S43B-0994 1340h
Dipping Structures beneath Japan, inferred from analysis of teleseismic waves.
Northern Japan: I analyze high frequency receiver functions from Hinet and Fnet stations throughout Northern Japan to discern velocity contrasts within the subducting Pacific Plate. The study region contains an upper and lower band of Wadatti-Benioff seismicity associated with the subduction of the Pacific Plate and Philippine Sea Plate. The basaltic upper crust of the subducting plate is observed to have lower P-wave and S-wave velocities than the surrounding mantle material and this roughly coincides with the upper band of seismicity. A second velocity contrast is believed to exist in the lower band of seismicity as observed recently is tomography studies. This velocity contrast consists of a lower P-wave velocity coupled with a high S-wave velocity. Synthetics receiver functions from such a model indicate a noticeable conversion produced by the lower velocity contrast, however, the timing of the phase occurs in the region that contains strong crustal multiples. Receiver Functions for Fnet data are obtain from approximated 140 events spanning from 1999 to 2004. We use 10 Fnet stations in the study. The data is supplemented with data from Hinet stations. I compare receiver function stacks for Fnet and Hinet stations to synthetic models to determine whether models that contain the lower velocity contrast can better fit the observed data than a model that excludes the contrast. Southern Japan: I analyze receiver functions for two regions, Kii Peninsula, where an active offshore-onshore seismic experiment will take place this year, and the Tokai region, where a similar experiment has taken place recently. Information from the wide angle experiments provides constraints on crustal structure allowing a clearer analysis of the subducting plate structure. In both regions, the subducting Philippine Sea plate can clearly be observed. Further analysis is required to determine if velocity contrasts exist within the subducting plate.
S43B-0995 1340h
Receiver Function Analysis in and around the Boso-Peninsula, Central Japan
We analyzed receiver functions to estimate a crustal structure in and around the Boso Peninsula, which is located east of Tokyo, Japan. The area includes a part of the source region of the 1923 Kanto Earthquake (M=7.9). The seismicity beneath the Boso Peninsula is reported to be lower than that around the area, and large events did not occur in recent 50 years. Several slow-slip events were also detected by continuous GPS observations. We applied the receiver function estimation technique using multiple-taper methods [Park and Levin, 2000]. We used seismograms of earthquakes occurred in teleseismic distances (epicenter distances between 30 to 90 degrees) and regional event that occurred in the subducting Pacific plate. In the Boso Peninsula, we installed a dense seismographic array which consists of 30 stations along a line of the northeast-southwest direction with a station spacing of about 2 km to 10 km. The array has observed teleseismic and regional events since February 2004. We also used seismograms recorded by regional seismographic networks from August 2002. A velocity structure beneath the Boso area is very complex. The Philippine Sea plate is subducting northward and the Pacific plate is subducting westward. We show some preliminary results although a recording period of a half year is not long enough for reliable results. First, we suggest that positive pulses at 4 -6 s correspond to the upper boundary of the northward subducting Philippine Sea plate. The depths expected from the positive pulses are relatively consistent with those estimated by the controlled source experiment. However, the direct P wave at 0 seconds is not clear and a thick sediment layer with a low velocity near surface delays the arrival by about 1-2 s. Some converted phase observed in local events also seems to appear on a common boundary. We expect to clarify the seismic velocity structure of deeper part of the crust to estimate the location of the upper plate boundary of the subducting Philippine Sea plate from these conversion phase analyses.
S43B-0996 1340h
A synthesis on lower mantle reflectors/scatterers and discontinuities beneath the western Pacific subduction zones
Strong seismic reflectors/scatterers and discontinuities have been observed in the lower mantle at various depths beneath western Pacific subduction zones. The lateral extension of these anomalous structures, however, is not well constrained. The information may hold the key to the understanding of the nature of these seismic structures as well as the related mantle processes. For example, are these seismic anomalies primordial chemical reservoirs, or pieces of subducted oceanic crusts or even parts of a highly heterogeneous global compositional boundary? In this study, we used teleseismic S to P and P to S converted waves to determine the locations and depths of lower reflectors/scatterers and discontinuities in the western Pacific. Our data are collected from several seismic arrays (J-array and Hinet in Japan, the Kaapvaal broadband seismic array, and short-period arrays in the western US) from recent deep earthquakes occurred in the study regions. The spread of conversion points among various arrays allow us to better constrain the lateral extension of the anomalous structure. For example, the mid-mantle discontinuity observed by {\it Niu and Kawakatsu} [1997] at the 110$\deg$E, 5$\deg$S appears to extend to further south and west by at least 100 km once the Kaapvaal data is combined. For the islands in the study region where seismic arrays are available, for example Japan and Indonesia, we also conduct receiver-function imaging to improve geographic coverage. By combining different types of seismic data, we expect to achieve a synthesized view of anomalous structures at middle to lower mantle depths.
S43B-0997 1340h
Detailed Structure and Thickness of Upper Mantle Discontinuities in the Tonga Subduction Zone From Regional Broadband Arrays
Recordings of deep Tonga earthquakes from two arrays of 12 broadband seismographs each in the Fiji and Tonga islands are stacked and searched for reflections and conversions from upper mantle discontinuities in the Tonga subduction zone. The arrays operated as part of the Seismic Arrays in Fiji and Tonga (SAFT) experiment from July 2001 to August 2002. In comparison with the commonly used teleseismic approaches, the short path lengths for the regional data provide smaller Fresnel zones and high frequency content for precise mapping of discontinuity topography and sharpness. This is particularly important for a subduction zone, where variations in temperature and water content may be expected which should cause changes in the elevation and sharpness of the discontinuities. We studied the phases $s410p$, $P660p$ and $S660p$. To enhance these low-amplitude phases, deconvolved seismograms from each event/array pair are aligned on the maximum amplitude of the direct $P$ wave and subsequently slant-stacked. For the 410-km discontinuity, the results show no systematic variations in depth with distance to the cold slab. The 660-km discontinuity varies between 656 and 714 km in depth. For the southern and central parts of the subduction zone, the largest depths occur in the core of the Tonga slab. For the northern part, two separate depressions of the 660 are observed. These anomalies are interpreted as being induced by the active, steeply subducting Tonga deep zone and a subhorizontally lying remnant of subducted lithosphere from the fossil Vityaz trench, respectively. Interpreting the deflections of the 660 in terms of local temperatures implies a thermal anomaly at 660 km depth of $-800$ to $-1200^{\circ}$K for the Tonga slab, and $-600$ to $-950^{\circ}$K for the piece of the Vityaz lithosphere. Except for the southern region where it thickens, the Tonga slab seems to penetrate the 660 with little deformation. Waveform modeling susggests that both the 410 and 660 discontinuities are sharp. The 410 thickness is estimated to be at most $\sim$10 km outside as well as within the slab, and the 660 width $\sim$5 km. This suggests that the subduction process has probabily little effect if any on the sharpness of the discontinuities.
S43B-0998 1340h
Array analysis of Taiwan short-period data and the transition zone morphology in the Fiji-Tonga region
One hundred plus short-period, 3-component stations operated by the Central Weather Bureau of Taiwan form a perfect array for teleseismic events from Fiji-Tonga subduction zone to study the structure of the lower mantle and the transition zone. By beamforming the seismograms with correct slowness, secondary phases whose magnitudes may approximate the noise level of individual records can merge as coherent energy against the background of destructive interference of the noise. We analyzed only one and a half year of CWB data using array techniques. Vespagrams were constructed for events with depth $>$ 200 km in search for consistent secondary phases that bear significance of the dynamics and structures of the mantle. Up to date, we have identified coherent energy that may be attributed to conversion/reflection of P waves at 410 km and 660 km discontinuities, with delays corresponding to tens of km. The delays of these phases, i.e., S660P, s410P, and p410P, indicate the variations in the thickness of the transition zone deformed or penetrated by the subduction of the Tonga slab. With limited samples we observed large variations in the thickness along the strike of the slab, but verification of a systematics awaits further analyses. Some of these results were bolstered by the frequency-wavenumber analysis. The PcP phase is identifiable, whereas we cannot identify confidently any PcP precursors that may be generated by refraction from a positive discontinuity above the core-mantle boundary. It is likely that such a discontinuity is either fuzzy or absent in the vicinity mid-way along the path. Other energy that stands prominent on the vespagram remains unexplained with structures in the mantle.
S43B-0999 1340h
Acoustic Velocity Of The Sediments Offshore Southwestern Taiwan
Along the Manila Trench south of 21>XN, deep-sea sediments are being underthrusted beneath the Taiwan accretionary prism which is composed of the Kaoping Slope and Hengchun Ridge. Offshore southwestern Taiwan, foreland sediments and Late Miocene strata of the Tainan Basin are being accreted onto the fold-and thrust belt of the syn-collision accretionary wedge of the Kaoping Slope. The Kaoping Slope consists of thick Neogene to Recent siliciclastics deformed by fold-and-thrust structures and mud diapers. These Pliocene-Quaternary sediments deposited in the Kaoping Shelf and upper slope area are considered to be paleo-channel deposits confined by NNE-SSW trend mud diapiric structure. Seismic P-wave velocities of the sediment deposited in the Kaoping Shelf and Kaoping Slope area are derived from mutichannel seismic reflection data and wide-angle reflection and refraction profiles collected by sonobuoys. Sediment velocity structures constrained from mutichannel seismic reflection data using velocity spectrum analysis method and that derived from sonobuoy data using tau-sum inversion method are compared, and they both provide consistent velocity structures. Seismic velocities were analyzed along the seismic profile from the surface to maximum depths of about 2.0 km below the seafloor. Our model features a sediment layer1 with 400 ms in thickness and a sediment layer2 with 600 ms in thickness. For the shelf sediments, we observe a linear interval velocity trend of V=1.53+1.91T in layer1, and V=1.86+0.87T in layer2, where T is the one way travel time within the layer. For the slop sediment, the trend of V=1.47+1.93T in layer1, and V=1.70+1.55T in layer2. The layer1's velocities gradients are similar between the shelf (1.91 km/sec2) and the slope(1.93 km/sec2). It means layer1 distributes over the slope and shelf widely. The result of the sediment velocity gradients in this area are in good agreement with that reported for the south Atlantic continental margins.
S43B-1000 1340h
An Investigation of the Subsurface Structure and P- and S-wave Velocities in the Taipei Basin, Taiwan
Over the past ten years, the Taiwan Central Geological Survey has conducted a wide-range of investigations of the Taipei basin by drilling over 30 wells (100-700m). During the same period, we have also scanned the basin area (20$\times$20 km) using over 300 shallow P-wave reflection seismic lines and many S-wave surveys (including S-wave reflection and surface-wave dispersion analysis: MASW). The purpose of this report is to present our compilation of these seismic data, incorporating it with the borehole drilling results to better describe the Tertiary basement, the Quaternary layers above the basement, and their P- and S-wave velocities. It is found that : 1) the deepest part of the basin basement is probably at the outlet of the Tanshui river, i.e., at the NW corner of the basin; 2) the Kanchiao fault forms a boundary bisecting the basin into a deep NW part and a shallow SE part; 3) the Sungshan and Chingmei formations are relatively flat deposits on the top, which implies the existence of a quiet deposition period since about 100,000 years ago; 4) the P- and S-wave velocities inside the basin are 1500-2200 m/s and 170-880 m/s in sediments, and 3000 m/s and 1500 m/s for the basement, respectively. A detailed 3D P- and S-wave velocity distribution is now available for other research in the Taipei basin.
S43B-1001 1340h
Anomalous Pn waves observed in the eastern Taiwan: indication of a thin crust beneath the Longitudinal Valley
We have found significant Pn waves for raypaths traveling along the Longitudinal Valley of eastern Taiwan with epicentral distances less than 200 km, where is supposed to be the suture zone of the arc-continent collision between Eurasia and Philippine sea plates. The waveforms analyzed in this study are generated from recent earthquake that occurred around 20 km depth at both ends of the Longitudinal Valley. This azimuth dependent Pn waves are characterized with its low amplitude wavetrain and the earlier arrival times. As revealed in the travel-time curve plot, the apparent velocity for this series of Pn arrivals is $\sim$7.9 km/s that is comparable with the crust-mantle boundary P wave velocity. Considering the focal depths and associated epicentral distances, on possible explanation of this observation is that the moho depth beneath the Longitudinal Valley is somehow shallower than elsewhere on land of Taiwan. This result is consistent with the latest tomographic image of Taiwan by Kim and Chiu (2004).
S43B-1002 1340h
Waveform modeling of Rayleigh waves and shear-velocity structure of the Southeast Indian Ocean upper mantle
Recent work of dynamic modeling demonstrated that an anomalously cold blob under the mid-ocean ridge tends to collapse by its own weight despite in the buoyant environment. However, there are two conditions under which this relatively dense blob can achieve dynamic balance with the upward flow field and be even lifted to the surface. One is when the low viscosity asthenosphere is absent, and another with a fast spreading rate or a flow field that is enhanced by additional tectonic forces, such as continental rifting. A test ground for this dynamics vs. kinematics tug-of-war is in the Australian-Antarctic Discordance (AAD). Earlier dynamic models have suggested that the associated topography and geochemistry result from a Cretaceous detached slab redrawn to the surface during the southeast Indian Ocean opening. Recent tomographic studies have delineated this structure with increasing details. We employ the waveform modeling technique of Cara and Leveque (1987) to invert for the path-averaged shear velocity structure from Rayleigh wave waveforms recorded by GSN, Geoscope, and Skippy stations for the ridge events, and then construct a 3-D image of the region. The method takes advantage of the secondary observable that is the cross-correlation function of the observed and synthetic seismograms, which stabilizes the inversion process for a multimode Rayleigh wave seismogram. Short path ensures relatively small Fresnel zone and better resolution. Preliminary results show that in addition to the off-axis anomaly to the west of the AAD zone as previously identified, positive anomalies underly the Australian continent-ocean boundary at 100-150 km depths. Seismological depictions of the upper mantle beneath AAD can aid formulation and distinction of dynamic models.
S43B-1003 1340h
Investigating topography of transition zone discontinuities in the Southwest Pacific using a migration method
Due to the temperature dependence of the phase transitions of upper mantle minerals the upper mantle discontinuities hold insight into Earth dynamics and can give information on temperatures in and around of regions of slabs. In regions of cold downwelling (i.e. slabs) the discontinuity at 410 km depth is expected to be shallower whereas the 660 km discontinuity should be deeper. The behaviour of the discontinuity at 520 km depth has yet to be analysed in more detail. We employ a migration method and underside reflections of PP off these discontinuities to investigate their topography in the upper mantle beneath the Southwest Pacific. The events are reflected in regions near and within slabs and the grid of potential reflectors spans from 115 to 150deg E and 0 to 30 deg N in 1degree steps and layers in steps of 10 km from the surface to 730 km. With this method the existence of the 520 km discontinuity is confirmed in this region although the reflector appears to be much deeper (at 560 km depth). In addition discontinuities at 220 km and 350 km are observed. The deflections of 660 and 410 km are studied in more detail. The underside reflection P660P that was recently considered to be absent is visible with the migration method in these regions. The results show that the 660 km discontinuity seems to be deeper up to 40 km whereas the 410km discontinuity seems to be only slightly shallower near subduction areas and the data show laterally varying topography on all observed discontinuities. Using the depths of the discontinuities we are attempting to constrain the temperature and Clapeyron slope within and in the vicinity of the slabs.
S43B-1004 1340h
Crustal and Upper Mantle Structure Beneath the Cape Verde Islands From Teleseismic Receiver Functions.
Cape Verde is a hot spot island archipelago around 450 km west of the westernmost point of West Africa. The islands stand in 4 km of water on a roughly circular swell 2 km shallower than expected for crust of its age. The aim of this work is to provide seismic constraints on the nature of the swell, which may be attributed to thermally less dense, hot plume material or compositionally less dense, melting residue. We deployed a temporary network of broadband stations in Cape Verde to record teleseisms to study the shallow structure beneath the entire archipelago. Receiver function analysis indicates that a $\sim$20 km thick crust exists beneath all the islands, under which wavespeeds increase and density decreases in a layer extending to around 80 km depth. There does not appear to be any simple relationship between the layer thickness and the onset of volcanism, suggesting that the layer is not caused by simple flow of less dense material from a plume source. We explore a suite of models that can account for the receiver function profiles, the 2 km bathymetric anomaly, and the volume of material erupted to form the island edifices. We find that temperatures $>$300 degrees hotter than ambient mantle or melt depletion of $\sim$10% can explain them. The results suggest that the bathymetric anomaly is due to localized melting beneath the islands which uplifts them and spreads away from the individual volcanic centers, not a single source of buoyancy.
S43B-1005 1340h
Seismic Imaging of Rifted Margins in the Southern Gulf of California
We present preliminary velocity models from two wide-angle seismic transects located in the southern Gulf of California along with coincident seismic reflection images. Continental rupture in the Gulf of California occurred recently and the conjugate margins can be studied with little uncertainty in plate reconstruction. Therefore it is considered an ideal place to analyze patterns of crustal extension and rift magmatism. The seismic reflection data were acquired with a 6-km-long streamer, 480-channel, aboard R/V Maurice Ewing, and ocean-bottom seismographs (OBSs) were deployed from R/V New Horizon. Onshore seismographs were also deployed along onshore extensions of the transects. Transect 5W runs northwest-southeast from the Los Cabos block, at the southern tip of the Baja California Peninsula, to the East Pacific Rise in the mouth of the Gulf. Transect 0E runs northeast-southwest from the hills of Sierra Madre in mainland Mexico near Mazatlan to approximately 115 km into the Gulf of California. Preliminary velocity models from Transect 5W margin show an abrupt transition from a 25-km-thick continental crust to an oceanic crust of normal thickness. Transect 0E crosses what is believed to be extended continental crust and lies in the initial direction of extension characteristic of the protogulf.
S43B-1006 1340h
Finite-Difference Modeling of Seismic Reflection Data in a Hard Rock Environment: An Example from the Mineralized Otago Schist, New Zealand
The interpretation of seismic reflection data in non-sedimentary environments is problematic. In the Macraes Flat region near Dunedin (South Island, New Zealand), ongoing mining of mineralized schist has prompted the development of a seismic interpretation scheme that is capable of imaging a gold-bearing shear zone and associated mineralized structures accurately to the meter scale. The anisotropic and complex structural nature of this geological environment necessitates a cost-effective computer-based modeling technique that can provide information on the physical characteristics of the schist. Such a method has been tested on seismic data acquired in 1993 over a region that has since been excavated and logged. Correlation to measured structural data permits a direct comparison between the seismic data and the actual geology. Synthetic modeling utilizes a 2D visco-elastic finite difference routine to constrain the interpretation of observed seismic characteristics, including the velocity, anisotropy, and contrast, of the shear zone structures. Iterative refinements of the model result in a more representative synthetic model that most closely matches the seismic response. The comparison between the actual and synthetic seismic sections provides promising results that will be tested by new data acquisition over the summer of 2004/2005 to identify structures and zones of potential mineralization. As a downstream benefit, this research could also contribute to earthquake risk assessment analyses at active faults with similar characteristics.
S43B-1007 1340h
Shallow Structure of the Dead Sea Transform From Pre-stack Migration of DESERT Data
The Dead Sea Transform (DST) is a major active continental shear zone at the boundary between the African and Arabian plates. The DESERT project was conducted as the first crustal-scale geoscientific investigation of the DST fault system. Here we focus on the seismic imaging of shallow structures in the close vicinity of the Arava fault (AF), the dominant fault of the DST in the study area. The seismic data used in this study represent a sub-set selected from a 100 km long vibroseis reflection line across the DST. Whereas the entire data set provided images of the whole-crustal reflectivity, the selected data were processed with special focus on the detailed, preferentially steep-dipping structures of the upper 5 km in the region around the Arava and Zofar faults. Firstly, the data were analysed with a semblance-based algorithm to detect and extract the events with strongest coherency. Then, pre-stack depth migration was carried out using isochrone calculations for a 2-D tomography-based velocity model. In the final step, the migrated reflection events were analysed and selected by consideration of several attributes from the semblance analysis and imaging procedure. The results establish a link between the crustal-scale reflection image and the near-surface studies including seismic velocity tomography and inversion of densely measured magnetotelluric (MT) data. Increased sub-horizontal reflectivity correlates with features of the MT-based conductivity model. Steep-dipping structures are imaged in the vicinity of the AF and follow the horst-like anomalies of the velocity tomogram.
S43B-1008 1340h
Seismological Signature of Chemical Differentiation of Earth's Upper Mantle
Chemical differentiation from a primitive rock (such as pyrolite) to harzburgite due to partial melting and melt extraction is one of the most important mechanisms that causes the chemical heterogeneity in Earth's upper mantle. In this study, we investigate the seismic signature of chemical differentiation that helps mapping chemical heterogeneity in the upper mantle. The relation between chemical differentiation and its seismological signature is not straightforward because a large number of unknown parameters are involved although the seismological observations provide only a few parameters (e.g., $V_P$, $V_S$, $Q_P$). Therefore it is critical to identify a small number of parameters by which the gross trend of chemical evolution can be described. The variation in major element composition in natural samples reflect complicated processes that include not only partial melting but also other complex processes (e.g., metasomatism, influx melting). We investigate the seismic velocities of hypothetical but well-defined simple chemical differentiation processes (e.g., partial melting of various pressure conditions, addition of Si-rich melt or fluid), which cover the chemical variation of the natural mantle peridotites with various tectonic settings (mid ocean ridge, island arc and continent). The seismic velocities of the peridotites were calculated to 13 GPa and 1730 K. We obtained two major conclusions. First is that the variations of seismic velocities of upper mantle peridotites can be interpreted in terms of a few distinct parameters. For one class of peridotites which is formed by simple partial melting (e.g. mid-ocean ridges peridotites), seismic velocities can be described in terms of one parameter, namely Mg$\#$ (=Mg/(Mg+Fe) atomic ratio). In contrast, some of the peridotites in the continental (cratonic) environment with high silica content and high Mg$\#$ need at least two parameters (such as Mg$\#$ and Opx$\#$ (the volume fraction of orthopyroxene)) are needed to characterize their seismic velocities. Second is the jump of seismic velocity at 300 km in harzburgite that is caused by orthorhombic (opx) to high-pressure monoclinic phase transition in $MgSiO_3$ pyroxene. If opx-rich harzburgite (the maximum content of opx in continental harzburgite is $\sim$45 vol$%$) exists at around 300km, the maximum contrast of jump would be 2.5 $%$ for $V_S$ and 0.9 $%$ for $V_P$. This phase transition will correspond to the seismological discontinuity around 300km (X-discontinuity).