Subduction Zone Interface Research Articles

Influence of material contrast on fault branching behavior

[1] Material contrasts across faults are a common occurrence, and it is important to understand if these material contrasts can influence the path of rupture propagation. Here we examine models, solved numerically, of rupture propagation through one type of geometric complexity, that of a fault branch stemming from a planar main fault on which rupture initiates. This geometry, with a material contrast across the main fault, could be representative of either a mature strike-slip fault or a subduction zone interface. We consider branches in both the compressional and extensional quadrants of the fault, and material configurations in which the branch fault is in either the stiffer or the more compliant material as well as configurations with no material contrast. We find that there are regimes in which this elastic contrast can influence the rupture behavior at a branching junction, but there are also stress states for which the branch activation will not depend on the orientation of the mismatch. For the scenarios presented here, both compressional and extensional side branches are more likely to rupture if the branch is on the side of the fault with the more compliant material versus the stiffer material. The stresses induced on the branch fault, by rupture traveling on the main fault, are different for the two orientations of material contrast. We show how the interactions between rupture on the two faults determine which faults are activated.

Inhibition of subduction thrust earthquakes by low-temperature plastic flow in serpentine

Serpentine is considered a candidate hydrous mineral in explaining aseismic zones along faults; however, recent frictional experiments have shown a marked increase in its shear strength with increasing depth, implying its deformation via unstable sliding at a plate interface in subduction zones. Here we report the results of a series of simple-shear experiments designed to determine the dominant deformation style of low-temperature serpentine species under P– T conditions that correspond to the mantle wedge corner in cool subduction zones (ca. P = 1 GPa and T = 200 °C). We found that with increasing shear strain, the dominant flow mechanism of the serpentinite evolves from (1) inhomogeneous semi-brittle flow by strain localization into sample-scale shear bands to (2) homogeneous ductile flow by intracrystalline deformation within individual serpentine grains, with the development of a planar shape fabric and a strong crystallographic-preferred orientation. TEM observations revealed that grain size reduction in lizardite crystals occurs due to (001) interlayer glide, represented by the occurrence of tiny sheet-like sectors oriented parallel to (001) liz lattice fringes, which evolved from large, elongated sectors with long axes oriented normal to (001) liz. Given that a low-viscosity serpentinized layer upon a subducting plate interface produces strain localization within the layer and subsequent large bulk shear strains, the above results indicate that the presence of low-temperature serpentine species at the plate interface in cool subduction zones inhibits the initiation of subduction thrust earthquakes, as stress is preferentially accommodated by plastic flow. This hypothesis explains the occurrence of an anomalous non-seismic region (devoid of large thrust earthquakes) in locally hydrated forearc mantle within northeast Japan.

Earthquake mechanisms and active tectonics of the Hellenic subduction zone

We use improved focal mechanisms and centroid depth estimates of earthquakes, combined with GPS velocities, to examine the tectonics of the Hellenic subduction zone, and in particular the processes occurring at both ends of the Hellenic Arc. Nubia-Aegean convergence is accommodated by shallowly dipping thrust-faulting along the subduction–zone interface, as well as by steeper splay faults in the overriding material. From a comparison of observed and expected seismic moment release over the last 100 yr, combined with existing knowledge of the longer-term documented historical record, we confirm earlier suggestions that most (80 per cent) of this convergence is accommodated aseismically, that is, that the subduction zone is uncoupled. This conclusion is robust, even allowing for rare very large earthquakes on splay faults, such as that of AD 365, and also allowing for the contribution of small earthquakes. The downgoing Nubian plate deforms by arc-parallel contraction at all depths, from 200 km seaward of Crete to at least 100 km within the subducting slab. Extensional (T) axes of earthquakes are aligned downdip within the descending slab suggesting that, even if the aseismic prolongation of the slab has reached the 670 km mantle discontinuity, it does not transmit stresses to shallower depths. Shallow thrust-faulting earthquakes on the subduction interface show a divergence of slip vectors round the arc, and GPS measurements show that this is accommodated mainly by E–W extension on normal faults in the overriding Aegean material. The eastern end of the subduction zone, south of Rhodes, displays distributed deformation in the overriding material, including a mixture of strike-slip and splay-thrust faulting, and probably involves rotations about a vertical axes. Here slip on the interface itself is by thrust faulting with slip vectors oblique to the arc but parallel to the overall Nubia-Aegean convergence: there is no evidence for slip-partitioning in the traditional sense. In the west, the subduction zone terminates in a distributed zone of parallel NE–SW strike-slip faults, of which the most prominent is the Kefalonia Transform Fault (KTF). A flexural gravity anomaly confirms that the deep bathymetric escarpment of the KTF is a lateral ramp, formed as the Ionian islands are emplaced SW onto the Apulian lithosphere, and enhanced by minor thrust faulting with slip vectors perpendicular to the scarp. Distributed parallel strike-slip faults both SW and NE of mainland central Greece terminate in E–W graben in central Greece, which accommodate the overall NE–SW shear by clockwise block rotation. Central Greece therefore acts as a relay zone between the strike-slip faulting of the NE Aegean and the Ionian Islands–western Peloponnese.

Deep low‐frequency earthquakes in tremor localize to the plate interface in multiple subduction zones

Deep tremor under Shikoku, Japan, consists primarily, and perhaps entirely, of swarms of low‐frequency earthquakes (LFEs) that occur as shear slip on the plate interface. Although tremor is observed at other plate boundaries, the lack of cataloged low‐frequency earthquakes has precluded a similar conclusion about tremor in those locales. We use a network autocorrelation approach to detect and locate LFEs within tremor recorded at three subduction zones characterized by different thermal structures and levels of interplate seismicity: southwest Japan, northern Cascadia, and Costa Rica. In each case we find that LFEs are the primary constituent of tremor and that they locate on the deep continuation of the plate boundary. This suggests that tremor in these regions shares a common mechanism and that temperature is not the primary control on such activity.

Predicted Ground Motions for Great Interface Earthquakes in the Cascadia Subduction Zone

Ground motions for earthquakes of moment magnitude ( M ) 7.5–9.0 on the Cascadia subduction zone interface are simulated based on a stochastic finite-fault model and used to estimate average response spectra for firm-site conditions in Vancouver, Victoria, and Seattle. We also express the response spectra as ground-motion prediction equations (GMPEs) for Cascadia events. The simulations are calibrated by modeling the wealth of ground-motion data from the M 8.1 Tokachi-Oki earthquake sequence of Japan. Adjustments to the calibrated model are made to consider average source, attenuation, and site parameters for the Cascadia region. We perform best estimate simulations for a preferred set of input parameters. Typical results suggest mean values of 5%-damped pseudoacceleration in the range from about 100 to 200 cm/sec2, at frequencies from 1 to 4 Hz, for firm-ground conditions in Vancouver, Victoria, and Seattle. Uncertainty in stress drop causes uncertainty in simulated response spectra of about ±50%. Uncertainties in the attenuation model produce even larger uncertainties in response spectral amplitudes—a factor of about 2 at 100 km, becoming even larger at greater distances. It is thus important to establish the regional attenuation model for ground-motion simulations. Furthermore, combining data from regions with different attenuation characteristics—in particular Japan and Mexico—into a global subduction zone database for development of global empirical GMPEs may not be a sound practice. Time histories of acceleration for the stochastically simulated motions are provided for reference sites in Vancouver, Victoria, and Seattle. An alternative set of motions, based on lightly modifying real recordings from the Tokachi-Oki earthquake to match expected conditions for Cascadia cities, are also provided. These alternative records have similar spectral content to the simulated motions but contain additional complexity and more realistic phasing. The provision of alternative record sets allows users to conduct studies to determine the importance of these effects for structural response.

Interseismic uplift rates for western Oregon and along‐strike variation in locking on the Cascadia subduction zone

We quantify the spatial pattern of uplift rate in western Oregon and northernmost California using tidal and leveling records to better understand the pattern of interseismic locking on the Cascadia subduction zone. We extend relative sea level time series of the six primary NOAA tide gauges to include all observations from 1925 to 2006. Previously unidentified bench mark instability biases portions of tidal records by ≤1.6 mm a−1 before correction. We determine precise relative uplift rates at the six tidal sites with an adjustment that includes rates of differenced time series as additional constraints. Our analysis of National Geodetic Survey leveling data between tide gauges corrects errors in 1941 leveling, and 184 secondary ties double the number of highest quality uplift rate estimates. Relative uplift rates from leveling are adjusted to the tidal rates, accounting for uncertainties in both data types. Tidal and leveling uplift rates agree within error for all but one of the coastal segments, where we infer systematic leveling error affects the 1988 line. Uplift rates are made absolute using an interval and location‐specific geocentric sea level rise rate of 2.3 ± 0.2 mm a−1. Total propagated one sigma errors for the absolute uplift rates of bench marks are ∼0.4 mm a−1. Along‐strike changes in uplift rate near 45°N and 42.8°N require two distinct changes in locking depth, as inferred from elastic dislocation modeling. The along‐strike changes in locking on this portion of the Cascadia subduction zone interface correlate to the western and southern extent of the mafic Siletzia block in the fore arc.

Non-volcanic tremor resulting from the combined effect of Earth tides and slow slip events

Swarms of non-volcanic tremor in southeastern Japan are associated with slow slip events and tend to occur with a periodicity of 12 or 24 h. This periodicity can be reproduced by a combination of stresses due to Earth tides and transient stress changes caused by slow slip events. Non-volcanic tremors may therefore be useful for understanding stress relaxation at the subduction-zone interface. Slow slip events1 are accompanied by swarms of non-volcanic tremor2,3 along the subduction zone of the Philippine Sea plate in southwest Japan: the swarms often occur with a periodicity of about 12 or 24 h (refs 4, 5). These episodic events are considered to be a manifestation of stress relaxation at the subducting plate interface1,6,7. Here, we analyse seismic data to record the locations and durations of non-volcanic tremor swarms that occurred in May 2005 and February 2006. We evaluate the magnitude of stress changes produced by slow slip events as well as the effective normal stress on the plate interface where slow slip events occur. We find that the observed periodicity in tremor occurrence originates from a combined effect of the periodic stress due to Earth tides and the transient stress due to slow slip events. Our calculations show that non-volcanic tremor is sensitive to stress change. This phenomenon can therefore be effective in monitoring the process of stress relaxation at subducting plate interfaces.

The Seismogenic Zone of Subduction Thrust Faults

This book is essentially B.S.—Before Sumatra. Before December 26, 2004, when part of the thrust fault interface between the subducting Indian and overriding Burma plates off the coast of Sumatra broke. Before the rupture raced 1,200 km northward at 10,000 km/hr (Mach 8). Before an area of the subducting plate comparable to that of California moved 10 meters, causing a magnitude 9.3 earthquake that was the largest in 40 years. Before the resulting seafloor motion generated a tsunami that crossed the Indian Ocean at the speed of a jetliner and killed more than 200,000 people. Is a B.S.—Before Sumatra—book about subduction zones still useful? The almost unbelievable size of this earthquake and its effects brought home how little we still know about the physics of the subduction zone interface where large earthquakes including the rare giant (M 9) earthquakes occur. (As part of efforts to entice students into the earth sciences we might call these “extreme” or “X-quakes.”) Two previous giant earthquakes, off Chile in 1960 and Alaska in 1964, occurred just before the discovery of plate tectonics explained that such earthquakes occur when years of strain …

Comparative Deformation Behavior of Minerals in Serpentinized Ultramafic Rock: Application to the Slab-Mantle Interface in Subduction Zones

The layer-structure minerals serpentine, brucite, and talc are postulated to form in the mantle wedge above a subducting slab as a result of progressive hydration and silica metasomatism. Tectonic mixing at the slab-mantle interface generates serpentinite mélanges that contain blocks of high-pressure (HP) or ultrahigh-pressure (UHP) metamorphic rock derived from the subducting slab. Such serpentinite mélanges may provide a means of exhumation of HP/UHP metamorphic rocks, and may define the lower limit of locked regions on the subduction interface that fail in large earthquakes. We review recently obtained frictional strength data for brucite and talc over the temperature range 25-400°C at 100 MPa effective normal stress and compare them with new data for antigorite. These minerals respond to heating in different ways, causing their frictional strengths to diverge. Water-saturated antigorite strength increases toward the fixed dry value of μ ≈ 0.75-0.80 with heating: μ ≈ 0.50 at 25°C and μ > 0.60 at 400°C. The difference in μ between dry and watersaturated talc gouge also decreases with increasing temperature, but both the dry and watersaturated values of μ are lower at elevated temperatures. For dry talc, μ decreases from 0.35 to 0.25 between 25° and 300°C, whereas for water-saturated talc, μ is approximately 0.20 at 25°C and 0.10-0.15 at elevated temperatures. Weakening of the interlayer bond of talc with heating may be responsible for the overall reduction in its frictional strength. The strength of dry brucite also is fixed at μ = 0.45-0.50, but the water-saturated value of μ decreases from ≈0.30 at 25°C to 0.20-0.25 at 200°-400°C. The water-saturated brucite gouge has extensively recrystallized along the shear surfaces, and its weakening may be attributable to solution-transfer processes. Because the serpentine minerals become stronger at elevated temperatures, to achieve low frictional strength in a serpentinized mantle wedge or serpentine-rich mélange would require some other cause such as nearly lithostatic fluid pressures or the addition of lower-strength minerals. Increasing abundance of brucite and talc in serpentinite at constant physical conditions would progressively reduce its frictional strength. The concentration of talc along a metasomatic front at the edge of the mantle wedge or in reaction zones surrounding HP/UHP blocks in serpentinite mélange should tend to localize shear in this extremely weak material.

Geodetic and seismic signatures of episodic tremor and slip in the northern Cascadia subduction zone

Slip events with an average duration of about 10 days and effective total slip displacements of severalc entimetres have been detected on the deeper (25 to 45 km) part of the northern Cascadia subduction zone interface by observing transient surface deformation on a network of continuously recording Global Positioning System (GPS) sites. The slip events occur down-dip from the currently locked, seismogenic portion of the subduction zone, and, for the geographic region around Victoria, British Columbia, repeat at 13 to 16 month intervals. These episodes of slip are accompanied by distinct, low-frequency tremors, similar to those reported in the forearc region of southern Japan. Although the processes which generate this phenomenon of episodic tremor and slip (ETS) are not well understood, it is possible that the ETS zone may constrain the landward extent of megathrust rupture, and conceivable that an ETS event could precede the next great thrust earthquake.

Subduction zone coupling and tectonic block rotations in the North Island, New Zealand

The GPS velocity field in the North Island of New Zealand is dominated by the long‐term tectonic rotation of the eastern North Island and elastic strain from stress buildup on the subduction zone thrust fault. We simultaneously invert GPS velocities, earthquake slip vectors, and geological fault slip rates in the North Island for the angular velocities of elastic crustal blocks and the spatially variable degree of coupling on faults separating the blocks. This approach allows us to estimate the distribution of interseismic coupling on the subduction zone interface beneath the North Island and the kinematics of the tectonic block rotations. In agreement with previous studies we find that the subduction zone interface beneath the southern North Island has a high slip rate deficit during the interseismic period, and the slip rate deficit decreases northward along the margin. Much of the North Island is rotating as several, distinct tectonic blocks (clockwise at 0.5–3.8 deg Myr−1) about nearby axes relative to the Australian Plate. This rotation accommodates much of the margin‐parallel component of motion between the Pacific and Australian plates. On the basis of our estimation of the block kinematics we suggest that rotation of the eastern North Island occurs because of the southward increasing thickness of the subducting Hikurangi Plateau. These results have implications for our understanding of convergent margin plate boundary zones around the world, particularly with regard to our knowledge of mechanisms for rapid tectonic block rotations at convergent margins and the role of block rotations in the slip partitioning process.

Probabilistic seismic hazard analysis for Sumatra, Indonesia and across the Southern Malaysian Peninsula

The ground motion hazard for Sumatra and the Malaysian peninsula is calculated in a probabilistic framework, using procedures developed for the US National Seismic Hazard Maps. We constructed regional earthquake source models and used standard published and modified attenuation equations to calculate peak ground acceleration at 2% and 10% probability of exceedance in 50 years for rock site conditions. We developed or modified earthquake catalogs and declustered these catalogs to include only independent earthquakes. The resulting catalogs were used to define four source zones that characterize earthquakes in four tectonic environments: subduction zone interface earthquakes, subduction zone deep intraslab earthquakes, strike-slip transform earthquakes, and intraplate earthquakes. The recurrence rates and sizes of historical earthquakes on known faults and across zones were also determined from this modified catalog. In addition to the source zones, our seismic source model considers two major faults that are known historically to generate large earthquakes: the Sumatran subduction zone and the Sumatran transform fault. Several published studies were used to describe earthquakes along these faults during historical and pre-historical time, as well as to identify segmentation models of faults. Peak horizontal ground accelerations were calculated using ground motion prediction relations that were developed from seismic data obtained from the crustal interplate environment, crustal intraplate environment, along the subduction zone interface, and from deep intraslab earthquakes. Most of these relations, however, have not been developed for large distances that are needed for calculating the hazard across the Malaysian peninsula, and none were developed for earthquake ground motions generated in an interplate tectonic environment that are propagated into an intraplate tectonic environment. For the interplate and intraplate crustal earthquakes, we have applied ground-motion prediction relations that are consistent with California (interplate) and India (intraplate) strong motion data that we collected for distances beyond 200 km. For the subduction zone equations, we recognized that the published relationships at large distances were not consistent with global earthquake data that we collected and modified the relations to be compatible with the global subduction zone ground motions. In this analysis, we have used alternative source and attenuation models and weighted them to account for our uncertainty in which model is most appropriate for Sumatra or for the Malaysian peninsula. The resulting peak horizontal ground accelerations for 2% probability of exceedance in 50 years range from over 100% g to about 10% g across Sumatra and generally less than 20% g across most of the Malaysian peninsula. The ground motions at 10% probability of exceedance in 50 years are typically about 60% of the ground motions derived for a hazard level at 2% probability of exceedance in 50 years. The largest contributors to hazard are from the Sumatran faults.

Franciscan subduction off to a slow start: evidence from high-precision Lu–Hf garnet ages on high grade-blocks

Lu–Hf analyses of garnet from metabasic amphibolite, glaucophane schist and eclogite facies blocks from the Franciscan complex give highly precise ages that allow us to place new constraints on the early thermal history of the Franciscan subduction zone. Garnets yield 176 Lu/ 177 Hf ratios ranging from 1.5 to 28 with the highest ratios from garnets with high spessartine/pyrope ratio. Sulphuric acid leaching (SAL) of garnets revealed the presence of inclusions with significantly higher Lu/Hf ratios than those of garnet itself (most likely apatite). Their removal by SAL brings the 176 Lu/ 177 Hf ratios in garnets down by as much as 40%. This suggests that 176 Lu/ 177 Hf ratios of apparently pure garnets can be greatly overestimated due to the presence of such inclusions. Sm–Nd garnet analyses were dominated by inclusions (mainly sphene), and failed to provide precise and accurate age information. The oldest Lu–Hf ages are 168.7F0.8 and 162.5F0.5 Ma on plagioclase-bearing garnet amphibolite from Panoche Pass and the Berkeley Hills, respectively, which suggests initiation of the subduction zone at about 169 Ma, coeval with the formation of the tectonically overlying Coast Range Ophiolite. Relatively high temperature conditions persisted for about 14 Ma as indicated by 153.4F0.8 Ma garnet growth recorded in epidote amphibolite and 157.9F0.7 in eclogite from Ring Mountain and Jenner, respectively. A 146.7F0.7 Ma age was obtained from garnet glacuophane schist, metamorphosed at around 400 jC. The sequence of ages from central and northern California shows a younging trend with decreasing metamorphic grade, which supports previous suggestions that the high-grade metamorphic blocks and slices resulted from progressive underthrusting and underplating in a cooling subduction system. Combining geothermometry with geochronological data allow us to estimate cooling rate along the subduction zone interface from amphibolite to blueschist facies conditions as ca. 15 jC/Ma. The thermal history requires high initial geothermal gradients within both the footwall and the hangingwall of the subduction zone and a relatively slow subduction rate of the order of 10 km/Ma during the initial stages of Franciscan subduction. Such conditions are consistent with initiation of the subduction zone at or close to an oceanic spreading centre. The data also suggest slow exhumation rates and significant residence time at depth of the earliest Franciscan rocks.

Constraints on the petrologic structure of the subduction zone slab-mantle interface from Franciscan Complex exotic ultramafic blocks

Ultramafic blocks within mud-matrix melange of the Franciscan Complex, California, preserve a series of metasomatic mineral zones generated by infiltration of Si-rich hydrous fluids during subduction. We describe the petrology and geochemistry of the metasomatic zones and compare them to current model predictions for the metasomatism of the mantle wedge by subduction zone fluids. Fluid flow affected a Cr-spinel lherzolite protolith to form first serpentinite, then a talc-dominated rock, and finally an amphibole-rich assemblage. A diverse suite of accessory minerals in the amphibole-rich zone (titanite + clinozoisite + zircon + apatite) suggests that the trace element signature of subduction zone fluids may be fractionated in this zone. Oxygen isotopic evidence suggests that the ultramafic blocks equilibrated with metasomatic fluids during serpentinization and that subsequent reactions occurred in equilibrium with these fluids in a temperature range of 450–500 °C. Whole-rock geochemistry indicates mobility of many elements into and out of the blocks during metasomatism, including elements such as Ti which are currently considered to have low solubilities in such fluids. Taken as a whole, the blocks appear to preserve the metasomatic structure of the slab-mantle interface in subduction zones and imply that the chemistry of slab-derived fluids is modified as they pass through these metasomatic zones in the mantle wedge. Our results suggest that the primary composition of subduction zone fluids is not likely reflected by arc magmas. Instead, we propose that arc magmas are derived from regions of the mantle fluxed by fluids residual to the metasomatic processes we observe.

Stress and deformation in subduction zones: insight from the record of exhumed metamorphic rocks

Abstract High pressure (HP) and ultrahigh (UHP) metamorphic rocks are exhumed from subduction zones at high rates on the order of plate velocity (cm/year). Their structural and microstructural record provides insight into conditions and physical state along the plate interface in subduction zones to depths of >100 km. Amazingly, many identified (U)HP metamorphic rocks appear not to be significantly deformed at (U)HP conditions, despite their history within a high strain rate mega-shearzone. Other (U)HP metamorphic rocks seem to be deformed exclusively by dissolution-precipitation creep. Indications of deformation by dislocation creep are lacking, apart from omphacite in some eclogites. Available flow laws for dislocation creep (extrapolated to low natural strain rates, which is equivalent to no deformation on the time scales of subduction and exhumation, i.e., 1 to 10 Ma) pose an upper bound to the magnitude of stress as a function of temperature along the trajectory followed by the rock. Although the record of exhumed (U)HP metamorphic rocks may only be representative of specific types or evolutionary stages of subduction zones, for such cases it implies: (1) strongly localized deformation; (2) predominance of dissolution-precipitation creep and fluid-assisted granular flow in the shear zones, suggesting Newtonian behaviour; (3) low magnitude of differential stress; which (4) is on the order of the stress drop inferred for earthquakes; and (5) negligible shear heating. These findings are easily reconciled with exhumation by forced flow in a low viscosity subduction channel prior to collision, implying effective decoupling between the plates.

Variation of interplate fault zone properties with depth in the japan subduction zone

The depth dependence of physical properties along the Japan subduction zone interface was explored using teleseismic recordings of earthquake signals. Broadband body waves were inverted to determine the duration of rupture and source depth for 40 interplate thrust earthquakes located offshore of Honshu between 1989 and 1995. After scaling for differences in seismic moment, there is a systematic decrease in rupture duration with increasing depth along the subducting plate interface. This indicates increases in rupture velocity or stress drop with depth, likely related to variation in rigidity of sediments on the megathrust.

Strong Ground Motion Attenuation Relationships for Subduction Zone Earthquakes

We present attenuation relationships for peak ground acceleration and response spectral acceleration for subduction zone interface and intraslab earthquakes of moment magnitude M 5 and greater and for distances of 10 to 500 km. The relationships were developed by regression analysis using a random effects regression model that addresses criticism of earlier regression analyses of subduction zone earthquake motions. We find that the rate of attenuation of peak motions from subduction zone earthquakes is lower than that for shallow crustal earthquakes in active tectonic areas. This difference is significant primarily for very large earthquakes. The peak motions increase with earthquake depth and intraslab earthquakes produce peak motions that are about 50 percent larger than interface earthquakes.

Elastic wave velocities in isochemical granulite and amphibolite: Origin of a low‐velocity layer at the slab/mantle wedge interface

Simultaneous measurements of compressional and shear wave velocities, Vp and Vs, in granulite and amphibolite with identical chemistry were carried out from room temperature to 850°C at 1 GPa. Both Vp and Vs in granulite and amphibolite decreased linearly with increasing temperature. The values of Vp and Vs at 700°C are 7.75 km/s and 4.37 km/s for granulite, and 7.40 km/s and 4.21 km/s for amphibolite, respectively. 3–4 % lowering of elastic wave velocities in hydrous amphibolite compared with anhydrous granulites may be consistent with seismological observations indicating the presence of a low‐velocity layer at the slab/mantle wedge interface in subduction zones.

Study of the structure of the slab-mantle interface using reflected and converted seismic waves

Seismic waves reflected and converted at the interface between a downgoing slab and the mantle above it can provide important constraints on the velocity contrast at the interface and hence on the properties of the downgoing lithosphere. Modelling studies suggest that different physical mechanisms may contribute to the velocity contrast at different depths along the interface. To explore the possible mechanisms, we examine the amplitudes of seismic waves reflected and converted at the slab-mantle interface in Japanese subduction zones. Slab-face P-reflection amplitudes suggest a velocity contrast ranging upwards to ~ 10 per cent at depths of between 300 and 400 km. P/S amplitude ratios from ScS-to-P conversions at the slab-mantle interface depend on the depth of conversion and are large, 0.14 and 0.12, at stations SHK and MAT. The size of the velocity contrasts and amplitude ratios suggests that a model of the interface more complicated than one involving merely temperature and/or compositional effects is required. We use these observations to constrain layered models of the slab-mantle interface using Thomson-Haskell matrix modelling, comparing observations with calculated amplitude ratios. The results suggest that the crustal layer of basaltic composition, persisting to depth during subduction, plays a role in the generation of the shallow converted waves. The deeper conversions appear to require layering as well, perhaps involving eclogite at the slab surface in addition to oriented olivine in the overlying mantle and the α−β (and α−γ) phase transformations in olivine within the slab.

14 February 1983 earthquake sequence in the Shumagin Islands, Alaska

Abstract Two moderate events (MS = 6.3 and 5.6) occurred just outside a local network in the Shumagin Islands, Alaska, on 14 February 1983. These were the first events of magnitude greater than 5.5 in the region since 1979. They occurred within the subduction zone at a depth of 26 km. Of the more than 50 aftershocks that were recorded at one or more stations, 12 events ranging from mb = 2.5 to 4.0 were recorded digitally and located. Focal mechanisms were calculated for the two main events from first-motion polarities and teleseismic body-wave modeling. The mechanisms for both events show similar nodal planes, with P axes in the approximate direction of plate convergence and nearly vertical T axes. The plane of motion is not constrained by the aftershocks but the shallowly dipping plane parallel to the subduction zone interface is the preferred solution. Dynamic stress drops of 80 and 70 bars are calculated from local strong-motion records. These events occurred in a mature seismic gap. They nucleated in a shallow portion of the main thrust zone that apparently has a relatively low state of stress and thus were not capable of rupturing into the higher strength region at the base of the main thrust zone.