T31C-2012
New Orogenic Model for Taiwan Collision Zone Inferred From Three-dimensional P- and S-wave Velocity Structures and Seismicity
The island of Taiwan is located in the site of ongoing arc-continent collision zone between the Philippine Sea Plate (PSP) and the Eurasian Plate (EUP). Numerous geophysical and geological studies are done in and around Taiwan to develop various models to explain the tectonic processes in the Taiwan region. However, their details have not been known enough, especially under the Central Range. We suggest a new orogenic model for Taiwan orogeny, named 'Upper Crustal Stacking Model', inferred from our tomographic images using three temporary seismic networks with the Central Weather Bureau Seismic Network. These three temporary networks are the aftershock observation after the 1999 Chi-Chi Taiwan earthquake and two dense array observations across central and southern Taiwan, respectively. Tomographic images by the double-difference tomography [Zhang and Thurber, 2003] show a lateral alternate variation of high- and low-velocity, which are well correlated to surface geology and separated by east-dipping boundaries. These images have reliable high-resolution by dense arrays to be able to discuss this alternate variation. We found three high-velocity zones (> 6.0km/s). The westernmost zone corresponds to the subducting EUP. Other two zones are located beneath the Hsuehshan Range and the Eastern Central Range with trends of eastward dipping, respectively. And, we could image low-velocity zone located beneath Backbone Range between the two high-velocity zones clearly. We interpret that these east-dipping high- and low-velocity zones can be divided into two layered blocks and the subducting EUP, each of which consists of a high-velocity body under low-velocity one. Layered blocks can be interpreted as stacked thrust sheets between the subducting EUP and the Northern Luzon Arc, a part of PSP. These thrust sheets are parts of upper- and mid-crust detached from the subducting EUP. The model of continental subduction followed by buoyancy-driven exhumation can explain the existence of stacked thrust sheets. Thus we propose a new orogenic model, as referred to as the 'Upper Crustal Stacking Model'.
T31C-2013
Study of the Taiwanese Orogen from Absolute Gravity Data
The island of Taiwan is located on the convergent boundary between the Philippine Sea Plate and the Chinese continental margin. In south-central Taiwan, the collision between the Luzon volcanic arc and the Chinese continental margin has caused intense crustal thickening and shortening in the rising mountain range. The mountain building processes are very alive and well illustrated by the rugged topography, rapid uplift and denudation, young tectonic landforms, active faulting and numerous earthquakes. The project AGTO proposes to study this natural laboratory of building of the mountain ranges using absolute gravimetry (AG). It also includes relative gravimetry (RG), GPS measurements and modelling. The target area is the southern part of the Island, for the Luzon volcanic arc to the east (Lutao island) to the western coastal plain (Tainan county), crossing the Coastal and Central ranges. This region probably suffers the highest rates of rising (between 1 and 2 cm/year of vertical movement in the Central range documented by permanent GPS measurements). Two AG measurements of the nine AGTO gravity bases have been performed respectively in November 2006 and November 2008. We will present these data and compare with models. Using simple models, we have estimated vertical movements and gravity field's variations which can be expected near the AG sites. Gravity variations due to the deformations are dominated by plate and free air effects, i.e. elevation of the topography. By comparison with these effects, those generated by mass transfers are weak: maximum 0.1 microgal/yr with the thin-skinned tectonic and 0.4 microgal/yr with the thick-skinned tectonic.
T31C-2014
Modeling a large co-seismic stress change associated with the Chi-Chi earthquake, Taiwan
Following the 1999, M 7.6 Chi-Chi earthquake, the Taiwan Chelungpu Drilling Project (TCDP), cored through the Chelungpu Fault in a 2-km-deep scientific borehole, near Dakeng, Taiwan. A comprehensive suite of geophysical logs were collected to measure physical properties of the fault zone (such as P- and S-wave velocity, density and resistivity) and borehole image logs were used to determine variations in stress orientation with depth through observations of stress-induced borehole breakouts. These data show that near the fault, the azimuth of the maximum horizontal principal stress (SHMAX) changes by about 90º from the regional tectonic stress direction (N120ºE), which is observed in the borehole both above and below the Chelungpu Fault. Mini-hydraulic fracturing tests were used to determine the magnitude of the minimum horizontal principal stress (Shmin) at multiple depths. Through dislocation modeling, we used these post- earthquake observations to constrain stress orientations and magnitudes before the earthquake and the stress changes on and near the fault as a result of the earthquake. Our simulations replicate the abrupt stress rotation observed in the images logs at the depth of the Chelungpu Fault. In addition, this modeling indicates that the magnitudes of SHMax and Shmin changed markedly during the earthquake. In order for the co-seismic stress changes to result in a 90º stress rotation near the fault, the state of stress prior to the earthquake had to have been a reverse faulting stress regime (as expected), but with SHMax equal to Shmin and larger than Sv. While such a stress state is quite unusual, it is consistent with the fact the ~N-S strike of the Chelungpu Fault changes abruptly to ~ E-W just north of the TCDP drill-site. During the earthquake, SHMax appears to drop from ~42 MPa to ~27 MPa in the vicinity of the fault, whereas Shmin changed much less, from approximately 42 MPa to about 40 MPa, resulting in the localized, 90º change in stress orientation.
T31C-2015
From passive continental margin to mountain belt: Insights from analytical and numerical models and application to Taiwan
The physics of deformation at plate boundaries is insufficiently understood. Taiwan is particularly interesting in this context since geological and geophysical arguments indicate that passive-margin sediments were exhumed into an active mountain belt over a relatively short period of time. Classically, Taiwan has been described as a critical wedge orogeny, in which deformation is entirely restricted to the crust. More recent data, however, indicate that deeper parts of the lithosphere also participate in deformation. In order to study the physical feasibility of various endmember hypothesis, we have performed two-dimensional (2-D) thermo-mechanical numerical experiments in which both mantle, lithospheric and surface processes are taken into account and in which crustal deformation evolves as a function of mantle flow. Our setup consists of ocean-ocean subduction, followed by passive continental margin deformation. Results indicate that in most cases the subducting slab becomes close to vertical upon arrival of continental material to the trench. Whereas crustal rheology, deformation and surface processes do not modify mantle flow significantly, they are important for crustal-scale processes. Relatively little crustal exhumation occurs if the crust is homogeneous and erosion rates are small. Significant exhumation in a dome-like structure is obtained if a weak lower crust and large erosion rates are present. The presence of an oceanic arc in the overriding oceanic lithosphere modifies, but does not control, crustal deformation. Syntheticmetamorphic facies maps and thermochronological data, computed from the numerical models, are in reasonable agreement with observations in Southern Taiwan if exhumation occurs in a dome-like structure. Further physical insight in the doming mechanisms is obtained by a combination of analytical techniques and crustal-scale numerical models. This analysis indicates that the dome-like exhumation mechanism can be attributed to a compressional mechanical instability, facilitated by both a weak lower crust and large rates of erosion. Both conditions are likely to be satisfied in Taiwan.
T31C-2016
Observation for high velocity gradients in the western plain of Taiwan from TAIGER experiment
Seismic data have been collected through the TAIGER (TAiwan Integrated GEodynamics Research) project for testing models of Taiwan orogeny. Both land broadband instruments and short-period geophones from across island transect have recorded high quality data from earthquakes and active sources. Especially, TAIGER explosion experiments provide an excellent opportunity through seismic refraction and wide-angle reflection studies to image the high resolution crust and upper-mantle structure beneath Taiwan. However, the most significant feature observed in the near source stations located at the western plain is presence of high-energy later arrivals. Those arrivals are identified as free-surface reflected multiples. Those high- amplitude multiples almost completely mask all other lower amplitude signals (seismic refraction and wide- angle reflection) from the deep crust. The nature and generation of those waves are demonstrated with the help of synthetic examples. It is found that the presence of high-velocity gradients in the shallow crust is necessary to generate these high-energy multiple diving waves. Detailed analysis those observation provided significant information of velocity gradients in the western Plain. The new velocity model may improve the accuracy of ground motion prediction or enhance images of weak signals from deep crust reflection and refraction.
T31C-2017
New Lithospheric Model of Taiwan based on the Receiver Function Method
Taiwan is situated on the junction area between two subduction systems. The complex orogeny was developed by collision between the Eurasian continental plate and Philippine Sea plate and is still active in the present. Therefore, Taiwan provides unique opportunities for geophysical imaging of the ongoing process underneath. The TAiwan Integrated GEodynamics Research (TAIGER) combined a field program of active and passive seismology, which will undoubtedly be a major step forward in understanding mountain building process. In 2006, we developed a new crustal model of Taiwan from teleseismic waveforms by the receiver function method. We determined lateral variation of Moho discontinuity, crustal thickness (H), and Vp/Vs ratios (Kappa) for each permanent broadband station using all the available teleseismic data collected by BATS (Broadband Array in Taiwan for Seismology) and CWB (Central Weather Bureau). All the broadband stations are distributed uniformly over the whole Taiwan area so that we could delineate the Moho depth contour map. Recently, we concentrated on the three linear temporary arrays of the TAIGER project and obtained three high-resolution images of crustal structure across Taiwan along west-to-east direction from north to south by using the CCP (common-conversion-point) stacking of teleseismic P-to-S converted waves. Sharp impedance contrasts in these images clearly show the relief of each of seismic discontinuities in the crust and upper mantle. The preliminary results show that the Moho depth, 40 to 50 km of central Taiwan is deeper than in other parts of the island, which suggests crustal thickening due to collision. In addition, shallow part of western foothill area show highly acoustic impedance which probably results from thick sediment.
T31C-2018
Waveform Simulations For TAIGER Data Sets From Taiwan 3D Reference Velocity And Moho Boundary Models
Studying seismic waveform variations in space and time is an important issue to investigate structural heterogeneities and ground motion responses for seismic hazard mitigation. The available 3D reference velocity models from transmission tomography studies are mainly limited by depth resolution, refraction arrival picks without explicit considering later phases and the spatial distribution of earthquakes and stations. Seismic data collected from the TAIGER (TAiwan Integrated GEodynamics Research) project can provide a valuable opportunity for studying deep crust structures. Evaluation of 3D reference models and update their shallow velocity structure is presented through travel-time and waveforms studies. Even though a well-defined multi-scaled reference velocity model of Taiwan is being debated, existing models are still important to study the structural heterogeneities and path effects through parallel computation of 4th-order staggered grid FD 3D waveform simulation. Simulation utilizes both far-field point and finite-dimensional moment tensor sources to investigate effects on Moho reflections and lateral velocity variations. Constraints on Moho reference boundary obtained from receiver function studies is discussed and compared with data collected from TAIGER project. For controlled source experiments, synthetic simulations show clear and focused Moho reflections in the 3-C data. Simultaneous 3D simulation of all available seismic records provides unique constraints on reference velocity model known so far. The waveform simulation will provide a fundamental research platform for future full 3D waveform inversion.
T31C-2019
Crustal structure across Taiwan orogen based on the TAIGER 2008 land refraction experiment
The ongoing orogeny of Taiwan involves a complex interaction often considered as a typical example of arc- continent collision. Tectonic models to explain how this collision occurs and produces with mountain building, of which Taiwan's is one of the fastest rising in the world, range from thin-skinned to lithospheric-scale end- members. In the thin skinned model, subduction of continental Eurasian mantle and lower crust is separated from a deforming crustal wedge by a plate boundary decollement. In the latter, deformation of crust and mantle occurs within a vertically contiguous system with the core of the mountain belt. The 3D geophysical signatures are fundamentally different between these models. The Taiwan Integrated Geodynamics Research (TAIGER) project is a joint USA-Taiwan effort to understand this arc-continental collisional system. Which focus on the questions including: How does the continental subduction play an important role in arc-continent collision? How does such an orogen evolve over time? And so on. The TAIGER project uses passive seismology, controlled-source seismology, magnetotellurics, petrophysics, and geodynamical modeling to obtain new 3D subsurface constraints. Results of part of these methods are presented in companion posters. During Feb-Mar 2008, the TAIGER team conducted an active source land refraction seismic experiment across Taiwan. This is the land source part of TAIGER project that includes 10 high energy sources in two transects across northern and southern Taiwan ranging between 500-3000 kg, which were recorded by PASSCAL Texan, R-130, Q-330, SAMTAC801H and Micro OBS instruments. All of the data are processed to the industrial SEGY format and one active source can produce about 8 to 10 profiles. Preliminary examination of the data reveals several interesting observations: 1) a PmP phase appears around the cross-over distance 150~180km which predicts a Moho depth of 35~45 km. 2) thicker crust in the north (estimated 45 km) than in the source (35 km), 3) a 3 km thickness sedimentary layer with P velocity 3.7km/s above the basement (5.7 km/s) in the northern Taiwan¡¦s western coast verse a 2 km layer (2.7 km/s) over a basement (4.9 km/s) in the southern Taiwan, 4) high-velocity gradients in the shallow crust of the western coast exists to generate high-energy multiples, 5) flat layers in the western coast area, however, complicated 3D structures in the mountain area are needed to explain complicated phases.
T31C-2020
Anomalous seismic attenuation beneath southeastern Central Range of Taiwan: Observations from a linear cross-island array
A remarkable aseismic zone in southeastern Central Range of Taiwan was noted for sometime in many well- located seismicity maps of Taiwan. Its cause remains a puzzle. Seismic attenuation may shed some light on possible cause of the aseismic zone. In this study seismic attenuation across southern Taiwan is investigated in terms of t* data and Q models. The t* data are obtained by fitting the observed spectra of P and S waves with a ω2 model using broadband records of a 25-station linear array deployed across southern Taiwan. The Q models are then determined by inverting the observed t* data. For P and S waves coming from northeast of the array, both tp* and ts* data at the eastern stations show remarkable decreasing trends with depth down to 40 km in contrast to the western stations. The inverted Qp and Qs profiles across southern Taiwan reveal a high attenuation zone at depth of about 20 km beneath the eastern stations traversing the aseismic zone in southeastern Central Range. On the contrary, no high attenuation zone is found from the tp* and ts* data of P and S waves coming from south of the array. This implies the high attenuation zone does not extend south beyond the linear array. In addition to absence of seismicity, this high attenuation zone is also marked by other attributes of very active collision orogeny, such as high mountains, fast crustal uplift, high heat flow, low Vp and Vs, and strong anisotropy. Consistent with these corroborating features the anomalously high attenuation zone appears to signify presence of high geothermal and/or partial melting regimes in the middle crust of southeastern Taiwan due to intense collision of the Eurasian and Philippine Sea plates.
T31C-2021
Inversion of explosive and teleseismic data for two-dimensional velocity structure beneath southern Taiwan
While relative arrival times of teleseismic events represent integrated effects of receiver-side velocity anomalies, the arrival times of wide-angle reflection enable us to constrain exclusively the shallow velocity structure. As a result, the deeper velocity structure is better resolved with teleseismic observations when combined with wide-angle reflection traveltimes. We thus conduct a joint inversion for two-dimensional velocity structure beneath southern Taiwan, using data both from the land-based active source during the Taiwan Integrated Geodynamics Research (TAIGER) project and from the Tonga-Kermadec deep earthquakes. The Pg phases of the four active explosions in southern Taiwan as recorded by a linear array of around 580 Texan instruments across southern Taiwan are identified and arrival times picked. The maximum offset distance is about 120 km. The teleseismic data are recorded by a nearly co-linear array of 25 broadband stations. The relative arrival times of all observable P, pP, and sP phases from six Tonga deep earthquakes are determined by the genetic algorithm with the cost function based on sum of cross- correlation over each pair stations. Topographic effects of all stations are considered. We then apply the Fast Marching Tomography (FMTOMO) Package (Rawlinson et al., 2006) to invert the two-dimensional velocity beneath southern Taiwan, which best fits both datasets. Preliminary results show that the velocities at shallow depth (less than 10 km) are faster beneath the eastern Taiwan than those beneath the western Taiwan. There is also a low velocity zone beneath Central Range at greater depth. The result will be refined by adding more observations.
T31C-2022
Evaluation of Regional Seismic Models in Taiwan by the TAIGER Active-Source Experiments
In the Taiwan region, the complex geometry and mechanism in the collision and subduction zone between Eurasia and the Philippine Sea plates result in extremely complicated structure in the crust and uppermost mantle. As part of the TAiwan Integrated GEodynamics Research (TAIGER) project, a series of active-source experiments were conducted in the spring of 2008 providing high-quality records of a number of explosions at nearly 1000 recording sites distributed on the island of Taiwan and the surrounding seafloor. In this work, we use the TAIGER explosion records to examine the effect of three-dimensional (3D) crustal and uppermost mantle structure beneath Taiwan on the traveltimes of seismic waves and evaluate the performance of currently available 3D models for the Taiwan region in predicting seismic traveltimes. We measure the onset times of the first-motions from the records of the TAIGER explosions. The spatial distribution of the observed first-arrival times exhibits clearly the possible pattern in the lateral variations in the P-wave speed. We also calculate the synthetic seismograms in both 1D and 3D models using two different methods: the frequency-wavenumber (FK) and finite-difference (FD) methods. Both methods calculate the seismograms accurately, although the FK method can only be used for 1D models. We also measure the first-arrival times in all of the synthetic seismograms. Comparisons between the model-predicted first-arrival times and the observations demonstrate the improvement of the 3D models over the 1D model. Detailed examination of the differences in theoretical and observed first-arrival times also allows us to evaluate the relative strengths and weaknesses of the current 3D models in different parts of the region.
T31C-2023
Relocation of Earthquakes and Tectonics of offshore Taiwan with the addition of TAIGER BBOBS
Ten BBOBS (broadband ocean bottom seismometers) were deployed from 2007/11 to 2008/05 as a part of the TAIGER (Taiwan Integrated GEodynamic Research) project. Out of the ten stations eight recorded local earthquakes with high signal-to-noise ratios. In this study, a total of 861 events offshore Taiwan are relocated by combining the readings of P wave arrival times from BBOBS and from the CWB short period stations on land. The analysis of integrated off-shore and on-shore data has resulted in tightening of three obvious clusters of offshore earthquakes. The three clusters of earthquakes are located in (i) the source region of the 2006 Hengchun M7.1 earthquake southwest of Taiwan; (ii) the NS trending Luzon forearc southeast of Taiwan; and (iii) east of the Luzon Arc near Lu-Tao. The data are not yet sufficient for 3-D tomographical velocity determination or relocation; in this study different methods (HYPOSAT, JHD, HYPODD and HYPO71), initial locations and velocity models are used to assess the robustness of the results. For cluster (i) the relocated events define a NW striking zone in which the hypocentral distributions is highly method/velocity-structure dependent, with RMS errors remain similar. The events show either as a more diffused group around 40 km or in two well-defined layers at ~20 and ~50 km. Further tests are being conducted to resolve the discrepancy. More BBOBS data to be retrieved in June 2009 may provide better velocity constraints. For cluster (ii) the active seismicity in west of Luzon Arc is shown to be a NS striking and east 65 degree dipping zone. This seismicity extends northward to the southern end of the Longitudinal Valley, striking N15°E. For cluster (iii), a M6 occurred in northeastern Lu-Tao on 2008/04/23, shortly before the recovery. The NW trending and nearly vertical-dipping seismic zone coincides with one of the planes of the Harvard CMT solution (strike: 158°; dip: 76°; and rake: -10°), i.e., the main shock may be an essentially vertical left-lateral strike-slip fault. While cluster (i) is related to subduction of the Eurasian plate under the Philippine plate, clusters (ii) and (iii) reflect forearc tectonics
T31C-2024
Ambient noise levels in the Taiwan region
To characterize the island-wide background seismic noise in Taiwan, we estimate the power spectral density (PSD) at broadband stations of both the BATS (Broadband Array in Taiwan for Seismology) and the TAIGER experiment (Apr. 2006~Apr. 2008) for periods ranging from ~0.2 to 100 seconds. A new approach to calculate the probability density functions of noise power (PDFs, MaNamara and Buland, 2004) is used in this study. The results indicate that the cultural noise at higher frequencies is significant at populated area, which shows diurnal and weekly variation as what we expected. The noise power for microseisms centered at a period of ~5 seconds around the western costal plain show ~20dB higher than what observed at eastern Taiwan. This observation supports the inference that the coastal regions having narrow shelf with irregular coastlines are know to be especially efficient at radiating the predominat microseisms. Results from the linear array across central Taiwan demonstrate that the average noise power is quietest at the eastern Central Range. We have mapped the PDF mode for stations at various periods to see the spatial distribution of ambient noise levels, which could be used as the basic information for future station siting. Temporal variation of noise PSD is also present to provide a quantitative description of the seismic data quality collected by both BATS and TAIGER experiment. Some operational problems like base tilt, sensitivity change can be identified easily as well.
T31C-2025
The Tectonic Significance of 3-D Density Model in Taiwan
Taiwan is located in the collision and subducting area follows complex tectonic background made varied geologic structures. The researches of seismic and geologic study in Taiwan are very adequate but the researches of gravity are insufficient. In this study, we converted the P wave velocity to density parameter by the referenced formula. The proper density structure is inversed by using the initial density model converted from tomography data. Hence after inversion computing from using the converted model, we get the finial density model fitting Bouguer anomaly map. The inversion result shows the observed and inversed gravity isogal maps fit very well and the correlation coefficient value between these two maps is 0.98. For further understanding the tectonic structures beneath Taiwan, the 3-D density model is shown by 2-D horizontal profile from up to down and 2-D vertical profiles from north to south. There are 14 horizontal profiles in 70 km depth and the interval distance is 5 km. We also project the seismicity on each horizontal profile from 2.5 km up to 2.5 km down. Roughly, the seismicity beneath Taiwan are almost happened in 0 to 30 km depth. The distributions of seismicity in each layer are along the border between the positive and negative density perturbation areas especially in the east Taiwan and the border between Central Range and Western Foothill. In 2-D vertical profiles from north to south, we cut six profiles across Taiwan from west to east. The horizontal length of profiles is 200 km and the depth of profiles is 70 km. From north to south, we can find out the Moho interface change from east to west roughly. In CC¡¦ and FF¡¦ profiles, the sediment layer is thicker than other profiles and also provides evidence for the existence of the low anomaly area exist. We also select the 3.2 g/cm3 and 7.5 km/s as the density and velocity value of Moho interfaces from models. The density Moho depth beneath Central Range is almost 40--50 km. The thickest crust is located in the north Central Range about over 55 km. The same result is shown in tomography model. The density Moho depth is more smooth relief than tomography Moho depth under Central Range area. These places are also plentiful filled with the complex tectonic activities and the orogeny process. The other tectonic profiles are either compared with density profiles in this study.
T31C-2026
Elements of Lithospheric Structure Beneath Taiwan from Seismic Reflection and Teleseismic Conversion Recordings during Project TAIGER
Recordings of explosive sources collected during both the test phase and main phase of land seismic surveying during Project TAIGER exhibit near-vertical reflections from intracrustal interfaces as well as the Moho beneath western Taiwan. A least some of the intracrustal events may correspond to basal decollements into which foreland thrusting is rooted; however these markers cannot be traced with any confidence beneath the central range. The Moho reflections correspond with converters mapped by sP and Ps receiver functions computed from TAIGER broadband stations. sP receiver functions were also used to estimate lithospheric thickness, but the results are ambiguous due largely to the short deployment period for these instruments. The most prominent observation from the land source recording is an anomalously strong reflector observed on two distinct shots in NE Taiwan. This "bright spot" is similar to amplitude anomalies encountered by deep seismic surveys elsewhere and usually associated with fluids at mid-crustal depths (in this case ca 20 km). Either magma, water or both may be involved. The location of the bright spot beneath the Central Range adjacent to the Lian plain suggest that it may result from partial melting triggered by exhumation, or alternatively to melting associated with propagation of the Okinawa trough at depth into NE Taiwan proper.
T31C-2027
New insights on the structures and age of northwestern West Philippine Basin
From 2006 to 2007, swath multi-beam bathymetric and geomagnetic data were collected in the northwestern part of the West Philippine Basin (121¢XE ~ 125¢X20¡¦E and 19¢X50¡¦N ~ 23¢X50¡¦N). The bathymetric images show obvious NE-SW trending fracture zones and seafloor fabrics 123¢XE to 125¢X20¡¦E. Moreover, an overlapping spreading center (OSC) near 22¢X20¡¦N; 125¢XE was identified. The apparent morphology indicates an excess supply of magma during its formation and suggests a southeastward migration of mantle materials. We have also analyzed the newly collected geomagnetic data. The results show that the direction of the magnetic lineations is NW-SE and the magnetic anomalies pattern displays a pattern of symmetry. We have conducted an age model simulation to infer a possible age of the oceanic crust of the northwestern part of the West Philippine Basin. According to our age model, we propose that the spreading direction in this area is NE-SW, and the spreading phase ceased at 43.5 Ma with a half spreading rate of 20 mm/yr in the northern part and 28 mm/yr in the southern part. This portion corresponds to the earliest age found in the West Philippine Basin.
T31C-2028
Development of Eulerian numerical procedure for free surface toward plate-mantle simulation
In the geophysical simulation study,
one of the great challenges is to
reproduce a realistic plate tectonics
with mantle convection simulation.
We develop an Eulerian numerical scheme for the steady Stokes flow
to solve the deformation of rigid material (plate tectonics)
induced by thermal convection of soft fluid (mantle convection).
Our simulation scheme combines (i)
the multigrid method together with
a fast and robust smoother algorithm named ACuTE
by Kameyama et~al. (2005),
and (ii) an low diffusive semi-Lagrangian advection algorithm named
CIP-CSLR-CS by Furuichi et~al. (2008).
Since it is easy to optimize in vectorization/parallelization,
our method is suitable for large scale simulation.
According to our recent study,
in which we carry out the validity test
of our simulation scheme
for a large deformation problem
by using the fluid rope coiling event,
the current approach in the grid resolution
of our large scale simulation successfully
reproduces not only qualitative but also
quantitative behavior of a deformation of curved rigid plate.
This indicates that
by introducing a proper treatment of a free surface,
our scheme may solve the whole system of solid earth
on the mechanical model
including the surface deformations
without serious quantitative errors.
In this study, we are trying to simulate
self-gravitationg motion of the Stokes flow as
the free surface problem.
It means that the spherical shape like the real earth
is simulated in the Cartesian grid.
We would like to show the development of our numerical
treatment toward free surface problems.
http://www.jamstec.go.jp/esc/research/Solid/index.en.html
T31C-2029
Characterizing the lithospheric-sublithospheric upper mantle system: its thermal, compositional, seismological, and rheological structure in 3D.
The thermal and compositional features of the lithospheric/sublithospheric mantle need to be well resolved in order to understand its origin and evolution. In this session we present LitMod3D, an interactive computer code developed to perform combined geophysical-petrological modelling of the lithospheric/sublithospheric upper mantle down to 400 km depth. LitMod3D is a collection of FORTRAN subroutines organized in two main modules: a forward calculation module and an interactive interface module used to visualize and modify data. The forward module solves the heat transfer, thermodynamic, rheological, geopotential, and isostasy (local and flexural) equations for any given lithospheric-sublithospheric upper mantle model, and outputs temperature, surface heat flow, density, seismic wave velocities, geoid and gravity anomalies, elevation, and lithospheric strength. Outputs from the model are compared with available geophysical and petrological observables to constrain possible compositional and thermal fields within the crust, lithospheric mantle, and sublithospheric upper mantle. In contrast to other available 3D codes, LitMod3D is built within a thermodynamic self-consistent framework, where all thermophysical properties are functions of the Gibbs free energy of the stable assemblage. The latter is obtained by Gibbs free energy minimization using the available software PERPLEX07. The final result is a self-consistent lithospheric/sublithospheric model that simultaneously fit all available geophysical and petrological observables, which reduces the uncertainties associated with the modelling of these observables alone or in pairs, as commonly done in the literature. LitMod3D is available from the first two authors upon request.
T31C-2030
Dense Root Removal by Asymmetric Delamination in Sierra Nevada, California: Insights from Numerical Modeling
Recent studies provide compelling evidence for an event of removal of lithospheric mantle in southern -and possibly central- Sierra Nevada (SN) mountains, California (Zandt et al., Nature, 431, 2004). A sequential history of foundering of the ultramafic root of the Sierra Nevada batholith, with a pronounced asymmetric flow, is proposed to explain a number of geophysical and geological observations, including a fast seismic velocity in the mantle located to the west of the SN crest, a gap in the Moho, recent subsidence and tilting of the Sierra Nevada, and a change in mineralogy of the xenolith population recorded at the surface. In the present study we focus on the quantitative evaluation of this conceptual model. We apply new thermo- mechanical algorithms, developed in MATLAB code, suitable to study the temporal evolution of laterally migrating lithospheric delamination. The motion equation, formulated in terms of the stream function, and the coupled thermal equation are solved applying finite difference techniques. Our physical modeling is shown to properly reproduce the first order features of the conceptual model for lithospheric delamination in the Sierra Nevada. We investigate the evolution of a dense ultramafic root, which brings about a Rayleigh-Taylor gravitational instability. Following our preliminary results, the presence of a fluid-weakened lithosphere, located just east of Sierra Nevada, is required to reproduce the asymmetric development of this instability, as previously proposed by Zandt et al. (2004). This weak rheology zone, which is modeled by means of a reduced viscosity, is shown to enable the ascent of asthenospheric material and westward propagation of delamination. Our predictions are also consistent with the previous inference of the V-shaped cone of crust being dragged down into the downwelling mantle (i.e., the Moho gap). Present results highlight that viscous drag is also likely responsible for present-day surface subsidence.