Tectonic tremor monitoring occasionally detects events in an anomalous zone in southern Cascadia, 50-100 km west of the main tremor band, near the expected southern edge of the subducting Gorda slab at the Mendocino triple junction. To investigate the geometry and temporal behavior of this tremor, we examine its constituent low-frequency earthquakes (LFEs) by developing 27 stacked LFE waveform templates that we use to detect events from 2018 to 2024. We then relocate LFE sources together with regional seismicity. We find that LFE hypocenters form a northeast-dipping alignment at 22-29 km depth, extending eastward from a zone of micro-earthquakes, similar to 15 km south of the southern edge of Gorda slab seismicity. These LFE families exhibit small bursts of activity every few days. Considering the strong world-wide association of tremor and LFEs with high slip-rate, plate-bounding faults, we hypothesize these LFEs may demark the southern edge of Cascadia subduction.

The 2022 Mauna Loa eruption began on November 27 23:21 HST (November 28 9:21 UTC) at the summit, based on the onset of seismic tremor and visual observations of nighttime incandescence from lava viewed by webcameras. Continued visual observations noted early southwestward migration of the summit flows followed by dike propagation and new fissures on the northeast rift zone in the early hours of November 28th HST. Northeast rift zone activity subsequently settled into persistent activity primarily from Fissure 3 (located about 7 km from the Summit) until the eruption stagnated by December 9th HST.This contribution provides a detailed retrospective assessment of the performance of four Hawaiian Volcano Observatory (HVO) infrasound arrays (AIND, AHUD, MENE, SHEEP) and an International Monitoring System array operated by the University of Hawaii Infrasound (I59US) data during the onset and progression of the eruption. We use results from a standard least-squares beamforming algorithm which is widely used for infrasound processing across the USGS Volcano Science Center and compare other multidisciplinary observations such as visual and seismic amplitude.We find that that the standard array processing approach performed adequately as a real-time assessment tool with high correlation back-azimuth computations in reasonable agreement with visual observations. High winds associated with storms impacted the quality of our results, particularly toward the end of the eruption when infrasound signals were comparably small. It was possible to distinguish between summit and rift fissure activity using the three long-term arrays operated by HVO, despite the large source to receiver distance. The SHEEP array (on the south flank of Mauna Kea) was established in response to the Mauna Loa eruption and only recorded the waning phase of the eruption. Regardless, the new array should further improve azimuthal coverage for future Mauna Loa eruptive activity.

A seismic swarm near Taʻū Island, a volcanic island in eastern American Samoa, occurred from July to October 2022. The earliest unrest was noted as felt shaking reports in late July, and instrumentation varied in the beginning of the sequence as the U.S. Geological Survey (USGS) Hawaiian Volcano Observatory responded by installing temporary and then permanent seismometers to monitor the activity. This network variability made it difficult to characterize the earliest seismicity and contextualize the entire sequence to discriminate between an underlying tectonic or volcanic source. Here, we present results analyzing hydroacoustic detections from an International Monitoring System hydrophone array near Wake Island, 4500 km northwest of Taʻū Island volcano. Using least-squares beamforming analysis, we create a catalog of T-wave detections from the direction of Taʻū Island to track the earthquakes, some of which were located by the USGS National Earthquake Information Center. Both the rate and hydroacoustic pressures, which we interpret as a proxy for earthquake size, gradually increased from late July to August, peaking on August 19 (rate) and August 24 (size), before decreasing to background in late September. Minutes-long bursts of tremor were also contemporaneously recorded as local network data became available on August 20. Tremor activity continued throughout the rest of August, peaking on August 25, before ending in early September. These tremor bursts were band-limited to 1–5 Hz and recorded as S waves at a regional station on the island of Upolu in Samoa, 250 km to the west of Taʻū Island. Our results do not constrain the tremor locations, but comparisons of earthquake and tremor reduced displacements recorded locally and regionally suggest a deeper tremor source. We interpret the increase in earthquake size and rate, together with the occurrence, characteristics, and relative depth of the tremor to be the result of magmatic activity beneath Taʻū Island volcano.

Cascadia slow slip events (SSEs) are often envisioned as smooth, continuous ruptures, progressively activating tremor asperities as they propagate. Macroscopically, geodetic inversions and spatiotemporal maps of tremor epicenters show steady, uniform migration. In detail tremor is more chaotic and discontinuous. Larger long-term SSEs observed in daily geodetic solutions are inferred to exhibit intermittent pauses that reflect temporary re-locking of the fault, but this temporal resolution limits tests for similar re-locking on shorter timescales. We use temporal measurements of the areal growth and radiated energy of tremor clusters to investigate SSE intermittence. We find that ruptures mirror tremor pauses. Areal growth rate, however, does not reset, and removing the pauses results in smoother and more similar growth measurements among all SSEs. The rupture similarity occurs regardless of size or location and hints at an underlying uniformity and lack of predeterminism in eventual SSE size. Epicentral uncertainty precludes quantifying early rupture stages, but for larger events areal growth follows a power-law and slows with increasing size. Temporal correlations in tremor energy with inferred SSE propagation velocities and tremor rates suggest its use as a proxy for slip velocity. We find that tremor energy is tidally modulated at daily and sub-daily frequencies, and this modulation is continuous through pauses, suggesting a memory of slip state is sustained through them. We argue these pauses reflect unsteady propagation of the slip front, marked by rapid re- and un-locking, and excluding them removes rupture complexity to reveal a diffusive-like slip process and underlying universality in growth.

Abstract The fracture of Earth materials occurs over a wide range of time and length scales. Physical conditions, particularly the stress field and Earth material properties, may condition rupture in a specific fracture regime. In nature, fast and slow fractures occur concurrently: tectonic tremor events are fast enough to emit seismic waves and frequently accompany slow earthquakes, which are too slow to emit seismic waves and are referred to as aseismic slip events. In this study, we generate simultaneous seismic and aseismic processes in a laboratory setting by driving a penny‐shaped crack in a transparent sample with pressurized fluid. We leverage synchronized high‐speed imaging and high‐frequency acoustic emission (AE) sensing to visualize and listen to the various sequences of propagation (breaks) and arrest (sticks) of a fracture undergoing stick‐break instabilities. Slow radial crack propagation is facilitated by fast tangential fractures. Fluid viscosity and pressure regulate the fracture dynamics of slow and fast events, and control the inter‐event time and the energy released during individual fast events. These AE signals share behaviors with observations of episodic tremors in Cascadia, United States; these include: (a) bursty or intermittent slow propagation, and (b) nearly linear scaling of radiated energy with area. Our laboratory experiments provide a plausible model of tectonic tremor as an indicative of hydraulic fracturing facilitating shear slip during slow earthquakes.

The Alaska Volcano Observatory (AVO) uses multidisciplinary data to monitor and study dozens of active and potentially active volcanoes. Here, we provide an overview of internally and externally generated data types, tools and resources used in their management, and challenges faced. Data sources include the following: (1) a multiparameter (seismic, infrasound, GNSS, web cameras) ground-based monitoring network that spans 3000 km and transmits data in real time; (2) a variety of satellite-borne sensors that provide information about surface change and volcanic emissions; (3) geologic and gas field campaigns; and (4) other external data products that provide situation awareness. Each data type requires distinct acquisition, processing, storage, visualization, and archiving approaches. AVO uses a variety of externally and internally developed tools to handle individual data types as well as multidisciplinary volcanological data. A primary tool is the Geologic Database of Information on Volcanoes in Alaska (GeoDIVA), which stores detailed, searchable information on more than 140 volcanoes and over 1000 eruptions and unrest events, including images, eruption descriptions, and geologic station and sample data, metadata, and analyses. It interacts with other internal tools that store monitoring reports and other operational records. Additional data management resources used by AVO assist with alarms and alerts, state-of-health monitoring, and multiparameter visualization. Requirements for 24/7 accessibility, the ever-expanding portfolio of data, and transitioning new tools from development to operations are all challenges faced by AVO and other volcano observatories. AVO strives to meet FAIR data practices and ensure that data are available to national and international community efforts using external repositories as well as those hosted by AVO and its parent institutions.

Volcanic tremor is a semi-continuous seismic and/or acoustic signal that occurs at time scales ranging from seconds to years, with variable amplitudes and spectral features. Tremor sources have often been related to fluid movement and degassing processes, and are recognized as a potential geophysical precursor and co-eruptive geophysical signal. Eruption forecasting and monitoring efforts need a fast, robust method to automatically detect, characterize, and catalog volcanic tremor. Here we develop VOlcano Infrasound and Seismic Spectrogram Network (VOISS-Net), a pair of convolutional neural networks (one for seismic, one for acoustic) that can detect tremor in near real-time and classify it according to its spectral signature. Specifically, we construct an extensive data set of labeled seismic and low-frequency acoustic (infrasound) spectrograms from the 2021-2022 eruption of Pavlof Volcano, Alaska, and use it to train VOISS-Net to differentiate between different tremor types, explosions, earthquakes and noise. We use VOISS-Net to classify continuous data from past Pavlof Volcano eruptions (2007, 2013, 2014, 2016, and 2021-2022). VOISS-Net achieves an 81.2% and 90.0% accuracy on the seismic and infrasound test sets respectively, and successfully characterizes tremor sequences for each eruption. By comparing the derived seismoacoustic timelines of each eruption with the corresponding eruption chronologies compiled by the Alaska Volcano Observatory, our model identifies changes in tremor regimes that coincide with observed volcanic activity. VOISS-Net can aid tremor-related monitoring and research by making consistent tremor catalogs more accessible. Volcanic tremor is a persistent vibration of the ground, atmosphere, or both that can occur before and during volcanic eruptions. Despite its importance in volcano monitoring and eruption forecasting, volcano observatories do not have a reliable way of automatically detecting and identifying tremor due to the variable intensities and frequencies at which it occurs. In order to accomplish this, we develop and test a pair of machine learning models that classify spectrograms (i.e., images representing a signal's frequency content over time) from seismic and low-frequency acoustic data. The models are trained on manually labeled images derived from the recent 2021-2022 eruption of Pavlof Volcano, Alaska, which demonstrated substantial signal diversity (e.g., different tremor types, earthquakes, explosions and noise). Our models achieve 81.2% and 90.0% accuracy on the seismic and low-frequency acoustic test sets respectively, and perform well when applied to data recorded from past Pavlof Volcano eruptions. In addition, transitions in tremor sequences identified from our analysis generally coincide with shifts in eruptive patterns from Pavlof Volcano. Our tools can help volcano observatories systematically monitor tremor, and advance tremor research by making catalogs of their occurrences more consistent and accessible. We develop a pair of convolutional neural networks that detect and classify volcano seismic and acoustic signals We apply our models to Pavlof Volcano eruptions (2007, 2013, 2014, 2016, and 2021-2022) and derive volcano seismoacoustic timelines The seismoacoustic timelines reveal shifts in unrest regimes linked to explosions and effusive activity

First posted May 11, 2023 For additional information, contact: Alaska Volcano ObservatoryU.S. Geological Survey4210 University DriveAnchorage, AK 99508 The Alaska Volcano Observatory responded to eruptions, volcanic unrest or suspected unrest, increased seismicity, and other significant activity at 17 volcanic centers in Alaska in 2019. The most notable volcanic activity was an eruption of Shishaldin Volcano, featuring eruptive activity that produced lava flows, lahars, and ash. Weak explosive activity also took place at Great Sitkin Volcano and Semisopochnoi Island. Mount Cleveland had one small ash-producing eruption followed by dome growth in early January but was quiet thereafter, and flank activity at Shrub mud volcano produced new mud deposits. Other activity documented in 2019 consists of declining unrest at Mount Veniaminof after its 2018 eruption; large ice and rock avalanches at Iliamna Volcano and Mount Spurr; anomalous seismicity and an increase in degassing at Pavlof Volcano; long-term inflation at Westdahl volcano, Akutan Volcano, and Mount Okmok; steam plumes and anomalous seismicity at Makushin Volcano; elevated seismicity at Mount Martin; and resuspended ash from the 1912 Novarupta-Katmai eruption deposits.

In April 2022, a seismic swarm near Mt. Edgecumbe in southeast Alaska suggested renewed activity at this transform fault volcano, which was last active ≈800 years ago. Previously, thin rhyolitic tephras were deposited 5 and 4 ka. Satellite radar data from 2014 to 2022 resolves line‐of‐sight rapid inflation up to 7.1 cm/yr beginning in August 2018. Bayesian modeling suggests a transcrustal system of a deflating (−0.528 km3) dipping sill at 20 km depth recharging a magma chamber at 10 km (0.222 km3). A near‐vertical conduit could capture the volume difference without noticeable surface deformation. Reanalyzed seismicity, recorded 25 km away, shows increases since July 2019. Magma ascent through ductile material and brittle strain release in a stressed overburden could explain the time delay. Cloud‐native open data and workflows enabled discovery and analysis of this signal within days after going unnoticed for >3 years.

For the past ∼12 years the Pacific Northwest Seismic Network has been automatically detecting and locating tectonic tremor across the Cascadia subduction zone, resulting in a catalog of more than 500,000 tremor epicenters to date, which has served as a valuable resource for tremor and slip research. This manuscript presents an updated methodology for routine tremor detection in Cascadia and a new catalog of over 180,000 tremor epicenters including amplitudes detected along the subduction zone margin from 2017 to 2021. The events are detected via cross‐correlation of continuous vertical envelope data of 128 stations from northern California to northern Vancouver Island. The modified approach results in less scatter and a 55% increase in detected epicenters than previously observed, as well as a newly identified tremor source offset updip from the main tremor and slip region at the southern edge of the subduction zone. Radiated seismic energy in the 1.5–5 Hz band is used to assign epicenters an energy magnitude (MeL), which is calibrated to the ML of local earthquakes. Southern Cascadia is most active, but the highest tremor energy rates occur in northern Cascadia. Tremor in central Cascadia is systematically weaker and less frequent. Individual epicenter magnitudes range from ∼0.5–2 and spatiotemporally cluster into 1,060 swarms with cumulative MeL ranging from ∼0.8 to 3.7. The swarms reflect underlying slow slip events and occur with an earthquake‐like energy distribution with a b value ∼1. Tremor epicenters, however, follow a tapered Gutenberg‐Richter distribution with high b values, suggesting individual tremor bursts and their constituent low‐frequency earthquakes are fault‐dimension limited.

Bogoslof volcano, Alaska, experienced at least 70 explosive eruptions between 12 December 2016 and 31 August 2017. Due to its remote location and limited local monitoring network, this eruption was monitored and characterized primarily using remote geophysical and satellite techniques. SO2 emissions from Bogoslof were persistently detected by the Infrared Atmospheric Sounding Interferometer (IASI) satellite sensors. Of Bogoslof’s 70 explosive events, 50% produced measurable SO2 masses ranging from 0.1 to 21.5 kt, with a median and standard deviation of 0.7 ± 4.0 kt SO2, respectively. Here, we compare IASI-derived SO2 masses from Bogoslof events to complementary geophysical datasets to provide insights into eruption source processes, namely the degree of seawater scrubbing of water-soluble SO2 and variations in magma flux. Correlations with the number of lightning strokes and infrasound energy are expected to indicate magma-flux as a controlling process, while correlations with infrasound frequency index are expected to indicate variations in vent-water content as a controlling factor. These comparisons suggest that the measured SO2 masses are primarily a function of eruption magnitude (degassed magma mass) and that scrubbing of SO2 emissions by vent seawater may have exerted a minor effect on the observed SO2 masses. SO2 masses were combined with petrologic constraints on melt inclusion and matrix glass S concentrations to calculate degassed magma masses and volumes. The cumulative SO2-derived degassed magma mass and estimated volume (dense-rock equivalent) for the full Bogoslof eruption were found to be 2.8 × 1010 kg and 9.3 × 106 m3, respectively. When individual event masses are compared against event masses calculated using an empirical plume-height method, a strong correlation is found (R2 = 0.83), with better than order-of-magnitude agreement in most cases. These estimates of eruption masses provide useful information on the magnitude, behavior, and associated hazards of the 2016–2017 eruption, and potentially future unrest at Bogoslof volcano.

The Alaska Volcano Observatory (AVO) seismic network has been in operation since 1988 and during this time has grown from 29 to 217 seismic stations providing real-time monitoring of 32 active volcanoes in Alaska, as well as useful data for regional earthquake monitoring. Since 1988, AVO has detected 59 volcanic eruptions at Aleutian arc volcanoes, and 31 of these have been captured by local seismic instrumentation. As part of this monitoring effort, AVO has cataloged more than 120,000 earthquake hypocenters and magnitudes associated with volcanic processes throughout the arc. This high rate of volcanic activity provides an excellent opportunity to study seismicity associated with magmatic and eruptive processes and develop and refine analytical techniques to track volcanic seismicity and warn of hazardous eruptions. The network is currently undergoing an extensive upgrade, replacing aging short-period analog seismometers with digital broadband instruments. These are expected to improve AVO's seismic capability and further facilitate other geophysical instrumentation such as continuous Global Positioning System receivers, infrasound sensors, and web cams.

The 2016–2017 shallow submarine eruption of Bogoslof volcano produced numerous infrasound signals over 9 months that were recorded on six Alaska Volcano Observatory (AVO) arrays at ranges of 59 to over 800 km from the volcano. The lack of geophysical monitoring near Bogoslof and the repeated production of volcanic clouds to flight levels made monitoring by remote infrasound critical during the eruption; for the first time, AVO relied extensively on automated infrasound detections from regional arrays to dispatch timely notifications of the ongoing activity. Most of the 70 eruptive events were detected on at least one array, but no array detected all of the events mainly because atmospheric conditions were highly variable during the eruption. Acoustic propagation modeling helps explain some of the variation in array detections but also highlights limitations in regional propagation models. To our knowledge, this is the first example of well-recorded infrasound from an explosive eruption occurring in shallow seawater, providing extensive insights into eruption dynamics in this unique environment. The dominance of low-frequency infrasound (0.1–1 Hz) is attributed to eruptions occurring beneath tens of meters of seawater. Higher-frequency infrasound signals were mostly limited to eruptions where the vent was isolated from major interaction with seawater or in several cases where a lava dome grew above sea level.

Characterizing the state of the volcanic vent is key for interpreting observational datasets and accurately assessing volcanic hazards. This is particularly true for remote, complex eruptions such as the 2016–2017 Bogoslof volcano, Alaska eruption sequence. Bogoslof’s eruptions in this period were either shallow submarine or subaerial, or some combination of both. Our results demonstrate how low-frequency sound waves (infrasound), integrated with seismic and satellite data, can provide unique insight into shallow vent processes, otherwise not available. We use simple metrics, such as the infrasound frequency index (FI), event duration, and acoustic-seismic amplitude ratio, to look at changes in the elastic energy radiation and infer changes in seawater access to the vent. Satellite imagery before and after selected eruptions is used to ground-truth inferences on vent conditions. High FI and gradual increases in infrasound frequency content at Bogoslof correspond with transitions from submarine to subaerial vent conditions and a diminished or absent role of water, likely resulting in a drying out of the vent region. Event durations generally correlate with high FI and the range of FI values for each event, suggesting long duration events were more effective at drying out the vent region. A trend from low to high acoustic-seismic amplitude ratios for some long duration events also suggests an increase in acoustic efficiency as the vent dried out. We demonstrate that infrasound can serve as a robust indicator of seawater involvement for Bogoslof and other shallow submarine eruptions that may not be inferable from other datasets, particularly in near-real-time.

Deep long-period earthquakes (DLPs) are an enigmatic type of volcanic seismicity that sometimes precedes eruptions but mostly occurs at quiescent volcanoes. These earthquakes are depleted in high-frequency content and typically occur near the base of the crust. We observed a near-periodic, longlived sequence of more than one million DLPs in the past 19 years beneath the dormant postshield Mauna Kea volcano in Hawai`i. We argue that this DLP sequence was caused by repeated pressurization of volatiles exsolved through crystallization of cooling magma stalled beneath the crust. This "second boiling" of magma is a well-known process but has not previously been linked to DLP activity. Our observations suggest that, rather than portending eruptions, global DLP activity may more commonly be indicative of stagnant, cooling magma.

The injection of gas and pyroclastic material from volcanic vents into the atmosphere is a prolific source of acoustic waves. Infrasound arrays offer efficient, cost-effective, and near real-time solutions to track the rate and intensity of surface activity at volcanoes. Here, we present a simple framework for the analysis of acoustic array data, based on least-squares beamforming, that allows to evaluate the direction and speed of propagation of acoustic waves between source and array. The algorithms include a new and computationally efficient approach for quantitative assessment of the uncertainty on array measurements based on error propagation theory. We apply the algorithms to new data collected by two 6-element infrasound arrays deployed at Mt. Etna during the period July–August 2019. Our results demonstrate that the use of two infrasound arrays allowed detecting and tracking acoustic sources from multiple craters and active vents associated with degassing and ash-rich explosions, vigorous and frequent Strombolian activity, opening of new eruptive fractures and emplacement of lava flows. Finally, we discuss the potential use of metrics based on infrasound array analyses to inform eruption monitoring operations and early warning at volcanoes characterized by episodic intensification of activity.