Alaska-Aleutian Subduction Zone Research Articles

Local-S shear wave splitting along the length of the Alaska–Aleutian subduction zone

SUMMARY The Alaska–Aleutian subduction zone represents an ideal location to study dynamics within a mantle wedge. The subduction system spans several thousand kilometres, is characterized by a slab edge, and has ample seismicity. Additionally, the majority of islands along the arc house broad-band seismic instruments. We examine shear wave splitting of local-S phases originating along the length of the subduction zone. We have dense measurement spacing in two regions, the central Aleutians and beneath Alaska. Beneath Alaska, we observe a rotation in fast splitting directions near the edge of the subducting slab. Fast directions change from roughly trench perpendicular away from the slab edge to trench parallel near the boundary. This is indicative of toroidal flow around the edge of the subducting Alaska slab. In the central Aleutians, local-S splitting is primarily oriented parallel to, or oblique to, the strike of the trench. The local-S measurements, however, exhibit a depth dependence where deeper events show more consistently trench-parallel directions indicating prevalent trench-parallel mantle flow. Our local-S shear wave splitting results suggest trench-parallel orientation are likely present along much of the subduction zone excited by the slab edge, but that additional complexities exist along strike.

Coseismic Coulomb stress changes induced by a 2020–2021 MW > 7.0 Alaska earthquake sequence in and around the Shumagin gap and its influence on the Alaska-Aleutian subduction interface

Three MW > 7.0 earthquakes in 2020–2021 occurred in the Shumagin seismic gap and its adjacent area of the Alaska-Aleutian subduction zone, including the MW7.8 Simeonof thrust earthquake on July 22, 2020, the MW7.6 Sand Point strike-slip earthquake on October 19, 2020, and the MW8.2 Chignik thrust earthquake on July 29, 2021. The spatial and temporal proximity of these three earthquakes prompts us to probe stress-triggering effects among them. Here we examine the coseismic Coulomb stress change imparted by the three earthquakes and their influence on the subduction interface. Our results show that: (1) The Simeonof earthquake has strong loading effects on the subsequent Sand Point and Chignik earthquakes, with the Coulomb stress changes of 3.95 bars and 2.89 bars, respectively. The Coulomb stress change caused by the Sand Point earthquake at the hypocenter of the Chignik earthquake is merely around 0.01 bars, suggesting the negligible triggering effect on the latter earthquake; (2) The triggering effects of the Simeonof, Sand Point, and Chignik earthquakes on aftershocks within three months are not well pronounced because of the triggering rates of 38%, 14%, and 43% respectively. Other factors may have played an important role in promoting the occurrence of these aftershocks, such as the roughness of the subduction interface, the complicated velocity structure of the lithosphere, and the heterogeneous prestress therein; (3) The three earthquakes caused remarkable coseismic Coulomb stress changes at the subduction interface nearby these mainshocks, with an average Coulomb stress change of 3.2 bars in the shallow region directly inwards the trench.

Aseismic slip and recent ruptures of persistent asperities along the Alaska-Aleutian subduction zone

The frictional properties and slip behaviors of subduction thrusts play a key role in seismic and tsunami hazard assessment, especially in weakly coupled “seismic gaps”. Here, we rely on GPS observations in the Shumagin Gap of the Aleutian subduction zone to derive the slip distribution of the 2020 Mw 7.8 Simeonof Island, Alaska earthquake and of the subsequent afterslip during the first 87-day period. Our modeling results show that the mainshock ruptured at depths of ∼30–40 km beneath Simeonof Island. Kinematic and stress-driven models indicate that the afterslip occurred both updip and downdip of the mainshock rupture. Physically plausible locking models derived from interseismic GPS velocities suggest that the 2020 Simeonof and 2021 Mw 8.2 Chignik earthquakes ruptured persistent asperities on the subduction thrust. We infer that there are several additional persistent asperities at depths of 20–50 km west ∼157°W. However, it is still uncertain whether there are additional locked asperities at shallow depths because of the current lack of geodetic observations close to the trench.

Sensitivity of Tsunami Data to the Up-Dip Extent of the July 2021 Mw 8.2 Alaska Earthquake

Abstract A large tsunamigenic earthquake of magnitude Mw 8.2 occurred on the Alaska-Aleutian subduction zone in July 2021. To reveal the characteristics of the event, we first applied spectral and wavelet analyses to the induced tsunami recorded both at the local and Pacific-wide sea level observation networks. Because the earthquake was relatively deep (∼30 km), the resultant maximum tsunami amplitudes were only ∼5 and ∼50 cm in the open ocean and coastal area respectively. However, owing to the unique geological feature of the region, the tsunami had dominant periods of 57–73 min, which are longer than that typically generated by similar-size megathrust earthquakes. Furthermore, we compared multiple source models inferred from various data sets and evaluated their performances in reconstructing the observed tsunami waveforms. The comparison results suggest that the up-dip limit of the rupture area must be restricted at depth of ∼20 km to accurately reproduce the observed tsunami waveforms. Shallower slips beyond the prescribed limit led to an overestimation of the tsunami amplitude. This implies that the earthquake was unlikely to rupture the plate interface on the near trench section.

Subduction earthquakes controlled by incoming plate geometry: The 2020 M > 7.5 Shumagin, Alaska, earthquake doublet

In 2020, an earthquake doublet, a M7.8 on July 22nd and a M7.6 on October 19th, struck the Alaska-Aleutian subduction zone beneath the Shumagin Islands. This is the first documented earthquake doublet involving a megathrust event and a strike-slip event. The first event partially ruptured a seismic gap, which has not hosted large earthquakes since 1917, and the second event was unusual as it broke a trench-perpendicular fault within the incoming oceanic slab. We used an improved Bayesian geodetic inversion method to estimate the fault slip distributions of the major earthquakes using Interferometric Synthetic Aperture Radar (InSAR) wrapped phase and Global Navigation Satellite Systems (GNSS) offsets data. The geodetic inversions reveal that the Shumagin seismic gap is multi-segmented, and the M7.8 earthquake ruptured the eastern segment from 14 km down to 44 km depth. The coseismic slip occurred along a more steeply, 26∘ dipping segment, and was bounded up-dip by a bend of the megathrust interface to a shallower 8∘ dip angle connecting to the trench. The model for the M7.6 event tightly constrained the rupture depth extent to 19-39 km, within the depth range of the M7.8 coseismic rupture area. We find that the M7.6 event ruptured the incoming slab across its full seismogenic thickness, potentially reactivating subducted Kula-Resurrection seafloor-spreading ridge structures. Coulomb stress transfer models suggest that coseismic and/or postseismic slip of the M7.8 event could have triggered the M7.6 event. We conclude that the segmented megathrust structure and the location of intraslab fault structures limited the rupture dimensions of the M7.8 event and are responsible for the segmentation of the Shumagin seismic gap. Our study suggests that the western and shallower up-dip segments of the seismic gap did not fail and remain potential seismic and tsunami hazard sources. The unusual earthquake doublet provides a unique opportunity to improve our understanding of the role of the subducting lithosphere structure in the segmentation of subduction zones.

Stress Interactions between an Interplate Thrust Earthquake and an Intraplate Strike-Slip Event: A Case Study of 2018 Mw 7.9 Gulf of Alaska Earthquake

ABSTRACT Most major earthquakes that have occurred in Alaska are related to rupture of the megathrust along the Alaska–Aleutian subduction zone. Large intraplate earthquakes in the oceanic lithosphere are rare. Recently, one large intraplate earthquake occurred near the Alaska subduction zone: the Mw 7.9 Gulf of Alaska earthquake (23 January 2018). The stress interaction between interplate megathrust and intraplate earthquake has been of interest because it helps to understand stress transfer and seismic hazard estimation. To explore the phenomenon of stress transfer, a viscoelastic spherical finite-element-method model accounting for topographic relief and lateral viscosity variations was used to calculate time-dependent stress transfer associated with the 1964 Mw 9.2 Alaska earthquake. We used postseismic Global Positioning System measurements to probe the differences in rheological properties between the continental mantle and oceanic mantle. The best-fitting model determined the viscosities of the continental mantle and oceanic mantle to be 3×1019 and 1×1020 Pa·s, respectively. Then, we calculated the viscoelastic stress transfers based on this viscosity structure. Because of viscoelastic stress loading associated with the 1964 Alaska earthquake, the ΔCFS values on the 2018 Alaska earthquake rupture fault exceeded the threshold (0.01 MPa) for triggering earthquakes.

Changing impacts of Alaska-Aleutian subduction zone tsunamis in California under future sea-level rise

The amplification of coastal hazards such as distant-source tsunamis under future relative sea-level rise (RSLR) is poorly constrained. In southern California, the Alaska-Aleutian subduction zone has been identified as an earthquake source region of particular concern for a worst-case scenario distant-source tsunami. Here, we explore how RSLR over the next century will influence future maximum nearshore tsunami heights (MNTH) at the Ports of Los Angeles and Long Beach. Earthquake and tsunami modeling combined with local probabilistic RSLR projections show the increased potential for more frequent, relatively low magnitude earthquakes to produce distant-source tsunamis that exceed historically observed MNTH. By 2100, under RSLR projections for a high-emissions representative concentration pathway (RCP8.5), the earthquake magnitude required to produce >1 m MNTH falls from ~Mw9.1 (required today) to Mw8.0, a magnitude that is ~6.7 times more frequent along the Alaska-Aleutian subduction zone.

Identifying Dynamic Response of a Twenty-Story Instrumented Building to 2018 M7.1 Anchorage, Alaska Earthquake and Its Aftershocks

ABSTRACT The dynamic responses of an instrumented twenty-story steel-moment frame office building were identified with the data from mainshock (M 7.1) occurred on November 30, 2018 and its six aftershocks ranging from M4.2 to 5.7 in the Alaska-Aleutian subduction zone. The deconvolution-based seismic interferometry was used to identify the traveling waves and intrinsic-damping ratio in the building. The median shear-wave velocity is 179 m/s for the east-west (EW), 201 m/s for the north-south (NS), and 174 m/s for the torsional responses. The building’s average intrinsic-damping ratio is estimated to be 4.4% and 3.7% for the EW and NS directions, respectively.

Tsunami Resonance Characterization in Japan Due to Trans‐Pacific Sources: Response on the Bay and Continental Shelf

AbstractThe tsunami resonance leads to the amplification of wave heights and extends the duration of wave activity. For tsunami warning and coastal planning, it is important to understand resonance behavior. We characterized the tsunami resonance of trans‐Pacific events in the bays and continental shelves of Japan, focusing on the Hokkaido and Sanriku regions. We considered five tsunami events with sources in the Peru–Chile subduction zone and the Alaska–Aleutian subduction zone. Spectral analysis was applied to observed and synthetic waveforms at coastal tide gauges and offshore GPS gauges to study the frequency content of tsunami energy, and modal analysis was conducted to investigate the eigen‐modes of the natural oscillation affected by regional and local bathymetry. It was found that the signatures of trans‐Pacific tsunami sources are lost in the spectra at most coastal gauges, but they partly remain in the spectra at offshore gauges, which are affected less significantly by topography or bathymetry. The predominant periods in the spectra at most coastal gauges match the eigen‐periods of natural oscillation calculated by modal analysis. Hence, we confirmed that regional or local bathymetry is the dominant factor in the resonance behaviors of trans‐Pacific tsunami events, rather than the source location or geometry. These findings are based on the systematic assessment of trans‐Pacific sources and have implications for tsunami warning and coastal planning along the Pacific coast of Japan.

Seismic tomography of compressional wave velocity and attenuation structure for Makushin Volcano, Alaska

Located in the central portion of the Alaska-Aleutian subduction zone, Makushin Volcano is among the most active volcanoes in the United States and has been classified as a very high threat by the National Volcano Early Warning System based on its eruptive history and proximity to the City of Unalaska and international air routes. In 2015, five standalone seismic stations and three mini seismic arrays of 15 stations each were deployed on Unalaska Island to supplement the Alaska Volcano Observatory permanent seismic network. This temporary array was operational for one year. Taking advantage of the increased azimuthal coverage and the array's increased earthquake detection capability, we developed body-wave Vp and Vp/Vs seismic images of the velocity structure beneath the volcano. Our results show a complex structure with the upper 4 km of the crust. The shallow high-Vp features possibly delineate remnant magma pathways or conduits. Low-Vp regions are found east of the caldera at approximately 4–8 km depth. This is in agreement with previous seismic tomographic work and InSAR-based geodetic models, which had identified this region as a possible long-term source of magma. A three-dimensional compressional wave attenuation model complemented the velocity models. In general, we observe regions of corresponding low P-wave velocity and high attenuation east of the summit caldera between 5 and 7 km depth, supporting the presence of a melt-rich zone. The absolute Vp values (5.5–6.0 km/s) in this region are consistent with the presence of a magma body with a mafic signature. We suggest that a triggering mechanism that involves mafic injections into a more evolved shallow magma chamber is representative of the current magmatic system below the volcano, similarly to what petrological studies have inferred for past eruptions.

Spatial Variation of Slip Behavior Beneath the Alaska Peninsula Along Alaska‐Aleutian Subduction Zone

AbstractWe resurveyed preexisting campaign Global Positioning System (GPS) sites and estimated a highly precise GPS velocity field for the Alaska Peninsula. We use the TDEFNODE software to model the slip deficit distribution using the new GPS velocities. We find systematic misfits to the vertical velocities from the optimal model that fits the horizontal velocities well, which cannot be explained by altering the slip distribution, so we use only the horizontal velocities in the study. Locations of three boundaries that mark significant along‐strike change in the locking distribution are identified. The Kodiak segment is strongly locked, the Semidi segment is intermediate, the Shumagin segment is weakly locked, and the Sanak segment is dominantly creeping. We suggest that a change in preexisting plate fabric orientation on the downgoing plate has an important control on the along‐strike variation in the megathrust locking distribution and subduction seismicity.

An approach for estimating the largest probable tsunami from far-field subduction zone earthquakes

Following the recent unexpected earthquake events of 2004 and 2011, it can be cautiously extrapolated that all major subduction zones bearing the capacity to produce mega-earthquake events will eventually do so given enough time, irrespective of the lack of such in the relatively short historical record. This notion has led to an effort of assigning maximum earthquake magnitudes to all major subduction zones, either based on geological constraints or based on size–frequency relations, or a combination of both. In this study, we utilize the proposed maximum magnitudes to assess tsunami hazard in Central California in the very long return periods. We also assessed tsunami hazard following an alternative methodology to calculate maximum magnitudes, which uses scaling relations for subduction zone earthquakes and maximum fault rupture scenarios found in literature. A sensitivity analysis is performed for Central California that is applicable to any coastal site in the Pacific Rim and can readily provide a strong indication for which subduction zones beam the most energy toward a study area. The maximum earthquake scenarios are then narrowed down to a few candidates, for which the initial conditions are examined in more detail. The chosen worst-case scenarios for Central California stem from the Alaska–Aleutian subduction zone that beams more energy and generates the biggest amplitude waves toward the study area. The largest tsunami scenario produces maximum free surface elevations of 15 m and run-up heights greater than 20 m.

Extending Alaska’s plate boundary: Tectonic tremor generated by Yakutat subduction

The tectonics of the eastern end of the Alaska-Aleutian subduction zone are complicated by the inclusion of the Yakutat microplate, which is colliding into and subducting beneath continental North America at near-Pacific-plate rates. The interaction among these plates at depth is not well understood, and further east, even less is known about the plate boundary or the source of Wrangell volcanism. The drop-off in Wadati-Benioff zone (WBZ) seismicity could signal the end of the plate boundary, the start of aseismic subduction, or a tear in the downgoing plate. Further compounding the issue is the possible presence of the Wrangell slab, which is faintly outlined by an anemic, eastward-dipping WBZ beneath the Wrangell volcanoes. In this study, I performed a search for tectonic tremor to map slow, plate-boundary slip in south-central Alaska. I identified ∼11,000 tremor epicenters, which continue 85 km east of the inferred Pacific plate edge marked by WBZ seismicity. The tremor zone coincides with the edges of the downgoing Yakutat terrane, and tremors transition from periodic to continuous behavior as they near the aseismic Wrangell slab. I interpret tremor to mark slow, semicontinuous slip occurring at the interface between the Yakutat and North America plates. The slow slip region lengthens the megathrust interface beyond the WBZ and may provide evidence for a connection between the Yakutat slab and the aseismic Wrangell slab.

Beach ridges as paleoseismic indicators of abrupt coastal subsidence during subduction zone earthquakes, and implications for Alaska-Aleutian subduction zone paleoseismology, southeast coast of the Kenai Peninsula, Alaska

The Kenai section of the eastern Alaska-Aleutian subduction zone straddles two areas of high slip in the 1964 great Alaska earthquake and is the least studied of the three megathrust segments (Kodiak, Kenai, Prince William Sound) that ruptured in 1964. Investigation of two coastal sites in the eastern part of the Kenai segment, on the southeast coast of the Kenai Peninsula, identified evidence for two subduction zone earthquakes that predate the 1964 earthquake. Both coastal sites provide paleoseismic data through inferred coseismic subsidence of wetlands and associated subsidence-induced erosion of beach ridges. At Verdant Cove, paleo-beach ridges record the paleoseismic history; whereas at Quicksand Cove, buried soils in drowned coastal wetlands are the primary indicators of paleoearthquake occurrence and age. The timing of submergence and death of trees mark the oldest earthquake at Verdant Cove that is consistent with the age of a well documented ∼900-year-ago subduction zone earthquake that ruptured the Prince William Sound segment of the megathrust to the east and the Kodiak segment to the west. Soils buried within the last 400–450 years mark the penultimate earthquake on the southeast coast of the Kenai Peninsula. The penultimate earthquake probably occurred before AD 1840 from its absence in Russian historical accounts. The penultimate subduction zone earthquake on the Kenai segment did not rupture in conjunction with the Prince William Sound to the northeast. Therefore the Kenai segment, which is presently creeping, can rupture independently of the adjacent Prince William Sound segment that is presently locked.

Modern Salt-Marsh and Tidal-Flat Foraminifera From Sitkinak and Simeonof Islands, Southwestern Alaska

We describe the modern distribution of salt-marsh and tidal-flat foraminifera from Sitkinak Island (Trinity Islands) and Simeonof Island (Shumagin Islands), Alaska, to begin development of a dataset for later use in reconstructing relative sea-level changes caused by great earthquakes along the Alaska-Aleutian subduction zone. Dead foraminifera were enumerated from a total of 58 surface-sediment samples collected along three intertidal transects around a coastal lagoon on Sitkinak Island and two intertidal transects on Simeonof Island. Two distinctive assemblages of salt-marsh foraminifera were recognized on Sitkinak Island. Miliammina fusca dominated low-marsh settings and Balticammina pseudomacrescens dominated the high marsh. These two species make up >98% of individuals. On Simeonof Island, 93% of individuals in high-marsh settings above mean high water were B. pseudomacrescens . The tidal flat on Simeonof Island was dominated by Cibicides lobatulus (60% of individuals), but the lower limit of this species is subtidal and was not sampled. These results indicate that uplift or subsidence caused by repeated earthquakes along the Alaska-Aleutian subduction zone could be reconstructed in coastal sediments using alternating assemblages of near monospecific B. pseudomacrescens and low-marsh or tidal-flat foraminifera.

New geophysical insight into the origin of the Denali volcanic gap

SUMMARY Volcanic gaps are segments of subduction zones that lack the volcanic activity usually found at these convergent margins. They are regions where the necessary conditions to produce melt may appear favourable, but where volcanoes are surprisingly absent from the surface. In this study, we present a new model that can explain the occurrence of such volcanic gaps. It is based on seismic imaging and geodynamic modelling of the Denali volcanic gap, a ∼400-km-wide region at the eastern end of the Alaska-Aleutian subduction zone. Here, the thick crust of the Pacific Plate and Yakutat terrane subduct at shallow angle beneath North America. A high-resolution seismic profile clearly images the subducting crust undergoing progressive dehydration between 50 and 120 km depth, and a negative subhorizontal velocity contrast at 60 km depth in the overlying mantle wedge. We interpret this 60 km discontinuity as marking the top of a layer of partial melt that pools at the base of the overriding plate. In steadystate subduction models, melt accumulates at the apex of a vaulted mantle wedge, the ‘pinch zone’, from where it may break through the overlying lithosphere to the surface. Beneath the Denali volcanic gap, the pinch zone is absent (or greatly reduced) because shallow subduction of the Yakutat terrane progressively cools the system, and causes the slab to advance and replace the hot core of the mantle wedge. This regime can be seen as the opposite of subduction rollback. It prevents the formation of a pinch zone, reduces the length of the melting column and causes melt to pool at the base of the overriding plate, thus inhibiting magma generation and extraction.

Coseismic and pre-seismic subsidence associated with great earthquakes in Alaska

Alternating beds of peat and mud in sediment sequences on the south-central Alaskan coast record coseismic and inter-seismic relative land and sea-level movements caused by repeated great earthquakes on the Alaska–Aleutian subduction zone. During the AD 1964 Mw=9.2 earthquake, tidal marshes and wetlands around upper Cook Inlet experienced up to 2 m of subsidence, burying peat-forming communities with intertidal mud. Here we use quantitative analyses of fossil diatoms within peat–mud couplets to reconstruct land/sea-level changes for the 1964 and five earlier great earthquakes during the past 3300 years. In contrast to geodetic observations that are limited to the present post-seismic phase, we quantify varying spatial patterns of uplift and subsidence through complete earthquake cycles. Relative land uplift characterises most of the inter-seismic phase of each cycle at our sites, whereas each great earthquake was preceded by a short period of pre-seismic relative land subsidence.

High interseismic coupling of the Alaska Subduction Zone SW of Kodiak Island inferred from GPS data

We use Global Positioning System (GPS) measurements to make the first geodetic study of the Semidi segment of the Alaska‐Aleutian subduction zone. This segment, which sustained an MW 8.2 earthquake in 1938, lies between Kodiak Island where the subduction interface appears to presently be fully locked, and the Shumagin Islands segment where substantial aseismic slip occurs. We invert the GPS station velocity estimates using a nonlinear least squares algorithm to solve for the width of the locked zone, the dip, and the interseismic coupling of a model subduction interface. The data are consistent with a shallow plate interface dipping ∼6°, a locking depth of ∼23 km, and high interseismic coupling of ∼80%.

Potential seismic hazards and tectonics of the upper Cook Inlet basin, Alaska, based on analysis of Pliocene and younger deformation

The Cook Inlet basin is a northeast-trending forearc basin above the Aleutian subduction zone in southern Alaska. Folds in Cook Inlet are complex, discontinuous structures with variable shape and vergence that probably developed by right-transpressional deformation on oblique-slip faults extending downward into Mesozoic basement beneath the Tertiary basin. The most recent episode of deformation may have began as early as late Miocene time, but most of the deformation occurred after deposition of much of the Pliocene Sterling Formation. Deformation continued into Quaternary time, and many structures are probably still active. One structure, the Castle Mountain fault, has Holocene fault scarps, an adjacent anticline with flower structure, and historical seismicity. If other structures in Cook Inlet are active, blind faults coring fault-propagation folds may generate M w 6–7+ earthquakes. Dextral transpression of Cook Inlet appears to have been driven by coupling between the North American and Pacific plates along the Alaska-Aleutian subduction zone, and by lateral escape of the forearc to the southwest, due to collision and indentation of the Yakutat terrane 300 km to the east of the basin.

Microfossil analysis of sediments representing the 1964 earthquake, exposed at Girdwood Flats, Alaska, USA

Abstract Diatom, pollen, foraminifera and thecamoebian assemblages from an outcrop of peat and silt at Girdwood Flats, in the upper Turnagain Arm of the Cook Inlet, Alaska, record four phases of relative land and sea-level changes. The first phase is the development of freshwater swamp above high marsh sediments during relative land uplift, caused by strain accumulation along the locked portion of the Alaska-Aleutian subduction zone. In second phase, the top 2 cm of the peat, all microfossil groups record pre-seismic relative sea-level rise (relative land subsidence). The third phase is rapid land subsidence, 1.7 m, during the earthquake of March 1964 that initiated intertidal silt accumulation above the peat. The final phase is the colonisation of mudflat by salt marsh communities during post-seismic land uplift. The microfossil data compare favourably with sequences from Washington, Oregon and British Columbia that record late Holocene submergence events caused by earthquakes. The comparable changes in microfossil assemblages record the different phases of relative land and sea-level changes and the magnitude of land subsidence caused by each earthquake (expressed relative to the tidal range at the site). These results raise the question whether preseismic sea-level rise represents any kind of warning of large earthquakes.