Lupei Zhu - US grants

EAR-0001016
Teng, Ta-liang


This project will carry out a 3D crustal structure study making use of the excellent data set recovered from the 1999 Chi-Chi (Taiwan) mainshock and several of its large aftershocks, plus three year worth of seismic background strong-motion recordings of several hundreds smaller (M ~4 - 5) events. These events are recorded by (1) a complement of 700 freefield digital strong-motion stations, (2) 75 3-componeent short-period network stations and (3) 12 broadband stations. The investigators will use the above data set to study a large sedimentation basin in southwestern Taiwan, with the objective to deduce a 3D crustal structure of resolution up to 3 s period. The upper crustal structure will be determined by a standard array processing technique using 1D surface-wave phase velocity inversion analysis for the period range 3 - 15 s. Lower crustal structure can be determined by dispersion data of long-period (10 s - 100 s) surface waves recorded at an array of broadband stations. A group of adjacent 1D crustal models will be pieced together to obtain a 3D crustal structure.

0214259
Koper

This grant provides partial support for an upgrade of the high-performance computing environment for geophysicists at Saint Louis University. Research carried out by the group is wide-ranging and includes Earth structure studies, seismic source studies, seismic hazard analysis, application of space-based remote sensing data to tectonic problems, and development of freely distributed software. The grant will especially benefit three recently hired assistant professors (Koper, Kusky and Zhu) in establishing research programs.

***

0217493
Mitchell

Description:
This award is for support of a joint research project by Dr. Brian Mitchell, Dr. Lupei Zhu, both at the Department of Earth and Atmospheric Sciences at Saint Louis University, Saint Louis, Missouri and Dr. Nihal Akyol, Department of Geophysical Engineering at Dokuz Eylul University, Izmir, Turkey. They plan to conduct a seismological study of the western part of Turkey using groups of both high-frequency (2 Hz) and broadband seismographs. They plan to use data from the high-frequency linear array to obtain a two-dimensional structural model of the grabens and underlying rock by using teleseismsic receiver functions and employing a stacking procedure recently developed by Dr. Zhu. They will use both the linear array and regional array recordings of local earthquakes to perform a combined inversion for precise event location and a tomographic velocity model of the region, and will use data from the broadband instruments for several additional studies. Objectives of this project are to obtain models for velocity structure, including possible anisotropy, for the region, to ascertain the degree of agreement or disagreement among crust/upper mantle models of anisotropy obtained in different locations and by different methods, to infer from those models the directions and consistency of mantle flow and orientation of the crustal stress fields, and to determine the best methods for obtaining information on anisotropic structure in a region of complex structure and tectonics.

Scope:
The selected sites of this study are particularly suitable for the proposed research. Western Turkey is one of the most seismically active continental regions in the world, and much of it is undergoing extensive north-south extensional deformation. Because the region experiences a large number of low to medium-magnitude earthquakes it is particularly suitable for a seismological study of the continental crust in Eurasia. The region's high attenuation values have been attributed to fluid-filled cracks, which tend to cause the crust to be anisotropic. The project will lead to a better understanding of the tectonics of Western Anatolia, which comprises a portion of the Tethysides orogenic belt. The project will enhance international collaborations between scientists in the U.S. and in Turkey. It will involve a recent PhD (Zhu), as well as collaboration with and training of a female seismologist from Turkey.

Mountain building by continent-continent collision is the most
important orogenic process on Earth whose mechanism is not
well understood. The collision between India and Asia is the
largest region of such collision and has produced the world's most
spectacular topography. Within this region, the Tibetan Plateau
stands out as a natural laboratory in which we can test hypotheses
regarding a fundamental process that has impacted continental
dynamics. Collision models, obtained in various studies, predict
significantly different upper-mantle structure beneath the plateau.
The upper mantle, therefore, holds a key to differentiating between
possible collisional models and to unraveling how this archetype
of plateaus came into being. The objective of this research is
to determine high resolution 3-D upper-mantle P and S velocity
structure beneath the Tibetan Plateau and neighboring regions.
The PIs will collect and process available seismic waveform
data of more than 300 broadband stations in the study area from
earthquakes in the upper-mantle distance range. These studies
should provide much improved lateral and depth resolution in the
upper mantle than it has been possible in all previous studies.
The results of the proposed study can be used to verify or dismiss
existing collision models and to develop new ones. The research
topic is also closely related to several undergraduate and graduate
courses (plate tectonics and geodynamics) that the PI is currently
developing and teaching at SLU. The proposed research involves
collaboration between Saint Louis University and the Institute of
Geophysics, Chinese Earthquake Administration. It will use seismic
data collected by several multi-national (China, France, Germany,
USA, etc) projects in Tibet.

Earthquakes are the result of rapid movement of crustal
blocks on faults, which are marked by a narrow fault zone
(FZ) with a width of several hundred meters. Most
earthquake rupture models, including the asperity and
stick-slip paradigms, suggest that FZ structure (geometrical
and material properties) controls earthquake rupture
initiation, propagation, and termination. However,
determining FZ structure at seismogenic depths where
earthquakes nucleate and the majority of slip occurs has
been shown to be extremely difficult. In this project,
seismologists at Saint Louis University use high-frequency
body-wave waveforms from aftershocks near the FZ to
determine fine-scale. Waveform characteristics
enables them to design a new strategy in which the FZ width
and velocities are determined separately, thus eliminating
the trade-off between the two. Furthermore, by using FZ
reflected waves, they can pinpoint the locations in the FZ
where they have obtained FZ width and velocity parameters.
This gives them a unprecedented high depth resolution of FZ
structure. In addition, they are able to relocate
aftershocks with an accuracy of a few tens of meters
relative to the FZ boundaries. They plan to apply the
method to data sets of various fault zones in California,
including the Landers, Hector Mine, San Jacinto, and San
Andreas Fault at Parkfield to determine FZ parameters
(strike, width, velocity and density drops, and Q) and
possible temporary variations.
Broader Impacts: The information obtained is crucial
to developing models of the earthquake source. Experience
gained in this study will help design future FZ seismic
experiments and the methodologies developed can be easily
applied to other regions and data sets, such as those from
the on-going USArray.

This award is funded under the American Recovery and Reinvestment Act
of 2009 (Public Law 111-5).

(a) Broader significance of the project

How does the Himalaya rise up to today's height? How is the Tibetan Plateau (TP) formed? How is the crust of plateau deforming to produce great earthquakes, such as the one we witnessed on May 12, 2008 in Wenchuan, China, which killed over 70,000 people? We know that the plateau is generated by the collision between India and Eurasia, which started about 60 million years ago. However, important details are missing. A major obstacle is our limited ability to "see" through 3D structure below the surface, making it difficult to relate surface geological structures and deformation to the underlying forces. The purpose of this project is to use a variety of seismic imaging techniques and unprecedented amount of seismic data that we will collect from global databases as well as those inside China to image the subsurface structure of central and eastern TP and its margins. These images will provide critical information to test key hypotheses on plateau formation and deformation.

Our research has broad implications for fundamental questions about the mechanisms and processes of mountain building, plateau formation, continental deformation, and seismic hazards in the region. The project will be an excellent opportunity for international scientific collaboration with China. It will support one graduate student from U. Illinois and one from Saint Louis U. We will also engage undergraduates for seismology training and seismic data processing skills.

(b) Technical description of the project.

A great variety of models have been proposed to explain the uplifting, formation, and deformation of the Tibetan Plateau. A major problem is the limited resolution of seismic imaging of the sub-surface 3D structure, making it difficult to relate seismic parameters to geological structures and processes. We propose to use joint-inversion methods involving multiple datasets to improve resolution of both P and S structures of the lithosphere in the central and eastern TP. We propose to jointly interpret P travel times, receiver functions, and surface-wave dispersion measurements from both ambient noise correlation and traditional earthquake-based method to derive 3D models of P and S velocities and anisotropies. In seismic inversion, model parameters often trade off with each other. To improve resolution and to resolve the ambiguity, a combination of different data sets that have sensitivities to different parameters is required or a priori constraints have to be imposed. The abundance of data now accessible makes our joint inversions feasible.

We are particularly interested in the crustal channel flow model, which suggests that mid-lower crust flows in response to topographic loading and the deformation of the upper crust is decoupled from the underlying mantle. We select central and eastern Tibet based on the need for sufficient data coverage and on our desire to study a sufficient large area to avoid possible bias from local heterogeneity and to compare the convergence regime (central Tibet) with the extrusion regime (E. Tibet).

The key questions we seek to address include:

(1) mid-crust channel flow: Is there evidence for widespread mid-crust channel flow? Where in the plateau does it occur?

(2) directions of crustal channel flow: What are the directions of the channel flow? How does the direction change from central Tibet to eastern Tibet and to the southeastern and northeastern margins?

(3) coupling or decoupling of crustal and mantle deformation: How does deformation change with depth? Is the upper crust deformation decoupled from the deformation in the mantle lithosphere?

(4) changes of structure and deformation from central Tibet to eastern Tibet: What is the extent of the India lithosphere underthrusting beneath the TP? What are the differences and connections in structures and deformation at depth among different regions? What controls do the major structures at the margins exert on the crustal and mantle deformation?

Researchers from Southern Illinois University (SIU) and Saint Louis University (SLU) are deploying 45 broadband seismographs and three short-period seismic arrays in the Wabash Valley in Illinois and Indiana. The experiment is part of the on-going EarthScope project and is aimed at better understanding the driving forces of the Wabash Valley Seismic Zone (WVSZ) such as regional lithospheric stresses exploiting a weak zone, glacial isostatic adjustment, and lithospheric basal tractions induced by mantle dynamics. By applying a variety of seismic techniques including body wave and surface wave tomography, receiver functions, shear-wave splitting, and mapping of background crustal seismicity, this seismic experiment will illuminate the seismic structure beneath the study area and impose valuable constraints on driving mechanism for this geologically enigmatic feature as well as further our understanding of intraplate earthquakes.

This project directly advances the Earthscope mission to advance our understanding of intraplate seismic zone dynamics and to improve seismic hazard assessments in the midcontinent. It allows for more quantitative comparisons between the WVSZ and the New Madrid Seismic Zone to its south. It furthers the education of a PhD student at SIU and one at SLU as well as sponsors undergraduate research at both institutions. The students are involved in a number of different aspects of the research from the fieldwork comprising station installation, servicing, and data collection to data analysis and interpretation. These students learn a number of seismic analytical methods that point them towards further research and study within the framework of EarthScope and other cutting-edge scientific initiatives. The PIs of the project work with science teachers at several high schools located in the study area. Two seismographs are located on school property, with display for students to view real-time seismic data and provide access to educational resources from Earthscope.

Worldwide, the number of earthquakes per year decreases rapidly with depth down to ~300 km, then peaks around 550 - 600 km, before terminating abruptly near 700 km. Deep-focus earthquakes (DFEQs), i.e., those occurring at depths below 300 km, are particularly mysterious, as we know that rocks generally deform by creep and flow, rather than by brittle fracture, at these depths, where pressures and temperatures are both very high. Understanding the mechanisms of DFEQs is important because these quakes occur in subduction zones and pose significant seismic hazards in many regions around the globe. It also helps understand properties and behaviors of rocks and how plate tectonics works in the Earth's interior. The experimental capabilities developed in the project will find broad applications in disciplines far beyond earth science, including materials science, physics, and engineering.

In this project, the investigators will combine advanced experimental techniques and state-of-the-art seismological analytical tools to obtain information on the physical mechanisms of fracturing under high pressure and high temperature. The materials to be studied are (Mg,Fe)2SiO4 olivine (the dominant mineral in the oceanic lithosphere and the upper mantle) and harzburgite (the dominant rock assemblage of the oceanic lithosphere). Samples will be deformed in a new class of deformation apparatus equipped with in-situ acoustic emission (AE) monitoring as well as x-ray diffraction and imaging, under a wide range of conditions of pressure, temperature, differential stress, strain, and strain rate. Controlled deformation will be conducted on these materials at pressures up to 14 GPa. A suite of state-of-the-art seismological methods of event detection, location, and source characterization will be applied to the nanoseismograms of AE events to determine rupture mechanisms. Our goal is to understand the physics that connects earthquake mechanics and minerals/rocks at laboratory scales, to provide fundamental insight as to how and under what conditions shear localization occurs, affecting, and affected by, mineral reaction equilibrium and kinetics, and triggers dynamic mechanical instability. Attention will be paid to controlling oxygen fugacity and minimizing water content during the experiments. It must be kept in mind the vast difference in scales between laboratory and subduction zone processes. The team will conduct comparison studies to examine AE source characteristics against those of DFEQs. Thermo-chemo-mechanical models will then be developed and evaluated based on experimental data and seismic observations, and large-scale subduction zone processes. Combining these approaches, the investigators anticipate a significant enhancement of our understanding of the mechanisms for DFEQs by establishing physical models for DFEQs whose testability and scalability can be further examined by computational simulations.