DI31B-1791
Imaging Upper Mantle Discontinuities by 3D Kirchhoff Migration of pXP, sXP and sXSH Reflections with Seismic Illumination Compensation
Teleseismic recordings of depth-phase precursors, pXP, sXP and sXSH, that travel upward from deep slab earthquakes and reflect from the underside of seismic velocity discontinuities at depths x above the sources, are analyzed using Kirchhoff migration with seismic illumination correction. The illumination weighting takes into account the earthquake radiation pattern, geometrical spreading of the wavefront and the earthquake-station geometry for each precursor relative to the reference depth phase. Building on a large-scale migration for the mantle wedge reflectivity structure under Tonga, we investigate the Java subduction zone using this imaging algorithm, finding multiple reflections from depths of around 100 km, 200 km, 300 km, 350 km and 420 km. Some of these reflectors are horizontally localized or intermittent. The presence of complex reflectivity in the mantle wedge near Java is qualitatively similar to that for Tonga. In this presentation, we consider additional subduction zones where our imaging is viable. The causes of the mantle wedge discontinuities are not definitively known, but are generally attributed to fluids expelled from the slab generating suites of hydrous phase transitions. Seismic tomography provides models of large-scale "smooth" velocity variations whereas our imaging algorithm produces small-scale "rough" components of the medium velocity field. Combined analysis of the tomographic and migration images holds promise of unveiling the complex mineralogy and dynamics of mantle wedges.
DI31B-1792
Bridging the Gap Between Seismic Observations and Mineral Physics Interpretations: a Hypothesis for Hydrous Majorite Associated With a Stagnant Slab Near the 660 km Discontinuity
Our broadband seismic body waveform analyses have determined a fairly broad region of high velocity anomaly (HVA) in the northwestern Pacific suggesting that the subducting plate is lying flat or piling up in the upper mantle transition zone (MTZ). That is in agreement with long- wavelength tomography studies although we found that a typical wavelength of HVA is shorter, and the total volume of the stagnant slab is considerably less than the tomographic image and much smaller than that of subducted plate during the known subduction history (Tajima and Grand, 1998). We also found variation of the discontinuity depth (~660 to 690 km) with HVA and a highly localized low velocity anomaly zone (LVAZ) near the 660 km discontinuity (Tajima and Nakagawa, 2006). We postulated a variable distribution of geochemical properties associated with a cold stagnant slab at the bottom of the MTZ where the contrast of a hydrous garnet-rich layer (subducted crust of MORB origin) versus bulk peridotite is involved. The estimated seismic properties in the MTZ, which are not unambiguous, can be checked for various conditions and validated through synthetic high pressure (P) and temperature (T) experiments. Two recent experimental studies under high P and T conditions provide supportive feedback for this hypothesis, i.e., a positive Clapeyron slope for hydrous garnet-perovskite phase transformation (Sano et al., 2006) and weaker hydrous garnet than peridotite in the MTZ (Katayama and Karato, 2008). On the other hand, seismic velocity measurements in laboratory experiments show that the garnetite with a MORB composition has significantly lower P- and S-wave velocities than wadsleyite and ringwoodite when these data are extrapolated to the conditions of the MTZ (Kono et al., 2007). However, the ultrasonic P- and S-wave velocity measurements of dry materials were performed at pressures up to 14.3 GPa at room temperature, and then the estimates of elastic properties for the MTZ condition are subjected to large uncertainties. They pointed out that the MORB majorite has very different bulk and shear moduli from those of the Mg end-member garnets and suggest possible misleading interpretations for seismic wave speed structures should they adopt elastic properties of simple end-members. Irifune et al. (2008) also have shown that majorite exhibits substantially lower seismic velocities relative to wadsleyite and ringwoodite based on combined in situ X-ray and ultrasonic measurements under the P and T conditions of the MTZ. Nonetheless, they suggested the possibility of progressive formation of CaSiO3-rich perovskite (which has high velocities) from the majorite phase leaving possible support for the increase of the seismic velocities at the bottom of the MTZ. We will present results of seismic waveform modeling and discuss our hypothesis for flattened HVAs with or without a depression of the 660 km discontinuity that the hydrous majorite (subducted crust, represented by the M2.0 model) is located next to the bulk peridotite (represented by M3.11 model) at the bottom of the MTZ is plausible given the present state of rheological knowledge.
DI31B-1793
What is the Density Distribution in Earth's Mantle?
The nature of the density distribution within the Earth is of major importance in understanding Earth processes yet is surprisingly poorly constrained. The main information comes from the mass and moment of inertia of the Earth, the frequency properties of the low order normal modes and satellite observations of the gravity field. An adiabatic profile through the lower mantle is entirely compatible with such observations but difficult to reconcile with the expected behaviour form mantle convection. Indeed numerical modelling would favour sub-adiabatic profiles whilst recent mineral physics results press for a super-adiabatic average state. Three-dimensional variations in density are frequently inferred by scaling relationships to velocity heterogeneity inferred from seismic tomography, yet such scaling depends on rather specific assumptions about the nature of heterogeneity. The reconciliation of the full range of density information is needed to provide new insights into Earth processes. In particular we need to consider what other sources of information we can bring to bear e.g. neutrino tomography?
DI31B-1794
Frequency dependence of attenuation: new measurements from normal modes and their geophysical implications
Constraining the frequency-dependence of intrinsic seismic attenuation in the Earth is crucial for: 1. correcting for velocity dispersion due to attenuation; 2. constructing attenuation and velocity models of the interior using datasets with different frequency contents; and, 3. interpreting lateral variations of velocity and attenuation in terms of temperature and composition. Frequency-dependence of attenuation q can be represented by a power law q ∝ q0 ωα. Despite its importance, efforts at determining α from surface-wave and free oscillation data have been thwarted by the strong trade-offs between the depth- and frequency-dependence of attenuation. We develop and validate a new method that eliminates this tradeoff, allowing a direct estimation of effective frequency dependence of attenuation. Using normal mode and surface wave attenuation measurements we introduce a new absorption band model between 80 and 3000s. Unlike the classical frequency independent model (Kananori & Anderson 1977) and the constant α model proposed by Minster & Anderson (1981), we suggest a model of the absorption band where α varies with frequency. It is 0.3 at periods shorter than 200s, it decreases to 0.1 between 300 and 800s, and becomes negative at periods longer than 1000s. We discuss the implications for dispersion corrections and show that for an α value of 0.3, the assumption of frequency-independent attenuation will result in 25% error for a frequency ratio of 10 and a 50% error for a frequency ratio of 100. Finally we also discuss the implications of our model for interpretations of lateral variations of seismic velocities in the mantle.
DI31B-1795
Continent-wide tomographic maps of Rayleigh-wave attenuation at intermediate periods in Eurasia inferred from 1-Hz Lg coda Q variation
We have used our recently developed continent-wide maps of 1-Hz Lg coda Q and its frequency dependence (Qo and η respectively) to develop a new method for estimating continent-wide Rayleigh-wave attenuation coefficients (γ) at intermediate periods. We estimated γ across Eurasia at periods of 5, 10, 20 and 50 s, using mapped values for Qo and η combined with an empirically-derived multiplicative factor for η. The method assumes that every mapped value of Qo, as suggested by earlier studies, approximates the average value of crustal shear-wave Q with depth there. The frequency-dependence parameter (η) of Lg coda Q varies between 0.3 and 1.0 across Eurasia with no predictable pattern, thus requiring us to determine an empirical relation between the frequency-dependence parameter of Lg coda Q at 1 Hz and the depth variation of average crustal shear-wave Q. Knowledge of Qo and this empirical factor allow us to map Rayleigh-wave attenuation coefficient variation across Eurasia. 5-s, 10-s, 20-s and 50-s Rayleigh-wave attenuation coefficients vary, respectively, between about 0.5 and 5.0 x 10−3 km−1, 0.2 and 2.2 x 10−3 km−1, 0.1 and 1.1 x 10−3 km−1, and 0.06 and 0.3 x 10−3 km−1. Rayleigh-wave attenuation coefficents (γ) at shorter periods (5 and 10 s) are dominated by small-scale sedimentary features while longer-period waves are insensitive to those features but are affected by more deep-seated crustal features. Variations shown by the γ maps, especially those for 20- and 50-s periods, show many similarities to, but also some significant differences from, the variations of our Qo map for Lg coda. The mapped 20- and 50-s Rayleigh-wave attenuation determinations, like those for our Lg coda Q map, show that attenuation is highest in regions of recent tectonic activity and lowest in old cratons. In some regions γ provides improved the ability to delineate crustal features that are only marginally resolved in the Qo map.
DI31B-1796 [WITHDRAWN]
Characteristics of Lateral Heterogeneities with Thermal and Chemical Origins in Pyrolitic Lower Mantle
Seismic tomography reveals various degrees of lateral heterogeneities of seismic velocities in the lower mantle, especially at the bottom of the lower mantle under Central Pacific and beneath Africa, where long wavelength anomalies have been consistently observed. The ratios of the relative change (scaling) of shear (VS) to compressional (VP) wave velocities (RSP), density to shear wave (RρS), and bulk sound speed to shear wave velocity (RCS) are often used as diagnostic for discrimination the origins of lateral heterogeneities, however, obtaining conclusive results of RSP, RρS, and RCS from current mineral physics data is still a challenge due to the lack of direct measurements on the elastic and anelastic properties at the conditions of the lower mantle. As a result, different mechanisms, such as anelasticity effect, variation of calcium content, as well as enrichments of iron and silica, have all been suggested as candidates to explain the extreme values of RSP (RSP>2.7 or RSP<1.0) and negative RCS and RρS in the deep mantle. In this study, a comprehensive investigation has been conducted on the variations of these scaling factors with respect to thermal and chemical anomalies for pyrolite lower mantle based on recent thermoelasticity data as well as constraints from previous seismic data. Corresponding to different assumptions used in the derivation of thermoelasticity data from different experimental techniques, low and high bounds for the scaling factors can be obtained. For the lower bounds, RSP of temperature origin increases from 1.7 to 2.0 from the top to the bottom of the lower mantle, however, it is also possible for temperature anomalies to produce higher values at the bottom of the lower mantle; the temperature anomalies can be closely estimated from seismic velocity anomalies alone if the assumptions required by the low bound RSP is satisfied and the density anomaly is small. In comparison, chemical variations (bulk iron and silica contents) result in slightly weaker correlation between shear and compressional waves at lower mantle depths; quantitative determination of chemical anomalies of bulk iron and silica contents, however, requires reliable knowledge of both velocity and density anomalies.
DI31B-1797
Modeling and Interpretation of Localized P-wave and S-wave Reflectivity in D"
Detailed P-wave and S-wave velocity structures for the lowermost mantle beneath the central and eastern Pacific obtained by broadband waveform stacking and modeling procedures indicate distinctive characteristics of low- and high-velocity regions in the deep mantle. The central Pacific region is located within the tomographically-resolved large low shear velocity province (LLSVP) beneath the Pacific that is variously interpreted as a relatively hot 'superplume' or large, dense chemically distinct 'pile'. The localized seismic reflectivity in this region includes rapidly varying depth of a simultaneous P-wave and S-wave velocity increase (D" discontinuity), evidence for a paired S-wave velocity decrease within the D" layer, and a thin ultra-low velocity zone (ULVZ) for both P- and S-waves right above the core-mantle boundary (CMB). The eastern Pacific region is located within the circum-Pacific band of high shear velocity that is variously interpreted as a relatively cold 'slab graveyard' and/or region with post- perovskite being present. The localized reflectivity in this region includes anti-correlated bulk-sound velocity and S-wave velocity contrasts at the D" discontinuity, small P- and S-wave velocity increases and changes in velocity gradient within the D" layer, and an S-wave velocity decrease in the lowermost 50-75 km of the mantle that may be paired with the D" discontinuity, but no resolvable ULVZ. The feature identified as the D" discontinuity has been attributed to the post-perovskite phase transition occurring in both regions, with regional differences in bulk chemistry and thermal structure being invoked to account for the observations. We consider whether predicted effects of Fe and Al variations on the perovskite-to-post-perovskite transition elasticity can be reconciled with the distinct P- and S-wave reflectivity structure in each region, including the D" discontinuity, the paired discontinuities that may involve reverse transformations back to perovskite, and the presence/absence of ULVZ.
DI31B-1798
Chemical Piles and Deep Mantle Plumes
Current tomographic models of the Earth display perturbations to a radial stratified reference model. However, if these are chemically dense structures with low Rayleigh numbers, they can develop enormous relief, perhaps with boundaries closer to vertical than radial. Several new methods have been developed to simulate 3D synthetics for such structures. The method we use approximates 3D effects by adding out-of- plane contributions from virtual receivers at neighboring azimuths with two related to the inner Fresnel zone and two longer-period contributors sampling the outer Fresnel zone. The four responses are scaled by diffraction operators that are defined by the source duration and travel time from the sharp edge structures. Here, we develop a new tool for processing array data based on such a decomposition referred to as a multi- path detector which can be used to distinguish between horizontal structure (in-plane multi-pathing) vs. vertical (out-of-plane multi-pathing) directly from processing array waveforms. We demonstrate the usefulness of this approach by processing samples of both P and S data from the Kaapvaal Array in South Africa. The strength difference between patterns produced by S vs. P-waves for the same event further validates the nature of these chemical boundaries. A detailed SKS wave field is assembled for a strip along the southern boundary by combining multiple events from the Kaapvaal Array. Applying this technique to this composite data set, we locate a prominent ultralow velocity zone at the edge of a 1000 km high jagged wall. We present evidence for a narrow plume with a diameter less than 100 km extending upward another 500 km from the top of the dome, in agreement with recent tomographic images and thermo-chemical convection models.
DI31B-1799
Multi-Scale Imaging of Earth's Deep Interior: New Constraints on Structure and Thermo- chemical Evolution of Earth's Mantle
Since seismic tomography began its revolution of global geophysics some 30 years ago we have made tremendous progress in our ability to image and understand structures and processes in Earth's deep interior. The long wavelength (global) structures discovered in pioneering studies in the early 198ies (e.g., Dziewonski and Woodhouse, 1984) have largely survived the test of time, and later studies have pushed the tomographic models to more-and-more detail. As a result, consensus has emerged on the large scale variations in mantle P and S wavespeed, the presence of compositional heterogeneity, and an intermediate style convection that is neither strictly layered nor unobstructed whole-mantle flow. Tomography constrains smooth variations in material properties. To understand better the radial structure of Earth's interior, along with (mass, heat) fluxes across interfaces and boundary layers, we also need constraints on rapid transitions in material properties. Seminal discoveries have been made through analysis of data associated with reflection, refraction, and phase conversion at interfaces. This trend continues, but the explosion in availability of waveforms from broadband seismograph networks all over the world, combined with advances in inverse scattering theory and high-performance computing, has begun to make global "exploration seismics" of deep Earth interfaces possible. We report new results of large-scale, high resolution imaging of the core-mantle boundary region (D") with inverse scattering of ScS and SKKS wavefields (either separately or jointly) and of the upper mantle transition zone with the wavefield that contains SS precursors due to underside reflection at mantle discontinuities. In the future, inverse scattering with wavefields recorded at global networks may enable the systematic scanning of Earth's mantle and CMB region, which in tandem with parallel advances in mineralogy and phase chemistry research may reveal that the mantle that we often think we know so well still has a few surprises.
DI31B-1800
Seismic Imaging of Multi-Scale Thermal and Chemical Complexities in Earth's Convecting Mantle
For nearly 25 years, seismic velocity heterogeneity in Earth's mantle has been mapped by tomographic
methods, and used to infer global mantle dynamics [e.g., Dziewonski, 1984]. Two nearly antipodal large low
shear velocity provinces (LLSVPs) in the lowermost mantle beneath the Pacific Ocean and Africa are
surrounded by higher than average velocities. The observation that LLSVPs tend to underlie hotspots and
the high velocities tend to underlie subduction has been long suggested to support whole mantle convection.
In more recent years, regional high-resolution waveform studies of structure throughout the mantle depend
on the global models as a long wavelength background framework. This presentation will highlight recent
regional work in the lowermost mantle, as well as in the mantle transition zone, which argue for a mantle with
diverse thermal and chemical complexities. In the deepest mantle, ultra-low velocity zones are found in
isolated patches, and may geographically concentrate near LLSVP margins; especially given the seismic
evidence for a chemically distinct origin to LLSVPs and geodynamic evidence for LLSVP margins being the
hottest mantle rock on Earth. Shear velocity discontinuities in the lowermost 200-300 km of the mantle (the
so-called D" discontinuity) are well-established features, with recent work documenting topography and
suggesting the discontinuity originates from the post-perovskite phase transition. In the mantle transition
zone, a richness in information is emerging about the topography on the 410 and 660 km phase boundaries,
as well as fine scale additional layering. These findings are discussed in reference to the global models,
which define the background thermodynamical setting for the high resolution findings. The high resolution
studies compliment inferences born from tomographically derived long wavelength volumetric heterogeneity,
in that they essentially depict 'roughness' which unveils and discriminates between regional
chemical/mineralogical versus thermal heterogeneity. That is, the two together infer geology and hence and
provide a strong conduit to research in other geodisciplines.
http://garnero.asu.edu/scirev
DI31B-1801
Repeating earthquakes in the southwest Pacific subduction zone and temporal change of the properties of the Earth's core
The seismic data recorded by the Global Seismographic Network (GSN), which Adam Dziewonski championed the establishment, allow researchers in the seismological community to improve the resolution of the Earth's deep interior in great details and to probe the temporal change of the properties of the Earth's deep interior. The permanent anchor of the GSN stations now makes it possible to detect repeating earthquakes with a year-order-of-magnitude recurrence interval, and the global deployment provides critical sampling coverage for the earthquake relative location determination in an unprecedented resolution. Here we present a case study for the repeating earthquakes occurring in the southwest Pacific subduction zone. We compile waveforms for earthquakes with magnitude greater than 4.6 occurring in the southwest Pacific subduction zone recorded by the GSN from January 1990 to December 2006. Waveform cross correlation is applied to screen through the entire waveform database. Among 12,000 earthquakes, we locate nearly 100 event pairs that have cross correlation coefficient greater than 0.8 at all depth ranges, from the interface between the downgoing slab and the overriding plate down to the bottom of the transition zone. Relocation results suggest that some intermediate- and deep- focus earthquake pairs are colocated, suggesting fault reactivation at intermediate and greater depths. The existence of repeating intermediate- and deep-focus earthquakes suggests that thermal (plastic) shear instability is considered the most likely mechanism for intermediate and deep repeating earthquakes. Along with the discovery of repeating earthquakes, we will also report the results of the temporal change of the properties of the Earth's core beneath the northwest and northeast Pacific, the Indian Ocean, Asia, and Antarctica.
DI31B-1802
The Effects of 3-D Q Structure on Surface Wave Phase Delay
3-D velocity tomography has rapidly progressed in the past decades while studies of 3-D anelasticity (Q) structure are somewhat lagging behind. It is known that using only S-wave velocity structure in the upper mantle, it's not possible to distinguish between thermal and chemical origins of mantle heterogeneities. The earth's anelasticity (Q) structure has strong sensitivity to temperature and weak sensitivity to compositional variations, and therefore can be applied as additional constraints to map thermal and compositional variations in the mantle. In present-day global anelasticity (Q) tomographic practices, the effects of 3-D anelasticity effects on seismic travel time (phase delay) -- 3-D anelastic dispersion -- have been ignored. In this work, we quantify the effects of 3-D anelasticity (Q) on surface wave phase delays by simulating wave propagation in 3-D anelastic earth models using the spectral element method (SEM). We compare the measured phase delays caused by 3-D anelasticity (Q) structure to those caused by 3-D velocity structure. The 3-D Q models and 3-D velocity models are both constructed from a 3-D temperature model, and the strength (rms) of the resulting 3-D Q models and 3-D velocity models are comparable to currently available seismic tomographic models. Our measurements based upon SEM seismograms show that (1) roughly 30% of the observed phase delays (travel times) are due to 3-D anelasticity (Q) structure; this implies that neglecting 3-D anelastic dispersion effects will lead to overestimated velocity perturbations in seismic tomography. (2) the significance of 3-D anelastic dispersion is frequency dependent, it's stronger in long period surface waves where the sensitivity in the low-Q asthenosphere is maximized; and (3) the 3-D anelastic dispersion effects are stronger in regions of "slow" anomalies where the temperature is higher than the background model.