Search

Abstract: Song and Kawakatsu (2012) have shown that the sub-slab fast splitting pattern observed in most subduction zones can be a direct consequence of subduction of the oceanic asthenosphere that has strong radial anisotropy. This model not only explains the non-intuitive trench-parallel splitting pattern in most of subduction zones, but also predicts the trench-normal behavior (fast polarization direction sub-parallel to the absolute plate motion of the incoming plate) observed in several shallow subduction zones. The general validity of such a scenario is crucial to fundamental understandings of the development of mantle anisotropy in sub-lithosphere or/and sub-slab conditions, the nature and formation of oceanic asthenosphere as well as the flow pattern and mass transport near subduction zones. To validate this scenario, we examine SKS splitting patterns observed across the fore-arc in central Alaska. Here the fast splitting direction varies from plate motion sub-parallel near the trench to mostly trench-parallel beyond the 100km slab-isodepth contour, while being strongly variable in between. After taking into account the rotation of anisotropy symmetry in the oceanic asthenosphere with respect to the local plate motion obliquity and down-dip variations in the slab dip, we reproduce a general 90-degree switch in fast splitting direction as well as the back azimuth dependent splitting pattern across the entire fore-arc. The current validation further augments the idea that, apart from anisotropy in the mantle wedge and the subducting slab, subduction of the oceanic asthenosphere is likely to be the dominant source of seismic anisotropy in central Alaska and potentially in many subduction zones. Furthermore, this result also provides alternative views to models of seismic anisotropy in the mantle wedge and on the length scale in which the 3D mantle flow may be important. [Copyright &y& Elsevier]

Abstract: Oceanic plates subducted at trenches penetrate into the deep mantle, and encounter a structural boundary at a depth of 410km where olivine, the dominant element of mantle rocks, transforms into a higher density form wadsleyite. This transformation may be delayed within the coldest core of subducting plates (slabs) due to kinetic effects, and it has been suggested that metastable olivine may persist deeper than 410km. Using high density seismic array data in Japan, we show the direct image of the structure corresponding to this metastable olivine wedge (MOW) beneath southwest Japan. Numerical simulation of a subducting slab, including the kinetic effect of water (H2O) on the olivine–wadsleyite transformation, indicates that the presence of the imaged MOW requires an insignificant amount of water (less than 100wt. ppm) be present in the slab mantle, thus a deep dry cold slab. We infer that the transportation of water into the deep mantle occurs along the top surface of the subducting slab, but no significant amount within the slab itself. We also demonstrate that a numerical simulation including the kinetics of 660-km phase transformation can reconcile the observed deep depression of the 660-km discontinuity with a gentle Clapeyron slope. [Copyright &y& Elsevier]

Abstract: The petrological consequences of deep subhorizontal deflection (“stagnation”) of subducting slabs should affect both apparent seismic velocity structures and slab morphology. We construct kinematic thermal models of stagnant slabs and perform thermodynamic modeling of the consequent perturbation of high-pressure phase transitions in mantle minerals, focusing upon Japan as our study area. We calculate associated thermo-petrological buoyancy forces and bending moments which (along with other factors such as viscosity variations and rollback dynamics) may contribute to slab deformation. We consider effects of variations in depth of stagnation, post-stagnation dip angle, phase transition sharpness, transition triplication due to multiple intersection of geotherms with phase boundaries, and potential persistence of metastable phases due to kinetic hindrance. We also estimate seismic velocity anomalies, as might be imaged by seismic tomography, and corresponding seismic velocity gradients, as might be imaged by receiver-function analysis. We find that buoyant bending moment gradients of petrological origin at the base of the transition zone may contribute to slab stagnation. Such buoyancy forces vary with the depth at which stagnation occurs, so that slabs may seek an equilibrium slab stagnation depth. Metastable phase bending moment gradients further enhance slab stagnation, but they thermally decay after ∼600–700km of horizontal travel, potentially allowing stagnant slabs to descend into the lower mantle. Stagnant slabs superimpose zones of negative velocity gradient onto a depressed 660-km seismic discontinuity, affecting the seismological visibility of such features. Seismologically resolvable details should depend upon both stagnation depth and the nature of the imaging technique (travel-time tomography vs. boundary-interaction phases). While seismic tomography appears to yield images of stagnant slabs, discontinuity topography beneath Japan resolved by migrated receiver functions appears to be consistent with slab penetration of the transition zone. However, model slabs which bottom around ∼780–810km and then bend upwards by a few degrees can match both the tomographic and receiver-function images. [Copyright &y& Elsevier]

Abstract: We present evidence of two layers of anisotropy in Earth''s upper mantle beneath the central orogenic belt of the North China Craton from teleseismic shear-wave splitting analysis. A one-layer model is in disagreement with the well-constrained azimuth-dependent splitting parameters for stations in the central orogenic belt; however, two layers of anisotropy improve the fit to the data. At the bottom layer, the fast polarization directions are subparallel to the absolute plate motion, indicating the presence of the current strain within the subcratonic asthenosphere related to the interactions between the surrounding plates. In contrast, at the top layer, the fast polarization directions are consistent with the strikes of the local surface faults, confirming a combined contribution of the past orogenies in the lithosphere. [Copyright &y& Elsevier]

Abstract: We have developed and implemented a new grid-based earthquake analysis system that continuously monitors long-period seismic wavefield of 20–50s recorded by broadband seismometers. The new analysis system automatically and simultaneously determines the origin time, location and seismic moment tensor of seismic events within 3min of their occurrence. This system has been in operation since 2003 at the Earthquake Research Institute (http://wwweic.eri.u-tokyo.ac.jp/GRiD MT/), and the locations and origin times are usually obtained within 3s and 20km away from the earthquake catalog values determined by the Japan Meteorological Agency (JMA). In addition, moment tensor solutions are comparable to the network solutions manually obtained. This new system should enable us to monitor long-period seismic wavefield continuously further to help identifying long-period (or low-frequency) events which are undetectable by the conventional monitoring of short-period seismic wavefields. [Copyright &y& Elsevier]

The hemispherical nature of heterogeneities of the isotropic velocity and attenuation of the inner core has been extensively studied by seismic observations, but consensus on the character of the boundary between two hemispheres has not yet been achieved. To investigate the hemispherical boundary in detail, we analyze core phase data whose ray paths propagate beneath the northern Pacific. We employ a waveform inversion approach based on simulated annealing to measure the differential traveltime and the attenuation parameter of the core phases. Measured differential traveltimes for data from European stations for events in the Fiji-Tonga region show positive anomalies consistent with the eastern hemisphere model, while those from USArray for events in Indonesia show relatively small values. Ray-theory based forward modeling of the differential traveltimes is conducted by varying the shape and the transition width of the boundary. An arc-like shape boundary that connects points (0°N, 159°W) on the equator and (79°N, 110°E) in far north with a transition width of ∼600 km best explains the observed differential traveltimes rather than a sharp boundary. This model also accounts for the observed variation of attenuation parameters. • The nature of hemispherical boundary of the inner core beneath the northern Pacific was investigated. • Core phase data were analyzed via a non-linear waveform inversion method. • Forward modeling of the differential traveltime anomaly was conducted for constraining the boundary. • The arc-like-shaped boundary with ∼600 km-wide velocity transition was observed. [ABSTRACT FROM AUTHOR]

High frequency energy radiation of large earthquakes is a key to evaluating shaking damage and is an important source characteristic for understanding rupture dynamics. We developed a new inversion method, Image Deconvolution Back-Projection (IDBP) to retrieve high frequency energy radiation of seismic sources by linear inversion of observed images from a back-projection approach. The observed back-projection image for multiple sources is considered as a convolution of the image of the true radiated energy and the array response for a point source. The array response that spreads energy both in space and time is evaluated by using data of a smaller reference earthquake that can be assumed to be a point source. The synthetic test of the method shows that the spatial and temporal resolution of the source is much better than that for the conventional back-projection method. We applied this new method to the 2001 M w 7.8 Kunlun earthquake using data recorded by Hi-net in Japan. The new method resolves a sharp image of the high frequency energy radiation with a significant portion of supershear rupture. [ABSTRACT FROM AUTHOR]

Abstract: The Indochina block is located at the junction of four major plates. It has a close relationship with the uplift of Tibet plateau and the sea-floor spreading of the South China Sea in the geological evolution histories. We deployed a temporary broadband seismograph array in Vietnam from February 2000 to October 2005, and measured shear-wave splitting of core-refracted phases to investigate the upper mantle anisotropy beneath the Indochina block and adjacent regions. The observed delay times of up to 1.5s are large, indicating a mantle source for the anisotropy. Two distinct regimes of seismic anisotropy are revealed. The fast polarization directions in the northern Indochina block with a main trend of N100°E are almost consistent with the direction of both the surface displacement and the absolute plate motion. These observations suggest the anisotropy is possibly generated by the present-day asthenospheric flow caused by the India–Eurasia collision. The fast polarization directions in the southern Indochina block and the Hainan island are nearly N60°E. The anisotropy in this area may reflect a combination of the India–Eurasia collision and the past orogenies in the lithosphere, or an abrupt variation in the asthenospheric flow. [Copyright &y& Elsevier]

Abstract: At Aso volcano, Kyushu, Japan, several different types of volcanic tremor have been observed for many years. One of them is the continuous tremor, the ground vibration which has dominant frequency between 3 and 10 Hz and has approximately constant amplitudes without any clear beginning and ending. We observed the continuous tremor at Aso using short period seismometer arrays for 3 days in 1999. We locate the source of the continuous tremor by seismic array data processing. We use the semblance coefficients in order to estimate the arrival azimuth and apparent slowness by grid search. The epicenters of the continuous tremor are located around the currently active crater, and the source depths are likely to be shallower than 600 m. We find that the estimated epicenter clearly migrates synchronized with the change in the tremor amplitude. The migration often occurs periodically with a period of about 80 s, but aperiodic occurrence of the migration is also often seen. In both cases, the epicenter is located southeastward (northwestward) when the amplitude is larger (smaller). We propose that there are two or more independent tremor sources with fixed locations, and that their amplitudes modulate either aperiodically or periodically with periods nearly 80 s. The tremor signals from those sources are mixed at the arrays, and the estimated epicenter parameters vary according to which signal dominates the seismograms. The simplest model is that there are two point sources, one at west and the other at south of the crater (we call the two sources as “NW source” and “SE source”, respectively), and the amplitude of the SE source changes with time. Consequently, the SE source dominates the seismogram when the observed amplitude is larger, whereas the NW source dominates when the amplitude is smaller. We generate synthetic seismograms, and apply the location technique to them to verify the validity of the two point source model and to search the locations of two sources which can explain the observed synchronization between the amplitude and the apparent epicenter location. We find that the distance between the two sources needs to be more than 400 m to agree with the observation. We also analyze the seismic array data observed in 2001, and infer that the NW source in 1999 may be identical to the tremor source of the 2001 data. [Copyright &y& Elsevier]

We measured shear wave splitting from SKS data recorded by the transcontinental NECESSArray in NE China to constrain lithosphere deformation and sublithospheric flows beneath the area. We selected several hundreds of high quality SKS/SKKS waveforms from 32 teleseismic earthquakes occurring between 09/01/2009 and 08/31/2011 recorded by 125 broadband stations. These stations cover a variety of tectonic terranes, including the Songliao basin, the Changbaishan mountain range and Zhangguancai range in the east, the Great Xing'an range in the west and the Yanshan orogenic belt in the southwest. We assumed each station is underlaid by a single anisotropic layer and employed a signal-to-noise ratio (SNR) weighted multi-event stacking method to estimate the two splitting parameters (the fast polarization direction φ , and delay time, δt ) that gives the best fit to all the SKS/SKKS waveforms recorded at each station. Overall, the measured fast polarization direction lies more or less along the NW–SE direction, which significantly differs from the absolute plate motion direction, but is roughly consistent with the regional extension direction. This suggests that lithosphere deformation is likely the general cause of the observed seismic anisotropy. The most complicated anisotropic structure is observed beneath the southern Great Xing'an range and southwest Songliao basin. The observed large variations in splitting parameters and the seismic tomographic images of the area are consistent with ongoing lithospheric delamination beneath this region. [ABSTRACT FROM AUTHOR]

We used ambient noise data recorded by the NECESSArray, a portable seismic array deployed in Northeast China, to construct Rayleigh-wave phase velocity dispersion curves and obtain phase velocity maps for periods from 6 to 40 s. S-wave velocity structure beneath NE China is obtained by joint inversion of the phase velocities and receiver functions. Our results show good correlation between crustal velocity structure and geological features. At shallow depths the Songliao Basin is separated into two small sub-basins by a fast velocity anomaly corresponding to a southwest uplifted horst and the southern sub-basin is associated with the Kailu depression. A low-velocity anomaly extending from the lower-crust to uppermost mantle beneath the Changbaishan mountain range is consistent with P- and S-wave tomography models. The mantle low velocity has been interpreted as a local mantle upwelling attributed to the subducting Pacific slab induced return flow. Finally, the S-wave velocity in the mid-and-lower crust is lower beneath the Xing'an–Mongolia Orogenic Belt (XMOB) than beneath the Changbaishan mountain range. This was not expected since the Changbaishan mountain range has experienced more extensive volcanic activity in the Cenozoic, including the currently active Changbaishan and Jingpohu volcanoes, than the XMOB. We infer that the mid-and-lower crust of the XMOB is mainly felsic in composition and propose that the low shear-wave velocity of the mid-and-lower crust beneath the XMOB could be the consequence of removal of deep mafic crust through a continental accretion and subsequent delamination event in the Paleozoic, or via convection removal induced by the rollback of the flat subducted Paleo-Pacific slab during the middle Mesozoic. [ABSTRACT FROM AUTHOR]

In order to better understand the deep subduction geometry of the Pacific plate and genesis of intraplate volcanism in northeast China (NE China), we computed a total of 45,505 receiver functions from 788 teleseismic events recorded by 255 stations (NECESSArray temporal and permanent stations) in NE China. We used a common-conversion-point stacking (CCP) method to generate a 3D reflectivity volume beneath the study area. To position the P-to-S conversions to the correct depths, we employed 3D crustal and mantle models as references to make time to depth conversion. The 3D reflectivity volume was generated in an area between 115°–135°E and 40°–49°N, in the depth range of 300 to 800 km. We found significant topographic relief on the 660-km discontinuity across the study area. In particular, in a westward Pacific plate subduction section between 40°N and ∼45.5°N, the 660-km discontinuity is depressed by as much as ∼30–40 km along the western extension of the deep seismicity. The depression is elongated along the strike of the deep seismicity and is confined to a 200–300 km region in the E–W direction of subduction. To the west of this depression the 660-km discontinuity is uplifted by 5–10 km in a rectangular area of ∼100 km by 200 km centered at about 125°E and 43°N. In the north, the 660-km discontinuity is moderately depressed (∼20 km) in a broad area that extends further west. The high and low regions in the 660-km topographic map correlate, respectively, with low- and high-velocity anomalies in the P- and S-wave tomographic velocity images at the same depth. Our results suggest that slab stagnation might not be occurring in the southern part of the NE China, where the Changbaishan volcanic complex is located, thus the magmatism is unlikely caused by dehydration of the flat-lying Pacific slab in the transition zone. The low velocity mantle upwelling arising from a gap of stagnant slabs is a likely source that feeds the volcanic complex in NE China. [ABSTRACT FROM AUTHOR]

Abstract: We analyzed tens of thousands of receiver-function data recorded by 185 broadband seismic stations to study the crustal structure beneath the northeast China. Moho depth and average crustal ratio were measured at each station using the H–κ grid searching technique. For stations located above unconsolidated sediments, we applied the H–β method to remove strong shallow reverberations and generate subsurface receiver functions that allow for effective H–κ analysis. The measurements show that the Songliao basin has a relatively thin crust (∼31 km), and the Moho depth increases significantly from southeast (∼27 km) to northwest (∼35 km). The northwestward tilting of the Songliao basin may suggest that it was initiated by lithospheric flexure due to the load of the Great Xing'an range in the Jurassic before the large-scale extension in the Cretaceous. Moho depth varies from 26.7 km to 42.3 km across the study area, with the shallowest and deepest Moho being located at the eastern flank of the Songliao basin and the Great Xing'an range, respectively. The Moho depth correlates well with the surface topography in the western part, but not the central and eastern parts of the study area. The residual topography computed based on the Airy's isostasy model indicates that the high topography at the eastern flank of the Songliao basin, the Changbaishan volcanic center, and the southern end of the Great Xing'an range is likely dynamically supported by the upper mantle. [Copyright &y& Elsevier]

Abstract: We determined the depths of the 410-km and 660-km seismic discontinuities (called “410” and “660,” respectively, hereafter) beneath the Philippine Sea and the northwestern Pacific by the receiver function method; we used the broadband data obtained from broadband ocean-bottom seismographs. We found a very deep “660” at 691km in the stagnant slab beneath the Philippine Sea. In the surrounding Philippine Sea regions and northwestern Pacific, the “660” appears at depths of 659–674km. Comparison of the “660” depth and P-velocity anomalies indicated that temperatures in the stagnant slab are lower than that in the mantle transition zone (MTZ) beneath the northwestern Pacific by about 500K. The temperature in the MTZ beneath the Parece Vela Basin and West Philippine Sea Basin, which surrounds the stagnant slab, is lower than that in the MTZ beneath the Pacific by about 200K. The water content in the stagnant slab (∼0.2wt% H2O) is higher than that in the other regions; however, the large uncertainty in the estimated water content (±0.2–0.3wt% H2O) prevents us from arriving at a definitive conclusion regarding the presence of water in the stagnant slab. Therefore, it is desirable to use other geophysical parameters (e.g., electrical conductivity) along with the seismic parameters in order to obtain an accurate estimate of the water content in the MTZ. [ABSTRACT FROM AUTHOR]

Abstract: In order to investigate the continuous volcanic tremor of Aso Volcano in Japan, we performed a series of temporary short-period seismic array observations near the Nakadake first crater (the active crater) during five years from 1999 to 2003. We deployed in all of the temporary observations a seismic array at the same location about 700 m west of the active crater, in order to investigate long-term changes in the tremor activity. In 1999 and 2003, another array was simultaneously deployed at a different location 700 m north of the crater to help locate the tremor sources. We developed a frequency domain semblance method and applied it to the waveform data of the frequency range where the continuous tremor is dominant (3–6 Hz). We measured arrival azimuths and slownesses of the continuous tremor signals as functions of frequency, which are then used to locate the epicenters of the tremor signals corresponding to the principal peaks of the power spectra. For the observations in 1999 and 2002, the continuous tremor amplitudes are relatively small, and the slowness of the tremor signal observed at the west array takes a local minimum (0.5 to 0.6 s/km) near the frequency (∼4.7 Hz for 1999 and ∼4.8 Hz for 2002) which corresponds to the highest spectral peak. This implies that body waves dominate the tremor signals at the west array around the frequency. The tremor epicenters corresponding to 4.7 Hz for the observation in 1999 are located at the west rim of the currently active crater. While the surface crater activity of Aso remains low and the tremor activity is not clearly linked with the surface activity until early 2003, a close link between the tremor and crater activity appears in the middle of 2003, when a small phreatic eruption occurred a month before the array observation (July 10, 2003). Tremor signals of the observation in 2003 show a large spectral peak (4.2 Hz) where the slowness measured for the west array is very large (1.1 s/km), clearly suggesting that surface waves are dominant. The epicenter is again located at the western rim of the active crater. We interpret these observations as follows: in 1999 and 2002 when the surface activities of Aso were low, the continuous tremor excitation was deep and inactive. In the middle of 2003 when Aso Volcano became active with a series of phreatic eruptions a shallower tremor source was activated, possibly masking the deeper sources. This shallowing of the dominant tremor source could be due to the increase in the volcanic gas flow rate triggered by the phreatic eruptions. [Copyright &y& Elsevier]

Abstract: We have determined the three-dimensional shear wave speed structure of the upper mantle in and around the Philippine Sea region using seismograms recorded by dense land-based and long-term broadband ocean bottom seismographic stations. We used a surface wave tomography technique in which multimode phase speeds are measured and inverted for a 3-D shear wave speed structure by incorporating the effects of a finite frequency and ray bending. The new ocean bottom data provided us with improved spatial resolution (~300 km) in the Philippine Sea region. In the upper 120 km, the shear wave speed structure is well correlated with seafloor age. At depths greater than 160 km, fast anomalies of the subducting Pacific Plate are clearly defined. We also found slow speed anomalies beneath the Kyushu–Palau ridge at depths greater than 120 km. Along the Izu–Bonin(Ogasawara)–Mariana arc, we have detected three separate slow anomalies in the mantle wedge at depths shallower than 100 km beneath the back arc. Each anomaly has a width of ~500 km. Moreover, these three anomalies have a close relationship with the three groups of frontal and rear arc volcanoes having distinct Sr, Nd, and Pb isotope ratios. We suggest that each of the anomalies is a site of large-scale flow of upper mantle into the mantle wedge, and that each already contains a component from the adjacent subducting slab. [Copyright &y& Elsevier]

We installed five broadband seismometers around Mt. Usu, Hokkaido, Japan, just before the first surface eruption on March 31, 2000. By using these broadband data with short-period seismograms recorded by Japan Meteorological Agency and Institute of Seismology and Volcanology of Hokkaido University, we located 590 earthquakes associated with the 2000 eruption of Mt. Usu from March 31, 12:58 to April 5, 2000, including 41 low-frequency earthquakes. Low-frequency earthquakes have clear predominant frequencies of about 1.0–1.5 Hz. While the seismicity of tectonic earthquakes was still active on April 5, there were almost no low-frequency events after April 2. All the low-frequency earthquakes occurred within a vertical and narrow zone near the craters of eruption, and they are shallower than 4 km. In contrast, the active area of tectonic earthquakes spreads out in a large area, particularly deep in the south of the craters. From these temporal and spatial patterns, together with GPS and gravity measurements that support the termination of magma activities before April 2, we suggest that low-frequency earthquakes were caused by magma or hydrothermal activities beneath the volcano while tectonic ones were caused by the regional tectonic stress affected by the intrusion of magma body near the craters. Related to long-term magma activities of this eruption, four deep crustal low-frequency events were identified on October 17, 1998 just in the south of Mt. Usu in the depth range of 20–30 km. As proposed for other active volcanoes, these deep crustal events may represent a part of deep magma activities beneath Mt. Usu. [Copyright &y& Elsevier]