the b-values, we infer that seismic activity is driven mainly by movement of fluids and gases associated with the regional magma zones and inflation centers. The seismic activity is energetic at times, and quieter at others, suggesting the presence of episodic magmatic and/or hydrothermal activity, focused at Lastarria. Our findings indicate that the previously observed inflation signals are indeed volcanic in origin. These results also demonstrate the potential for success of a future seismic monitoring system and provide a framework for interpreting the subsequent observations, both of which are critical to assessing the volcanic risk of the northern Chile-Argentina region.

Seismic stations and seismic arrays suffer from unwanted seismic noise as a result of inevi-table population growth and development. This encroachment of noise degrades stations' performance. Moving stations to a quieter location breaks the continuity of historical records and can be logistically complicated. This is especially true for seismic arrays that requires a larger footprint. In this study, we examine the feasibility and merit of an adap-tive denoising algorithm to reduce the impact of persistent anthropogenic noise. We build our algorithm on spectral subtraction techniques that have been commonly applied to speech and audio traces and develop a noise-suppression technique that is tailored for seismic data. Using the continuous wavelet transform, we subtract estimates of the noise in the frequency domain. We evaluate this algorithm on synthetic data, consisting of a set of carefully selected events on a low-noise array environment in Alaska. Then we apply this technique to a seismic array in Turkey known to suffer from persistent anthropogenic noise. Our results on individual seismic traces demonstrate that the noise-suppression technique is quite successful at improving the signal-to-noise ratio of key seismic phases. The strengths of this approach include its intuitiveness, its ability to adapt to changes in the background noise, and the ability to reduce noise while preserving the phase of the original signal-a prerequisite for use in array analysis. When the denoised traces are used for array analysis we do not find the noise suppression to be as effective as it is on indi-vidual traces. We explore a number of reasons why this performance is less than desired. Despite the results of our particular implementation, we demonstrate that the larger fam-ily of spectral subtraction techniques offer considerable adaptability and deserve more attention in the seismic community.