Seismology [S]

S13B MCC:level 2 Monday 1340h

Beneath the Continents: Crust and Mantle II Posters

Presiding:J Xie, Lamont Doherty Earth Observatory; F H Wagner, University of Arizona

S13B-1041 1340h

Cenozoic Crustal Extension and Regional Fault Interaction in Southern Arizona Revealed Through Seismic Reflection Investigations

* Wagner, F H (fwagner@geo.arizona.edu) , University of Arizona, Dept. of Geosciences 1040 E 4th St, Tucson, AZ 85721 United States
Johnson, R A (johnson@geo.arizona.edu) , University of Arizona, Dept. of Geosciences 1040 E 4th St, Tucson, AZ 85721 United States

Recent reprocessing and interpretation of industry seismic reflection data collected in the 1970's around southern Arizona have illuminated subsurface features related to Cenozoic crustal extension. Within the northern Tucson Basin, the Catalina detachment fault dips at 23-35 degrees to the southwest off of the western flank of the Catalina-Rincon Metamorphic Core Complex. Imaged on the detachment surface are broad corrugations on the scale of tens of kilometers, parallel to the extension direction, evident down to several seconds two-way travel time in the data. In the southern portion of the basin, the Santa Rita fault extends northwest from the western flank of the Santa Rita Mountains and dips 15-20 degrees and is cut by the Catalina detachment beneath the central Tucson Basin. This cross-cutting relationship is problematical in that the Santa Rita fault shows late Quaternary movement and also is oriented nearly perpendicular to extension directions in the region. We propose that the Santa Rita fault represents contemporaneous upper-plate deformation in the detachment system, and that recent movement inferred on the fault is simply a reactivation of this older fault surface. West of the Tucson basin, separated from it by the Sierrita Mountains, is Altar Valley. Seismic data reveal that Altar Valley is a half graben with a large-displacement normal fault bounding the western margin of the basin. This basin-bounding fault dips steeply toward the east from the eastern flank of the Baboquivari Mountains and appears to dip less steeply with increasing depth. The seismic data image multiple fault splays that merge in basement crystalline rocks into a main, shallowly eastward-dipping surface at around four seconds two-way travel time (approximately 10 km). This surface appears to anastomose with other surfaces imaged in the basement rocks and suggests that regional deformation has been distributed through discrete deformation zones. In a more regional context, the subsurface relationships between the Catalina detachment, the Santa Rita fault and the Baboquivari fault are intriguing. One aspect of interest is the unusual orientation of the Santa Rita fault, which is oblique to the direction of major extension in the area. The Santa Rita fault is located between areas of differing amounts of crustal extension, with a zone to the north of extreme crustal extension on the order of 100% or more across the Catalina / Rincon metamorphic core complex, and a zone to the immediate south that has experienced moderate crustal extension of perhaps 30%. Consequently, the nature and orientation of the Santa Rita fault could be the result of mountain-range-scale block rotations transferring strain between extensional domains. Also of interest is the relationship between the Catalina detachment system and the Baboquivari fault. The projections of these two fault systems suggest the two may interact at mid-crustal levels, perhaps joining into a common regional detachment. This places the Sierrita Mountains in the hanging-wall block between these two fault systems. Indeed, deformation within the Sierrita's seems to support contemporaneous movement of the Baboquivari fault and the Catalina detachment.

http://www.geo.arizona.edu/~fwagner/

S13B-1042 1340h

Coupled Basin Evolution and Core-Complex Exhumation in Southeastern Arizona: Another Look Into Subsurface Relations From Seismic Reflection Data

* Johnson, R A (johnson@geo.arizona.edu) , University of Arizona, Department of Geosciences, 1040 E. 4th Street, Tucson, AZ 85721 United States
Wagner, F H (fwagner@geo.arizona.edu) , University of Arizona, Department of Geosciences, 1040 E. 4th Street, Tucson, AZ 85721 United States

Records of lithospheric extension and mountain-range uplift are most continuously contained within syntectonic sedimentary rocks in basins adjacent to large structural culminations. Although subject to local conditions and other limitations, basin-fill sedimentary assemblages provide essential keys to understanding the sequence and timing of tectonic activity and crustal deformations in the southern Basin and Range Province where metamorphic core complexes document extreme crustal extension. In southeastern Arizona, major core complexes form mountain ranges with the highest elevations in the region and, paradoxically, supposedly much less extended terranes lie at lower elevations. These large structural and topographic culminations are flanked by sedimentary basins that have trapped the products of core-complex formation. Adjacent to the Catalina-Rincon metamorphic core complex, stratigraphic-sequence geometries evident in seismic reflection data within the Tucson Basin suggest a two-phase basin-evolution model. In its earliest stage, Phase I was characterized by voluminous volcanism that coincided with the early stages of crustal extension and tectonically induced collapse of post-Laramide crustal blocks. Extensive faults formed relatively small-scale proto-basins, which coalesced in later stages along larger basin-bounding extensional fault systems. Evidence of synextensional sedimentation within the enlarging basin is provided by sediment growth packages, derived from adjacent footwall material, which fan into brittle hanging-wall faults. During this phase, volcanism remained active, and growth packages contain interbedded sedimentary rocks and volcanic products. Phase II of basin evolution consisted of an overall broader deepening of the central basin, coupled with uplift and initial exposure of the Catalina-Rincon metamorphic core as supported by the introduction of mylonitic clasts present in the basin fill that were not present in the earlier phases of sedimentation. Well-log constraints from the central basin suggest that a significant depositional hiatus exists between the two phases of deformation and sedimentation, consistent with surface geological evidence for late-stage exhumation of the Catalina-Rincon metamorphic core complex. However, seismic sequence analysis suggests that mountain uplift was coupled with basin subsidence over an extended period of time. Moreover, the geometry of upper basin-fill units suggests lateral mass transfer of mid-crustal rocks to the adjacent core complex as exhumation of the crystalline core progressed.

S13B-1043 1340h

DEEP BOREHOLE RECEIVER FUNCTIONS IN LONG VALLEY CALDERA, CALIFORNIA

* Chavarria, J (jac4@duke.edu) , Duke University - Earth and Ocean Sciences, 103 Old Chemistry B90229, Durham, NC 27708 United States
Julia, J (jordi@duke.edu) , Duke University - Earth and Ocean Sciences, 103 Old Chemistry B90229, Durham, NC 27708 United States
Malin, P (malin@duke.edu) , Duke University - Earth and Ocean Sciences, 103 Old Chemistry B90229, Durham, NC 27708 United States

Receiver functions are now routinely calculated from teleseismic P-waveform records to investigate the subsurface geology. For these calculations broad-band seismometers are generally used since only the low end of the frequency spectrum (f $<$ 1.5 Hz) can be interpreted. With a number of ongoing efforts of deep drilling and monitoring of earthquakes, a number of short period seismometers has been installed in deep wells in various geologic settings. In this work we explore the use of these short period instruments in deep borehole environments for the analysis of high signal to noise ratio receiver functions. Receiver functions are obtained by deconvolving the vertical component from the corresponding radial component of the waveforms, which equalizes for the source effects in the original data. During the deconvolution process, the instrument response is also equalized regardless of the dynamic range of the recording sensors. It has been noted that the signal to noise ratio increases with depth and here we show that single stations, located in deep boreholes, have strong signals that allow for stable deconvolutions. We tested the receiver function technique with short-period instruments (fc = 4.5 Hz) located in two wells in the Long Valley Caldera in California. The first instrument is located in a ~200m deep geothermal well whereas the other one is in the 2.5km deep Long Valley Exploratory Well. A number teleseismic events (30 $<$ D $<$ 90) have been recorded over a few months of acquisition and with these data, stable receiver functions were computed. The deconvolved traces coherently show strong secondary waves that we have interpreted as a Ps conversion and a PpPhs multiple reverberberation on top of a feature that we interpret as the top of a magma body located about 10-12 km depth.

S13B-1044 1340h

Transition zone structure of California from Ps converted waves

* Cupillard, P (paulcup@ipgp.jussieu.fr) , IPG Paris, Tour 24-14 4, place Jussieu, Paris, 75252 France
Ritsema, J (jeroen@ipgp.jussieu.fr) , IPG Paris, Tour 24-14 4, place Jussieu, Paris, 75252 France

We present new estimates of the structure of seismic discontinuities in the upper mantle of California and the northern Gulf of California from the analysis of stacks of 10,000 TriNet, Berkeley Digital Seismic Network, and NARS-Baja waveforms. Throughout the region, Ps conversions from the 410-km and 660-km conversions arrive 2-3 s later than predicted by IASP91, due to low seismic velocities in the uppermost mantle of western North America. In addition, the differential traveltime of P660s and P410s is up to 1 s shorter than predicted by IASP91, which is among the lowest values observed worldwide, in accord with earlier observations of single-station stacks. This suggests that the transition zone beneath California is as thin as beneath various hotspot regions, and that the temperature of the mantle beneath California is elevated to at least 400 km depth. We will evaluate whether our observations can be reconciled with models of the upper mantle that invoke variations of shear and compressional velocity due to temperature alone.

S13B-1045 1340h

Preliminary Results of Seismic Refraction/Reflection Experiment in Northwestern Nevada and Northeastern California

* Colgan, J P (jcolgan@geo.stanford.edu) , Department of Geological and Environmental Sciences, Stanford University, Stanford, CA 94305-2115 United States
Lerch, D L (lerch@geo.stanford.edu) , Department of Geological and Environmental Sciences, Stanford University, Stanford, CA 94305-2115 United States
Gashawbeza, E M (ewenet1@geo.stanford.edu) , Department of Geophysics, Stanford University, Stanford, CA 94305-2215 United States
Wilson, C K (wilsonck@geo.stanford.edu) , Department of Geophysics, Stanford University, Stanford, CA 94305-2215 United States
Klemperer, S L (sklemp@stanford.edu) , Department of Geological and Environmental Sciences, Stanford University, Stanford, CA 94305-2115 United States
Klemperer, S L (sklemp@stanford.edu) , Department of Geophysics, Stanford University, Stanford, CA 94305-2215 United States
Miller, E L (miller@geo.stanford.edu) , Department of Geological and Environmental Sciences, Stanford University, Stanford, CA 94305-2115 United States

In September 2004, Stanford University is conducting a 260-km seismic refraction/reflection experiment across the northwestern margin of the Basin and Range Province between Winnemucca, Nevada and Alturas, California. This area was the locus of intense mid-Miocene (c. 17-15 Ma) volcanism and subsequent high-angle extensional faulting, and is often thought to be the breakout region of the Yellowstone hotspot. Today the region is characterized by high heat flow and is presumed to have relatively thin crust (c. 25 km), but little is known about the overall structure of the crust and how it relates to magmatism and extensional faulting. Our seismic experiment is designed to collect information on the crustal thickness, velocity structure, and crustal-scale reflectivity of this area, and consists of four parts: 1) A 260-km crustal refraction profile, with five in-line shots (1.25 to 2 tons each) and one 1.25 ton fan shot to the south, with over 1000 receivers spaced 100 to 300 m apart, acquired with ACS/PRF and NSF/EarthScope funding. The goals of the refraction profile are to determine crustal thickness and overall velocity and reflectivity structure. 2) While geophones are deployed along the refraction profile, we will collect reflection data in P, SV and SH modes using the tri-axial "T-Rex" vibrator truck operated by the Network for Earthquake Engineering Seismology (NEES) and the University of Texas at Austin. This experiment will asses the capability of this instrument to collect useful crustal-scale reflection data in conjunction with PASSCAL and EarthScope recorders, and if successful will help constrain the types of rocks and structures present beneath the flat-lying Miocene volcanic rocks that cover much of northwestern Nevada and largely obscure older structures. 3) 48 short-period 3-component receivers will be embedded in the main refraction line, supplemented by two 16-receiver offline deployments perpendicular to the main line. This experiment is designed to measure crustal S-wave splitting (PmS phase) from the active source experiment as an indicator of crustal anisotropy resulting from lower crustal flow. We may also detect differences in direct-S travel times using the SV and SH modes of the T-Rex vibrator from the same vibration location. 4) A 16-km high-resolution (40 m receiver spacing) reflection profile across Surprise Valley, CA. This experiment will test the capabilities of the T-Rex vibrator for shallow, high-resolution P and S-wave imaging in a low-cost experiment, and should yield geologically useful information on the depth of basin fill in Surprise Valley above the recently active Surprise Valley fault. We present preliminary data from our experiment and discuss their implications both for regional tectonics, and as a potential model for future EarthScope/NEES collaborations.

S13B-1046 1340h

Origins of the Yellowstone Hotspot and Implications to Western U.S. Tectonics

* Crosswhite, J A (jason@newberry.uoregon.edu) , University of Oregon, Department of Geosciences 1272 University of Oregon, Eugene, OR 97403-1272 United States
Humphreys, E D (gene@newberry.uoregon.edu) , University of Oregon, Department of Geosciences 1272 University of Oregon, Eugene, OR 97403-1272 United States

We assembled teleseismic data for the Yellowstone hotspot representing a $>$1,000 km aperture array. These data combined with finite-frequency ("banana-doughnut") tomography allowed resolution to the transition zone. The PASSCAL Yellowstone data was augmented with temporary deployments and regional arrays in Utah, Colorado, Montana, Idaho and Utah. In net, these data represent more than 200 seismometers running approximately one year. The finite frequency Fr\'echet kernels better predict sensitivity to velocity and include diffraction phenomena. Due to their large sampling volume, there is better resolution at depth. The combination of an extended dataset and newer techniques allowed imaging to the transition zone with good resolution. Yellowstone is an atypical hotspot in many ways. Theories of its origin include plume and non-plume hypotheses. We have tested these hypotheses for the Yellowstone hotspot in the context of western U.S. tectonic history and dynamics.

S13B-1047 1340h

Scattering and Attenuation of Seismic Waves in the Northeastern United States

* Cicerone, R D (rcicerone@bridgew.edu) , Bridgewater State College, Department of Earth Sciences and Geography, Bridgewater, MA 02325 United States
Doll, C G (doll@ll.mit.edu) , Earth Resources Laboratory, MIT, 42 Carleton Street, Cambridge, MA 02142 United States
Doll, C G (doll@ll.mit.edu) , now at Lincoln Laboratory, MIT, 244 Wood Street, Lexington, MA 02420 United States
Toks\"oz, M N (nafi@mit.edu) , Earth Resources Laboratory, MIT, 42 Carleton Street, Cambridge, MA 02142 United States

The energy-flux model of seismic coda, developed by Frankel and Wennerberg (1987), is used to derive path-averaged estimates of scattering (Q$_{S}$$^{-1}) and intrinsic attenuation (Q_{I}$$^{-1}$) for the northeastern United States. The model predicts the amplitude of the coda wave vs. time as a function of frequency, Q$_{S}$$^{-1}, and Q_{I}$$^{-1}$. A non-linear inversion scheme is developed that allows for the estimation of Q$_{S}$$^{-1} and Q_{I}$$^{-1}$ as a function of frequency by fitting the model to a narrow-bandpass filtered envelope of the seismic coda for each seismogram at discrete frequency points. The inversion is performed on seismograms from earthquakes recorded by the MIT New England Seismic Network (NESN) over a 15-year period between 1981 and 1995. Preliminary results indicate that scattering is the dominant mechanism of energy dissipation, and that the effects of intrinsic attenuation are secondary. The scattering is strongest at shorter propagation distances and decreases substantially as the propagation distance increases. Conversely, intrinsic attenuation is negligible at shorter propagation distances and increases as the propagation distance increases. These results are interpreted as indicative of a strong scattering region at shallow depth, with the scattering decreasing with increasing depth, and with a subsequent increasing of intrinsic attenuation at greater depth. We propose a second analysis to invert the path-averaged estimates of Q$_{S}$$^{-1} and Q_{I}$$^{-1}$ using a constrained linear method with regularization to obtain a one-dimensional model of Q$_{S}$$^{-1} and Q_{I}$$^{-1}$ vs. depth in the crust. Possible mechanisms for the scattering include the presence of a weathering layer near the surface, the presence of fractures in the shallow crust, and topography.

S13B-1048 1340h

Seismic Characteristics of the Pan-African Granulite Terranes of Southern Gondwanaland

Gaur, V K (gaur@cmmacs.ernet.in) , CMMACS, NAL Belur Campus, Bangalore, 560 037 India
Rai, S S (ssrai_ngri@rediffmail.com) , NGRI, Uppal Road, Hyderabad, 500 007 India

The Pan-African orogeny was a major tectono-thermal event which occurred at the Proterozoic-Palaeozoic boundary (~600 Ma) and is now marked by zones of high-grade metamorphism distributed from the Arabian-Nubian shield through Tanzania, Mozambique, Southern India and Sri Lanka to the eastern Antarctic. Teleseismic earthquake recordings from stations in these areas have been analyzed to compare the seismic characteristics of the crust and upper mantle for these regions. Inversion of P-wave receiver functions and surface-wave phase velocity data reveals a similar crustal structure in each of these areas of the now disrupted Pan-African belt. The Pan-African crust has a more complicated structure than does the adjacent Eastern Dharwar Craton of the south Indian shield. Most of the stations located on the Pan-African granulite terranes show a two-layer crust with a clear mid-crustal discontinuity and a lower velocity zone at upper to mid-crustal depths. The analysis of the splitting of Ps Moho converted phases indicates similar anisotropic fabric for the crust when considering their reconstructed locations. SKS splitting measurements show a similar anisotropic fabric for the upper mantle beneath the stations whose data we have analyzed. Both the crust and upper mantle anisotropy for the Pan-African granulite terranes is significantly different from that observed on the Eastern Dharwar Craton of the south Indian shield.

S13B-1049 1340h

Searching for the difference in attenuation between the fast and slow shear waves

Gao, S S (gao@ksu.edu) , Department of Geology, Kansas State University, Manhattan, KS 66506 United States
* Liu, K H (liu@ksu.edu) , Department of Geology, Kansas State University, Manhattan, KS 66506 United States
Zhang, Z (zhuzhang@ksu.edu) , Department of Geology, Kansas State University, Manhattan, KS 66506 United States
Gao, Y (gaoyuan@seis.ac.cn) , Institute of Earthquake Science, China Earthquake Administration, Beijing, 100036 China

Measurements of shear-wave splitting parameters (fast polarization direction PHI and splitting time DT) have become a powerful and routine technique to study the deformation field of the earth's interior. While it is evident that velocity anisotropy can lead to significant arrival-time difference between the fast and slow shear waves, the question of whether the two waves have observable difference in attenuation has rarely been addressed. In principle, such a difference in attenuation is likely to be observed, especially from local shear-waves recorded in areas with fluid-filled cracks; relative to the fast wave, the slow wave may experience higher attenuation. Successful detection and characterization of the difference in attenuation may provide additional constraints on the properties of the anisotropic media. We have developed and tested a grid-searching procedure to search for the optimal PHI, DT, and the relative attenuation factor (as quantified by t* which is travel-time over Q) between the fast and slow shear waves that give rise to the maximum cross correlation coefficient between the corrected fast and slow waves. The procedure can be used for both SKS/SKKS phases and shear-waves originated from local earthquakes. Details about the procedure and its applications on seismic data recorded at a number of locations will be presented.

S13B-1050 1340h

Construction of a 3D seismic velocity model for the Barents Sea region using sediment vs. crystalline crust thickness relationships

* Ritzmann, O (oliver.ritzmann@geo.uio.no) , Department of Geoscience - University of Oslo, P.O. Box 1047 Blindern , Oslo, 0482 Norway
Faleide, J (j.i.faleide@geo.uio.no) , Department of Geoscience - University of Oslo, P.O. Box 1047 Blindern , Oslo, 0482 Norway
Bungum, H (hilmar.bungum@norsar.no) , Norsar, P.O. Box 53, Kjeller, 2027 Norway
Maercklin, N (nils.maercklin@norsar.no) , Norsar, P.O. Box 53, Kjeller, 2027 Norway
Schweitzer, J (johannes@norsar.no) , Norsar, P.O. Box 53, Kjeller, 2027 Norway
Mooney, W D (mooney@usgs.gov) , United States Geological Survey, 345 Middlefield Road, Menlo Park, CA 94025 United States
Detweiler, S T (shane@usgs.gov) , United States Geological Survey, 345 Middlefield Road, Menlo Park, CA 94025 United States
Leith, W S (wleith@usgs.gov) , United States Geological Survey, 345 Middlefield Road, Menlo Park, CA 94025 United States

We present a 3D seismic velocity model for the extended Barents Sea region, including Svalbard, Novaya Zemlya, the Kara Sea and the Kola-Karelia Regions. The purpose of developing a higher-resolution velocity model is to improve generally the seismic event localization in the target region. The model should improve the future monitoring facilities and the accompanied travel-time modeling. Initial testing of the model will base on the modeling of a series of seismic ground-truth events recorded by the surrounding stations. The model has a spatial resolution of 50x50 km and includes 1490 nodes. Each node is filled with a 5-layer crustal model (plus water/ice- and additional mantle layers): Nodes within the oceanic and continental domains bear two sedimentary layers (low/high vp) and three "crystalline" crustal layers (low/intermediate/high vp). Basis of this model is a recent compilation of seismic velocities taken from published wide-angle profiles, unpublished ESP profiles and additional gravity modeling along deep MSC-profiles. Over 700 1D velocity profiles are collected. In order to interpolate the velocity/depth-information from the randomly distributed 1D profiles on the equal-spaced grid, the following technique was applied: Analyzing the database, we found a strong linear trend between the total thickness of the sediment layers and the remaining crystalline crust within pre-defined continental provinces (e.g. distinct sedimentary basins, plateaus, basement highs, etc.). Area-wide depth-to-basement information, based on the integrated analysis of seismic, gravity and magnetic data is used to calculate the crystalline and total crustal thicknesses as functions of sediment thickness. The mean seismic velocities and thickness-rates for each of the 5 crustal layers are calculated from the compiled database. Analysis of the regressions show that about 75-90% of the data input is fitted by the calculated functions with a maximum of 20% deviation relative to its total thickness. The compiled database provides further excellent statistical background for composition of crystalline crustal rocks in the target region. The overall distribution of seismic velocities within crystalline crust shows a clear bimodal structure with velocity peaks at 6.4 and 6.8 km/s. First modeling tests along four selected transects were carried out to evaluate the constructed 3D seismic model. According to the tests travel-time deviations can exceed 2 s at distances of 300-800 km (by comparison to a standard 1D model).

S13B-1051 1340h

A Shear Wave Velocity Structure in the Eastern Mediterranean From Surface Wave Dispersion

* Di Luccio, F (diluccio@ingv.it) , Istituto Nazionale di Geofisica e Vulcanologia, Via di Vigna Murata 605, Rome, 00143 Italy
Pino, A N (pino@ingv.it) , Istituto Nazionale di Geofisica e Vulcanologia, Via di Vigna Murata 605, Rome, 00143 Italy

Regional moderate earthquakes are analyzed to obtain a shear wave velocity model in the complex tectonic setting of the Eastern Mediterranean. The analysis of surface waves from broadband waveforms is based upon the fitting of the fundamental mode of Love and Rayleigh waves. Since the surface waves from different periods travel at different velocities, hence their displacements vary with depth and their analysis can give a constrain to the earth structure. Besides, analyzing dispersive signals is faster than modeling the whole waveforms, which requires the computation of the Green's functions. The behaviour of the computed group velocities as a function of period gives information on the crustal (short periods) and upper mantle structure (long periods). The study area was divided into separate regions, each characterized by specific paths, both pure and mixed. For each source to receiver path the group velocities are obtained using a Multiple Filter Analysis code (see Herrmann and Ammon, 2002), where the Rayleigh and Love waves are selected from the vertical and tangential displacement, respectively. Although the group velocities are quite low in the study area in the 10-100s period range, important differences are observed even in adjacent regions, at longer as well as shorter periods.

S13B-1052 1340h

Upper mantle structure beneath Ethiopia from a joint inversion of receiver functions and surface wave dispersion

* Dugda, M T (mtd172@psu.edu) , Pennsylvania State University, Department of Geosciences, University Park, PA 16802 United States
Nyblade, A A (andy@geosc.psu.edu) , Pennsylvania State University, Department of Geosciences, University Park, PA 16802 United States
Julia, J (jordi@duke.edu) , Duke University, Division of Earth and Oceanic Sciences, Durham, NC 27708 United States
Ammon, C (cammon@geosc.psu.edu) , Pennsylvania State University, Department of Geosciences, University Park, PA 16802 United States

We investigate upper mantle structure beneath the Afar Depression, the Main Ethiopian Rift (MER) and Ethiopian Plateau by jointly inverting receiver functions and Rayleigh wave group velocities. The main aim of this study is to characterize the nature of the thermal anomaly beneath the region and then to use this information to evaluate models for the origin of the hotspot activity. The data for this study come from the Ethiopian Broadband Seismic Experiment in which 27 stations were deployed throughout Ethiopia for two years (2000-2002). Preliminary results suggest contrasting upper mantle structure between the plateau stations and the stations in the MER and Afar Depression. For the plateau stations to the east and west of the MER, the upper mantle has a high velocity lid, with shear wave velocities of around 4.5 km/s that extends to between 70 and 100 km depth. Below the lid is a pronounced low velocity zone with shear wave velocities of about 4.1-4.2 km/s. The lowest velocities are found no deeper than 150 km. In contrast, there is no seismic lid beneath the stations in the MER and Afar Depression. The shear velocities are anomalously low at all mantle depths. These preliminary findings are consistent with a plume model for the hotspot activity.

S13B-1053 1340h

Crustal structure of the Arabian Plate: New constraints from the analysis of teleseismic receiver functions

* Al-Damegh, K (kdamegh@yahoo.com) , King Abdul Aziz City for Science and Technology, Department of Astronomy and Geophysics, Riyadh, 11442 Saudi Arabia
Sandvol, E (sandvole@missouri.edu) , University of Missouri, 101 Geology Building University of Missouri, Columbia, MO 65211 United States
Barazangi, M (mb44@cornell.edu) , Institute for the Study of the Continents, Snee Hall Cornell University, Ithaca, NY 14853 United States

Receiver function for teleseismic earthquakes recorded at 23 broadband and mid-band stations in Saudi Arabia and Jordan were processed to map variations in crustal thickness. The receiver functions were stacked using a slant-staking technique to estimate the Moho depth and the Vp/Vs for each station. The errors in the slant stacking were estimated using a bootstrap error technique. The grid search technique was also used to estimate the crustal structure for seven stations. A jackknife error estimation method was used to estimate the errors in the grid search results for three stations. The average crustal thickness of the Arabian shield is 39 km. The crust thins to about 23 km along the Red Sea and to about 29 km along the Gulf of Aqaba. In the northern part of the platform, the crust varies from 33 - 37 km. This is less than the average crust in the shield. However, the crust is thicker (41 - 49 km) in the southern part of the platform. Our Moho map shows a dramatic change in crustal thickness between the topographic escarpment of the Arabian shield and the shorelines of the Red Sea. The Moho depth in the Proterozoic Arabian shield is relatively stable, and ranges between 35 - 45 km with an average of 39 km. We compared our results in the shield to nine Proterozoic and Archean shields regions that include reasonably well-determined Moho depths. Overall, we do not observe significant difference between Proterozoic and Archean crustal thickness. We observed a dramatic change in crustal thickness along the Red Sea margin that occurred over a relatively short distance. The transition from oceanic to continental crust along the Red Sea margin occurs in < 250 km, while the transition in a typical west Atlantic-type margin occurs in over 450 km. This important result highlights the abruptness of the breakup of Arabia. We argue that a preexisting zone of weakness coupled with anomalously hot upper mantle could have initiated and expedited the breakup.

S13B-1054 1340h

An Investigation of Crust and Upper Mantle Structure in Western Argentina Utilizing Local Event Receiver Functions

* Calkins, J A (jcalkins@geo.arizona.edu) , University of Arizona, 1040 E. Fourth Street, Tucson, AZ 85719 United States
Zandt, G (zandt@geo.arizona.edu) , University of Arizona, 1040 E. Fourth Street, Tucson, AZ 85719 United States
Gilbert, H (hgilbert@geo.arizona.edu) , University of Arizona, 1040 E. Fourth Street, Tucson, AZ 85719 United States
Beck, S (beck@geo.arizona.edu) , University of Arizona, 1040 E. Fourth Street, Tucson, AZ 85719 United States

Images of the crust-mantle boundary and crustal structure obtained using the traditional analysis of teleseismic receiver functions (RFs) exhibit an unusually weak P-S conversion from the Moho in Western Argentina, where the subducting Nazca plate temporarily flattens out beneath the overriding South American plate. In order to better estimate depth to the Moho and search for mid-crustal impedance contrasts, we calculate and stack receiver functions using approximately 45 local earthquakes occurring in the downgoing slab between December of 2000 and February of 2001. The events occurred over a depth range of 76 to 165 km and were all within 128 km horizontal distance of the recording station and thus traveled with ray parameters less than .09 s/km. Radial receiver functions are calculated at two temporary broadband seismic stations located between San Juan and Mendoza, in the region where the Precordillera transitions eastward to the Sierras Pampeanas. Plots of stacked RFs as a function of ray parameter show a strong signal from the Moho at 7 seconds corresponding to a depth near 50 km, as well as conversions from interfaces within the crust at depths of $\sim$ 20 and 35 km. It should be noted that the narrow time interval between the P and S arrivals, due to the close proximity of events to the stations, precludes the analysis of reverberations within the crust to better constrain crustal Vp/Vs estimates and to refine the depth to interfaces. The observed Moho depth is in good agreement with estimates made using Pn apparent phase velocities along a transect through tectonically similar terrain 200 km to the north. In both cases, areas of relatively low topography are underlain by anomalously thick crust. The discrepancy in the clarity of the Moho Ps between RFs obtained using teleseismic versus local events currently remains unexplained but is an area of ongoing research.

S13B-1055 1340h

New Interpretation of Deep Seismic Refraction data from the Tocantins Province, Brazil

* Perosi, F (fabio@iag.usp.br) , US Geological Survey, 345 Middlefield Rd. MS 977, Menlo Park, CA 94025 United States
* Perosi, F (fabio@iag.usp.br) , University of Sao Paulo, Rua do matao, Sao Paulo, 05508 Brazil
Mooney, W D (mooney@usgs.gov) , US Geological Survey, 345 Middlefield Rd. MS 977, Menlo Park, CA 94025 United States
Berrocal, J (berrocal@iag.usp.br) , University of Sao Paulo, Rua do matao, Sao Paulo, 05508 Brazil

We interpret seismic refraction data in the central sector of the Tocantins Province, central Brazil, producing a seismic crustal model with a well-defined upper, middle, and lower crust with smooth velocity gradients in each layer. The depths to Moho vary from 32 to 43 km, and we find mean crustal P-wave velocity to be 6.3 - 6.4 km/s. The behavior of the lower crustal layer allows an improved understanding of regional gravimetric features of central and northern sectors of the Tocantins Province, and suggests that there was subduction of the Amazon plate beneath central Brazil. In the southeastern sector, the refraction experiment resulted in a detection of a thinner crust (38 km) below the Brasilia fold belt, and a thicker crust (41 km) below the Parana basin and Sao Francisco craton (42 km). The upper crust beneath the Parana basin is around 20 km thick, and thins to less than 10 km below the craton. Gravimetric measurements in the central sector of the Tocantins Province delineate a high and a low anomaly separated by a steep gradient with a NE direction. The axis of the gradient seems to bend still further to NE in the northern sector of that province, whereas the gravimetric high continuous northwards, defining a separation between them. This suggests those features belong to different tectonic processes. The gravimetric model, which incorporates seismically resolved structure beneath the Tocantins Province, matches the observed gravimetric data quite well. Although tectonic movements have been monitored with high precision GPS for only a short time interval (1999- 2001), the results suggest observable deformations. The main seismicity of central Brazil, the Goias-Tocantins seismic belt, seems to be spatially associated with a large gravimetric high and with observed tectonic deformation. These results bring new insights into the geological history of the central and southeastern sectors of the Tocantins Province.

S13B-1056 1340h

Upper Mantle P and S-wave Velocity Structure Beneath Eastern Anatolian Plateau

* Sandvol, E (sandvole@missouri.edu) , Department of Geological Sciences, 101 Geology Building, Columbia, MO 65211 United States
Zor, E (zore@missouri.edu) , Department of Geological Sciences, 101 Geology Building, Columbia, MO 65211 United States

The Eastern Turkey Seismic Experiment (ETSE) was designed to image the crustal and upper mantle velocity structure beneath the northernmost Arabian plate and beneath the Anatolian Plateau, constraining geodynamic models for young continent-continent collision. ETSE was composed of a 29 station broadband PASSCAL array which was deployed from October 1999 until August 2001. Tomographic mapping of the Rayleigh wave phase velocities suggests that the uppermost mantle ultra-low velocity zone is underlain by relatively normal to slightly low velocity mantle. This finding is roughly consistent with our three dimensional teleseismic P wave tomography results. The preliminary P-wave tomography, however, does not show as clear of contrast in seismic velocities across the Bitlis suture. We also observe interesting differences between the S-wave and P-wave tomography results in the center portion of the Anatolian plateau where S-wave velocities are the lowest. This might be an indication of the presence of water in the uppermost mantle in the easternmost Anatolian plateau. We have also investigated the depth extent of the azimuthal anisotropy beneath the eastern Anatolian plateau using Rayleigh wave phase velocities. It appears as if there are two anisotropic layers within the upper mantle. The top layers (periods 30 and 50 seconds) are consistent with the values obtained from Pn anisotropy and shear wave splitting. The bottom layers (periods 100 - 150 seconds) are not consistent with fast directions obtained from shear wave splitting and Pn azimuthal anisotropy. The two layers are separated by a region that has relatively little anisotropy.

S13B-1057 1340h

Crustal Attenuation Within The Anatolian Plateau And Surrounding Regions

* Zor, E (zore@missouri.edu) , Department of Geological Sciences, University of Missori-Columbia, Columbia, MO 65211 United States
Sandvol, E A (sandvole@missouri.edu) , Department of Geological Sciences, University of Missori-Columbia, Columbia, MO 65211 United States
Xie, J (xie@ldeo.columbia.edu) , Lamont Doherty Earth Observatory, Columbia University, Palisades, NY 10964 United States
Mitchell, B (mitchbj@eas.slu.edu) , Department of Earth and Atmospheric Sciences, St. Louis University, St. Louis, MO 63103 United States
Turkelli, N (turkelli@boun.edu.tr) , Department of Geophysics, Kandilli Observatory and Earthquake Research Institute, Bogazici University, Istanbul, 81220 Turkey

In order to map variations in crustal and uppermost mantle attenuation for a given region, we have collected waveform data from a wide variety of sources like national networks, GSN stations, various permanent and temporary sites in the Middle East (Jordan, Saudi Arabia), Caspian Sea region. A large number of digital seismograms (approximately 8000) yields a dense path coverage for all four regional seismic phases throughout most of the northern portion of the Arabian plate as well as the Anatolian Plateau. This dense coverage should dramatically improve the resolution of our Q model. Moreover, it allows the bias in Q measurements to be minimized by using methods that are more reliable, but require restrictive path geometries. The most efficient method for developing a lithospheric attenuation model is to use standard two-station method using computed Fourier spectra for the corresponding regional seismic phases: Pn, Pg, Sn, and Lg for many events and paths. This method virtually eliminates source, site and path effects which is even more critical for paths through a region such as the Middle East, where both crustal and uppermost mantle attenuation are very high. We began working on Lg with this method which allow us to obtain a model of Qo and n (Lg Q at 1 Hz and its power-law frequency dependence, respectively). Subsequently, the tomography will yield high-resolution maps of laterally varying Qo. This study will also yield a regionalized Q models for Pg, Pn and Sn. Our strategy is to begin in regions where we have very dense two station ray paths; therefore, we have chosen to begin our focus on the eastern Anatolian plateau and northern Arabian plate where we have data from Kandilli Observatory and Earthquake Research Institute, Jordanian Seismic Observatory and from the temporary PASSCAL network in eastern Turkey (ETSE). This region has been shown to be a region of anomalously high Lg and Sn attenuation and blockage. Our Q-models will be used to quantify the level of attenuation within the Anatolian plateau.

S13B-1058 1340h

Crustal Structure Study In Turkey With Controlled Seismic Sources

* Kuleli, S (kuleli@erl.mit.edu) , Massachusetts Institute of Technology, 77 Massachusetts Ave. , Cambridge, MA 02139
* Kuleli, S (kuleli@erl.mit.edu) , Bogazici University, Kandilli Obeservatory and Earthquake Research Institute, Cengelkoy, Istanbul, 81220 Turkey
Toksoz, M (toksoz@erl.mit.edu) , Massachusetts Institute of Technology, 77 Massachusetts Ave. , Cambridge, MA 02139
Gurbuz, C (gurbuz@boun.edu.tr) , Bogazici University, Kandilli Obeservatory and Earthquake Research Institute, Cengelkoy, Istanbul, 81220 Turkey
Gok, R (gok1@llnl.gov) , Lawrence Livermore National Laboratory, 7000 East Ave, Livermore, CA 94550
Schultz, C (schultz9@llnl.gov) , Lawrence Livermore National Laboratory, 7000 East Ave, Livermore, CA 94550

Seismic refraction data have been obtained from five strategically placed shots and several quarry blasts in Central and Eastern Anatolia. The largest shot, two tons in 14 boreholes, was in central Anatolia. It was monitored along three radial profiles (north, southeast, and southwest) with 90 temporary stations in addition to the permanent stations of the Turkish National Seismic Network. This shot was recorded to a distance of 300 km. Other smaller shots (up to one ton) were detonated in boreholes in northwest Anatolia, and southeastern Anatolia. Seismic data from three quarry blasts, recorded along linear profiles and distributed network stations, extended the coverage to eastern Anatolia. Travel time and waveform data were analyzed to obtain a 3-D crust, upper mantle velocity model. In central Anatolia crustal thickness varies between 37 and 42 km. The Pn velocity is about 8 km/sec. Crustal thickness changes laterally. It is thinner in northwest Anatolia and slightly thicker in eastern Anatolia. There are abrupt changes associated across the major fault zones, such as the North Anatolian Fault.

S13B-1059 1340h

Study of the Northern sea of Okhotsk Tectonic Zone Using Common Depth Point (CDP) Exploration Profiles

The study of seismo-tectonics in oceanic regions requires special scientific research. At the same time, many offshore regions that are oil-bearing by occur in the zones of high seismic activity, and are covered by the dense net of seismic profiles. These profiles are done by the method of common depth point (CDP). For example, the northern part of the Sea of Okhotsk is characterized by high seismic activity, and more than 50 000 km of seismic CDP profiles of have been done. The analysis of seismic sections has shown that modern active faults have characteristic features which allow to determine their major parameters. On the non-migrated reflection sections, faults are marked by the numerous diffracted waves. The active faults cross the layered sedimentary stratum from the sea bottom and are traced into the acoustic basement. The active tectonic faults are almost always grouped into zones up to a kilometer in width. The dip of the fault planes are steep; and as a rule they are located subparallel to each other. Within the zones of tectonic faults, displacements of the surface of the acoustic basement have a step-like character. Separate portions of the tectonic zones are represented by crush zones up to 200 m wide. At present, the most active tectonic faults are characterized by only horizontal displacement and are not reflected in the sea bottom relief. The most active tectonic zones are spread in latitudinal direction and have the length up to 150-350 km. The maximum amplitude of vertical displacements of the acoustic basement reaches 8-10 km. Such zones are marked by gradients in the gravity field. The length of tectonically active zones and the amplitudes of vertical displacement are the integral mark of seismicity. It makes possible to estimate the maximum possible energy of the earthquakes in the future. Besides the active faults, breaks in acoustic basement and within the low structural stages of the sedimentary stratum are distinguished on the reflection seismic sections of CDP. These faults are not active at the present time.

S13B-1060 1340h

Two-dimensional crustal velocity structure of the southern Korea from the first crustal refraction experiment in Korea

Baag, C (baagce@snu.ac.kr) , School of Earth and Environmental Sciences, Seoul National University, Seoul, 151-742 Korea, Republic of
* Cho, H (hmcho@seismic.snu.ac.kr) , School of Earth and Environmental Sciences, Seoul National University, Seoul, 151-742 Korea, Republic of
Moon, W (wmoon@eos1.snu.ac.kr) , School of Earth and Environmental Sciences, Seoul National University, Seoul, 151-742 Korea, Republic of
Lee, J (jung@knu.ac.kr) , Department of Geology, Kyungpook National University, Daegu, 702-701 Korea, Republic of
Jung, H (hjjung@kunsan.ac.kr) , Department of Ocean System Engineering, Kunsan National University, Kunsan, 573-701 Korea, Republic of
Kim, K (kykim@cc.kangwon.ac.kr) , Department of Geophysics, Kangwon National University, Chunchon, 200-701 Korea, Republic of

We present the data and a seismic velocity profile of the crust along a seismic transect in Korea. The first crustal refraction experiment in Korea was carried out along the 300 km long survey profile through the central part of the southern Korea in the NW-SE direction in December 2002. 198 portable seismometers including 28 three-component ones were deployed at approximately 1.5 km interval. The NE-SW survey profile is almost normal to the major structural trend in Korea. Two shots were exploded in 100 m deep drill wells at the western end and the mid-point of the profile using 1000 kg and 500 kg explosives, respectively. Seismic signals were recorded on 195 receivers out of 198 at a rate of 120 samples per second. The traveltimes of P wave phases, Pg, PiP, PmP, and Pn were identified and shear wave phases, Sg and SmS were also observable on a reduced traveltime plot. A forward modeling scheme of Zelt and Smith (1992) was used to produce the layered P velocity model. The traveltimes of P wave phases and Sg and SmS phases were matched simultaneously under the Poisson solid assumption of crustal material. An iterative modeling resulted in a layered crustal velocity model in which the thickness varies from 28 km to 34 km. Upper crustal velocities range from 5.44 to 6.03 km/s and lower crustal velocities from 6.47 to 6.77 km/s. The underlying mantle has a velocity 7.95 km/s. The mid-crustal reflections recognized in the western part were considered to be representative of the Conrad and it was modeled as a weak velocity discontinuity, vanishing to the east. The crust appears thickest around the middle area and thinnest at the eastern end, located in the Cretaceous Gyeongsang Basin characterized by about 5 km deep surface low velocities. The low velocities also appears in the area of the Okchon fold belt in the central part. The depth to the Moho and P velocities presented here are consistent with those by Park et al., (2003) estimated from the traveltimes of PmP phases recorded on the earthquake network in Korea.

S13B-1061 1340h

Surface Wave Tomography of India

Gaur, V K (gaur@cmmacs.ernet.in) , CMMACS, NAL Belur Campus, Bangalore, 560 037 India
Rai, S S (ssrai_ngri@rediffmail.com) , NGRI, Uppal Road, Hyderabad, 500 007 India

We present group velocity dispersion results from a study of regional distance range fundamental mode Rayleigh waves propagating across the Indian region. One-dimensional path average dispersion measurements have been made for 1001 source-receiver paths and these have been combined to produce tomographic images between 15s and 60s period. These results have significantly higher lateral resolution for the Indian region than is currently available from global and Asian group velocity studies. Testing of the model shows that the average resolution across the region ranges from 5 to 7.5 degrees for the periods used in this study. The Indian Shield is characterized by high group velocities for waves sampling the crust and uppermost mantle. Comparatively lower velocities are observed for the Himalaya due to the thickened crust and the Gangetic plains due to the mollasse sediments and recent alluvium cover in the Himalayan foredeep. The northern Bay of Bengal shows extremely low group velocities due to the thick sediment blanket of the Bengal fan. The Shillong Plateau has high group velocities similar to those observed for the south Indian Shield. The Katawaz Basin in southern Pakistan shows low velocities similar to those seen in the Bay of Bengal. The geometry of the velocity contours south of the Katawaz Basin matches well with the prograding Indus fan into the Arabian Sea. Finally, the Tibetan Plateau has the thickest crust in the region and shows lower group velocities compared to the Indian Shield and the Himalaya at all frequencies.

S13B-1062 1340h

Seismic Attenuation Structure of India Using Lg Coda Q Measurements

* Rowlands, D P (rowlands@esc.cam.ac.uk) , Bullard Laboratories, Madingley Road, Cambridge, CB3 0EZ United Kingdom
Mitra, S (mitra@esc.cam.ac.uk) , Bullard Laboratories, Madingley Road, Cambridge, CB3 0EZ United Kingdom
Bonner, J L (bonner@westongeophysical.com) , Weston Geophysical Corporation, 57 Bedford Road Suite 102, Lexington, MA 02420 United States
Priestley, K (keith@esc.cam.ac.uk) , Bullard Laboratories, Madingley Road, Cambridge, CB3 0EZ United Kingdom
Rai, S S (SSRAI\_NGRI@rediffmail.com) , NGRI, Uppal Road, Hyderabad, 500 007 India
Gaur, V K (gaur@ccmacs.ernet.in) , CMMACS, NAL Belur Campus, Bangalore, 560 037 India

Q measurements from the coda of Lg waves provide excellent constraint on the attenuation structure along source-receiver paths. Lg coda Q values are strongly dependent on crustal structure and generally correlate very well with the tectonic histories of the regions sampled -- high Lg coda Q is typical of tectonically inactive regions (e.g., cratons) whilst low Lg coda Q is typical of tectonically active regions (e.g., active orogenic belts). We use a new tomographic algorithm and the most comprehensive dataset yet obtained from permanent and temporary seismic stations distributed throughout the Indian sub-continent to investigate variations of Lg coda Q in this region. Of particular interest are possible variations between the different cratons which comprise the Indian peninsula. For example, the presence of the Deccan Traps within a much older craton provides an excellent test of the sensitivity of Lg coda Q to relatively recent tectonic events. Initial results show that Lg coda Q varies between 450 and 700 for different regions within India and further analysis will be used to constrain the data in terms of geologic and tectonic provinces. We compare our observations of the scattering Q structure in India with those from previous intrinsic Q studies in the region in order to relate the different attenuation mechanisms.

S13B-1063 1340h

Probing the Mantle Transition Zone beneath Tibet Using Triplicate Seismic Waveforms

* Tseng, T E (tseng1@uiuc.edu) , Department of Geology, University of Illinois at Urbana-Champaign, 1301 W. Green Street, Urbana, IL 61801 United States
Chen, W (wpchen@uiuc.edu) , Department of Geology, University of Illinois at Urbana-Champaign, 1301 W. Green Street, Urbana, IL 61801 United States

Tibet, the world's largest and highest plateau, is a key for understanding how the entire continental lithosphere responds to active collision. Many studies in the past 30 years focused on the intriguing double-thickness of Tibetan crust. However, equally important is the role of mantle lithosphere during continent collision, as density differences in the mantle ultimately drives tectonism. Nearly all tectonic models of Tibet invoke removal and sinking of the mantle lithosphere (of either India or Eurasia), including processes such as delamination, subduction, or convective instability. Dense, removed mantle lithosphere would have continued its descent and probably rests somewhere near the transition zone of the mantle where significant increases in density occur across major seismic discontinuities near depths of 410 and 660 km. As such, current state of the mantle transition zone beneath Tibet offers important constraints on mantle dynamics. Results from the mantle, in turn, have the potential of constraining how the crust achieves a double-thickness in Tibet. We construct a sequence of seismic profiles that sample the mantle beneath Tibet by combining broadband data from several permanent and temporary seismic arrays. The profiles comprise of triplicate waveforms that are particularly sensitive to anomalies near the mantle transition zone such as the size and topography of the 410- and the 660-km discontinuities. The data coverage is particularly favorable beneath the vast Qiangtang terrane of northern Tibet, with profiles extending over an aperture of more than 1,000 km, sampling both the 410- and the 660-km discontinuities. Preliminary results suggest that there are lateral (east-west) variations in the size of the 660-km discontinuity, on the order of about 1% in {\it P}-wave speed, over distances of only about 200 km.

S13B-1064 1340h

Seismicity and Velocity Structure of the Himalayan Collision Zone: Mantle Earthquakes and Fast Velocity Zone in the Lower Crust of Southern Tibet

* Monsalve, G (monsalve@colorado.edu) , University of Colorado at Boulder, 2200 Colorado Ave, Boulder, CO 80309 United States
Sheehan, A (afs@cires.colorado.edu) , University of Colorado at Boulder, 2200 Colorado Ave, Boulder, CO 80309 United States
Schulte-Pelkum, V (vera_sp@cires.colorado.edu) , University of Colorado at Boulder, 2200 Colorado Ave, Boulder, CO 80309 United States
Wu, F (francis@binghamton.edu) , SUNY at Binghamton, PO Box 6000, Binghamton, NY 13902 United States
Rowe, C (char@lanl.gov) , Los Alamos National Laboratory, EES-11 M.S. D408, Los Alamos, NM 87545 United States

P and S-wave arrival data from 28 broadband seismic stations deployed in eastern Nepal and southern Tibet were timed to determine hypocenters of local earthquakes, velocity structure of the crust and upper mantle, and geometry of subsurface interfaces. After picking first arrivals of P and S phases, we estimated hypocenters for over 1600 earthquakes in the area of our network for the time period between October 2001 and April 2003. Locations were determined using a weighted least squares algorithm and a priori 1-D velocity models for Nepal and southern Tibet. We relocated the earthquakes using a double-difference algorithm, recovering about 70% of the events. Cross-sections of relocated hypocenters show remarkable alignment of crustal earthquakes, with depths between 15 and 25 km, along the region of highest relief of the Himalayan Front. In addition to these shallow concentrations, two groups of upper mantle earthquakes clusters stand out: seismicity in the 1988 Udaypur Earthquake zone in southern Nepal and a belt of deep seismicity in southern Tibet, nearly 200 km long and approximately parallel to the Himalayan front. With the depth of the crust-mantle boundary constrained reasonably well via receiver functions, these groups of deep earthquakes can be associated with the existence of a strong lithospheric mantle in the Himalayan collision zone. We found conclusive evidence that at least one of the Udaypur aftershocks occurred in the mantle. The depths of the Tibetan earthquakes are more difficult to constrain, but irrespective of the velocity model used, their mantle depths are robust, even when noise is added to the seismic signals. We suggest the presence of a fast velocity layer in the lower crust of southern Tibet based on Wadati diagrams of deep earthquakes, the variations of travel-time residuals with depth, preliminary 3-D joint tomographic models and the receiver function profile. This fast layer could be a result of eclogitization of the Indian crust

S13B-1065 1340h

Strong Crustal Anisotropy Delineating the Himalayan Decollement

* Schulte-Pelkum, V (vera.schulte-pelkum@colorado.edu) , Dept. Geol. Sciences, University of Colorado at Boulder, 2200 Colorado Ave, UCB 399, Boulder, CO 80309 United States
Sheehan, A (afs@cires.colorado.edu) , Dept. Geol. Sciences, University of Colorado at Boulder, 2200 Colorado Ave, UCB 399, Boulder, CO 80309 United States
Bilham, R (roger.bilham@colorado.edu) , Dept. Geol. Sciences, University of Colorado at Boulder, 2200 Colorado Ave, UCB 399, Boulder, CO 80309 United States
Wu, F (francis@binghamton.edu) , SUNY Binghamton, PO Box 6000, Binghamton, NY 13902 United States

Receiver functions recorded during the 2001-2003 HIMNT PASSCAL experiment in eastern Nepal and southern Tibet display strong and coherent variations in backazimuth, for both radial and transverse components. A strong, shallow signal observed from the Nepalese foothills to the High Himalaya can be modelled as a several kilometer thick, highly ($\sim$ 20%) anisotropic layer with arc-normal plunging symmetry (cf. $SKS$ splitting by Wu and Sheehan, this meeting). Its depth and gentle northward dip coincide with the decollement between India and the Himalaya inferred from structural geology and seismicity. The anisotropy may be caused by shear along the decollement; alternatively, it may already exist at the top of the Indian shield prior to entering the collision zone. Although data coverage decreases to the north, a layer with similar characteristics may be traced to depth in Tibet, dipping more steeply than the Moho under the High Himalaya and the southernmost edge of the Tibetan plateau. Farther north, there is evidence of a midcrustal conversion from a high-velocity layer in the lower Tibetan crust. This layer is substantiated by local tomography (Monsalve et al., this meeting). Our observations of a possible pinching down of the lower Indian crust, merging into a thinner, high-velocity layer, may be caused by eclogitization with its attendant change in volume.

S13B-1066 1340h

Upper-Mantle Velocity Structure Beneath Northwestern Australia From Waveforms of P, PP, S, and SS Waves

* Kuge, K (keiko@kugi.kyoto-u.ac.jp) , Department of Geophysics, Kyoto University, Sakyo, Kyoto, 606-8502 Japan
Kennett, B L (brian@rses.anu.edu.au) , Research School of Earth Sciences, Australian National University, Canberra, ACT, 0200 Australia

In order to understand upper-mantle velocity structure for a region including an old continental lithosphere, we study waveforms of P, PP, S, and SS waves from shallow earthquakes around Papua New Guinea for the epicentral distance range from 10 to 45 degree. These events traverse the mantle beneath the north and west of Australia with Precambrian outcrop. The SKIPPY project with a set of deployments of portable broad-band instruments across Australia from 1993 to 1996, has been followed by further deployments by the Australian National University in the Kimberley region of north-west Australia in 1997-1998 and in Western Australia in 2000-2001. In this study, we exploit data from these deployments with a bandpass-filter between 0.03 and 0.1 Hz preceded by integration in time to obtain the displacement waveforms. Clear waveforms of P, PP, S, and SS can be observed in several record sections where the mid-points lie beneath the Arafura Sea and Northern Territory. We compute theoretical seismograms for 1-D velocity models derived from short-period travel-time data, including ak135 and the regional models njpb (Kennett et al., 1994), and az20sl3 (Kaiho and Kennett, 2000). The synthetic seismograms are compared with each other and with the corresponding observations. For S and SS waves, in addition to variations of the travel times depending on the nature of the models, differences in waveforms are expected for epicentral distances less than 20 degree and from 25 to 45 degree. For some record sections recorded in Western Australia at distances from 35 to 45 degree, we see that the njpb model, which has a thick lid of slightly high velocity, tends to match the observed waveforms better than ak135 and az20sl3.

S13B-1067 1340h

A Comparison of Pn Attenuation in Stable And Active Continental Regions

* Xie, J (xie@ldeo.columbia.edu) , Lamont Doherty Earth Observatory of Columbia University, 61 Route 9W, Palisades, NY 10964 United States

There have been a few studies on Pn attenuation on continents, including those in the shield-like regions of Scandinavia, Canada, central Asia, and active regions of the Tibetan plateau and Basin and Range province. Results of these studies are used to calculate Pn attenuations at a common reference distance of 1200 km, taking into account the different parameterizations used. In the three shield regions the attenuations are found to be similar, within a factor of 2, between frequencies of 0.1 and 5 Hz. In a region in and around north central Tibet (N.C.T.), where mantle lid may be partially molten, Pn attenuation differs from those in shield regions in a strongly frequency-dependent manner: at low frequencies near 0.1 Hz N.C.T. is only marginably more attenuative, by less than a factor of 2, than the shield regions. The differences increase with increasing frequency such that at 4-5 Hz, N.C.T. is about 100 times more attenuative then the shield regions. Pn attenuations in southern and eastern Tibet exhibit a puzzling behavior of being similar to those in shield regions at low frequencies, while being similar to that in N.C.T. at high frequencies. A possible explanation of this behavior is that Pn attenuation beneath Tibet is more affected by intrinsic Q at lower frequencies, and by scattering and multiple bouncing ("whispering gallery") mechanisms at higher frequencies.

S13B-1068 1340h

Temporal variations of crustal characteristics

* Liu, K H (liu@ksu.edu) , Department of Geology, Kansas State University, Manhattan, KS 66506 United States
Gao, S S (gao@ksu.edu) , Department of Geology, Kansas State University, Manhattan, KS 66506 United States
Nair, S K (snair@ksu.edu) , Department of Geology, Kansas State University, Manhattan, KS 66506 United States

One of the most outstanding puzzles related to the formation and evolution of continental crust is whether crust formation processes have always been the same or whether they have changed with time. The most important piece of information to resolve this puzzle is to study the temporal variations of seismologically detectable properties of the continental crust. Previous seismological studies gave contradicting results. Using a modified version of the procedure of Zhu and Kanamori (2000), we have measured three parameters at GSN and GEOSCOPE stations for the purpose of characterizing the properties of the earth's continental crust of different ages. The parameters that we measured include a) crustal thickness (H); b) crustal Vp/Vs; and c) the amplitude of teleseismic P-to-S converted waves relative to that of the direct P-waves (R). For most of the stations, the number of hand-picked, high-quality receiver functions participated in the stacking is between 100 and 600. The stations are approximately evenly distributed over the continental areas of the earth. The stations are divided into three groups. Those in group one are on Cenozoic-Mesozoic orogenic zones, those in group two are on Paleozoic orogenic zones, and those in group three are on Precambrian platforms or shields. On average, the resulting R measurements are the largest at group one stations, and are the smallest at group three stations. The average H values are between 37-40 km for the three groups and are statistically indistinguishable. The mean Vp/Vs value is 1.81+-0.02, 1.77+-0.02, and 1.76+-0.01 for groups one, two, and three, respectively. One of the models to explain the observed reduction of Vp/Vs over age involves increasingly pervasive delamination of the mafic lower crust over time. Thermal events that caused the delamination might have reduced the sharpness and/or the flatness of the original Moho and consequently led to observed decrease of the amplitude of P-to-S converted phases with the age of the crust.

S13B-1069 1340h

Extension to High Frequencies of the Experiment Capability of SEIS-UK

* Brisbourne, A (amb27@le.ac.uk) , SEIS-UK, University of Leicester University Road, Leicester, LE1 7RH United Kingdom
Horleston, A (ach17@le.ac.uk) , SEIS-UK, University of Leicester University Road, Leicester, LE1 7RH United Kingdom
Denton, P (pdt@le.ac.uk) , SEIS-UK, University of Leicester University Road, Leicester, LE1 7RH United Kingdom

The seismic equipment facility serving the UK academic community, SEIS-UK, has recently purchased 15 high frequency seismic recording systems capable of sampling at up to 24kHz. A suite of 3 component 30Hz geophones and piezoelectric accelerometers has also been acquired. The SAQS systems, designed and built by ISS International of South Africa, represent a significant extension in the capability of the SEIS-UK facility. The units record 24bit data from 6 channels, at sample rates of 50Hz to 24kHz, in continuous or triggered mode. An external GPS antenna and removable hard disk allow the systems to run autonomously. The 6 channels can be configured as any combination of tri-, bi- or uni-axial systems. The recording systems are also compatible with broadband seismometers. Although originally designed for the purpose of hardrock mine monitoring, the SAQS system is a good solution for a range of high frequency seismic monitoring experiments, especially controlled source or high-resolution surveys. The SEIS-UK systems have been modified with disk-heaters to extend the environmental operating capability. The addition of an external GPS antenna means the units can be used in the same way as any other passive seismic field system. However, the functionality of the recorder is significantly greater than that of many standard seismic systems. The equipment is currently loaned to the British Antarctic Survey for use as part of a large multi-disciplinary experiment on the Rutford Ice Stream in Antarctica. The instruments will be deployed in an array centred on the main sub-glacial access hole. Instruments will detect events from the bed of the glacier with the aim of investigating the ice flow mechanisms and for determining the relationship between the ice sheet, sub-glacial bed and tidal motion.

http://www.le.ac.uk/seis-uk