The TAIGER (Taiwan Integrated Geodynamics Research) project is a joint USA-Taiwan program that seeks to model the complex collision processes and the tectonic development of Taiwan, based on extensive observation. Funded by the NSF Continental Dynamics Program and by Taiwan's National Science Council and involving investigators at SUNY/Binghamton, UTIG, USC, Cornell, and Wisconsin, this project has acquired more than twice as much data as originally proposed due to synergistic contributions by Taiwan scientists and Taiwanese government science funding. From FY2006-2009, deployments of seismic instruments on land (IRIS) and at ocean-bottom (OBSIP) maximized the recording of local events and teleseisms, as well as artificial sources specially set off for TAIGER, in order to obtain high resolution 3-D images. To map the areas around Taiwan the PIs were able to use the NSF-supported research ship, the R/V Langseth, to create seismic waves that were recorded by a 6-km long streamer and also ocean-bottom and land recorders. These seismic and magnetotelluric profiles as well as petrologic lab data are currently being analyzed. TAIGER project data combined with geodynamics methods will be used to test a set of existing tectonic models that range from a thin skinned model, in which subduction of continental Eurasian mantle and lower crust is separated from a deforming crustal wedge by a plate boundary decollement, from a thick skinned model, where deformation of crust and mantle occurs within a vertically contiguous system, with progressive thickening of continental mantle beneath the core of the mountain belt.
The current project (TAIGER?s TALE), will expand on thier initial studies to carry out in-depth and advanced studies using the extensive TAIGER data volumes (no new data will be acquired). This work will lead to significant advances in understanding deformational processes associated with the evolution of Taiwan due to plate collision. Studies will focus primarily on seismological analyses to obtain physical properties and rheological indicators as well as the integrative ties with geodynamical modeling efforts. The over-arching strategy is to take advantage of the unique aspects of the acquired data sets, the tectonic environment, and the diverse and complementary participating/collaborating personnel. They will perform their analyses of various data sets in such a way that higher frequency/higher resolution results are nested in the context of broader scale results. They will make extensive use of ?crossover? seismological data sets recorded during TAIGER (active sources collected by passive-source arrays, and earthquakes collected in active-source arrays). Also, analysis/interpretation of these data will be closely tied to geodynamic modeling.
One of the best tools that Earth Scientists have for investigating the processes of contemporary tectonics is subsurface imaging. For that reason a large part of the effort spent in carrying out the NSF-funded TAIGER project, designed to investigate the tectonics of the ongoing arc-continent collision in Taiwan, involved the collection and analysis of seismic data to generate such images. The principal objectives of the current project were to test and refine these images using different data sets and more advanced imaging techniques than were available in the initial analysis of the TAIGER data. Of particular interest was an island-wide image generated using P and S wave arrival time data from a combination of locally and teleseismically recorded events by Kuo-Chen et al (Three?dimensional P velocity structures of the lithosphere beneath Taiwan from the analysis of TAIGER and related seismic data sets, J. Geophys. Res., 2012) which set crucial bounds on the amount of crustal shortening and consumption as a result of the collision. An initial test of this model came from the inversion of an independent source of arrival times (mostly from indigenous sources on Taiwan over a period of several decades) in combination with Bouguer gravity data. While the results of this analysis (Li et al., Moho depth variations in the Taiwan orogen from joint inversion of seismic arrival time and Bouguer gravity data, Tectonophysics, 2014) allowed a more precise definition of crustal thickness beneath Taiwan, particularly on the eastern edge of the island, the first order structures derived by Kuo-Chen et al. were largely corroborated. The other main effort involved a re-analysis of the TAIGER dataset using waveform techniques, mostly a combination of ambient noise tomography and an application of recently developed teleseismic full waveform tomography. The ambient noise analysis mostly used standard methodologies developed by personnel at the University of Colorado, Boulder (mainly Bensen, Shapiro, and Ritzwoller) while the waveform tomography techniques were developed by us (e.g., Roecker et al., 2010 and Baker and Roecker, 2014) and first applied to the TAIGER data set. Our approach was to use a the ambient noise image as a starting model for the upper crust in combination with the Kuo-Chen et al. model for the lower crust and upper mantle. Somewhat surprisingly, we discovered that we could fit both body and surface wave waveforms reasonably well with a model that did not deviate in a significant way from this starting model. This result confirms the basic integrity of the Kuo-Chen model, while at the same time offering a better resolved image of crustal thickening and consumption to the east of the central range of Taiwan. With these results we have achieved our principle objectives of testing the existing models for the subsurface of Taiwan, finding that while we were able to improve the level of resolution and accuracy of those images, the first order implications of those images were largely robust. This project resulted in the direct training of one student, Anastasia Rodzianko, who received a Masters degree for her work on the ambient noise and waveform tomography parts of this project. It also resulted in the peripheral training of several other graduate students through the exposure they received to the algorithms used in generating these images.
