| The uplifted and extended Tibetan plateau, the largest and highest mountain belt in theworld today, is resulted mainly from the collision and continuous convergence ofIndian-Eurasian plate since Eocene epoch. The Tibetan Plateau began construction as acomposite in the Mesozoic, when India was a separate continent. About50million yearsago, the Indian plate began to collide with the Eurasian plate. This massive collisioncompressed Earth’s crust and crumpled up a broad area into steep mountains surroundingand high flat plains inner the plateau. The Tibetan plateau, the “root of the world”, withunique set of events that formed it, and continue to build it, appears to be the best locationto study continental geodynamics.Since1924when E. Argand firstly proposed a hypothesis that India is underthrustbeneath southern Eurasia forming a thick crust and high topography, numerous studieshave been made on this subject and tens of competing models have been raised on the issueconcerning the tectonic process and mechanical evolution of the uplift of the TibetanPlateau. Each of the models has its rationale and supportive evidence, but none cansatisfyingly well explain the whole major characteristics of the tectonic process and crustaldeformation of the plateau. On the uplift history of the Tibetan Plateau, although themodels of non-coeval uplift in varied areas and of episodic uplift have been generallyaccepted, their details remain highly debatable. The reasons for these very different modelsand conclusions are related with the following two aspects. First, the uplift history of theTibetan plateau is long and complex. Many previous researches, especially early work,focused on records at local places or several sites, without considering regional differences.So, it is hard to obtain a robust conclusion. Second, many parameters of plateau uplift,such as time, periods, rates and altitudes, were determined based on such postulations likethat when the plateau rises to some height, deep state of lithosphere and surfaceenvironments, including tectonics, climate, ecology, sediments, and temperature, wouldchange notably, and from these variations the time of plateau uplift can be estimated.Apparently, such inferences must have remarkable uncertainties or even mistakes. From the above, it can be seen that in order to approach a more reliable model todescribe the tectonic process and mechanical evolution of the Tibetan Plateau, moreconvincing observations, investigations and constraints are indispensable. Since the middle1990s, a wealth of GPS measurements have been conducted around the Tibetan plateau,which makes it possible to pose quantitative constraints on the kinematic models of theTibetan plateau. This work uses GPS data collected at564stations around the Tibetanplateau since1999to obtain a3D velocity field of the present-day crustal motion of theplateau with high-resolution. This3D velocity field will facilitate better understandingabout the tectonic process the Tibetan Plateau, and provide important kinematic constraintson the dynamics of the present-day uplift and spreading of the plateau. The primarycontents of this thesis are summarized below.(1) During the last two decades, a great number of GPS measurements have beencarried out in China mainland. Among them, two major projects implemented by ChinaEarthquake Administration and other collaborated organizations are most outstanding. Oneis the Crusta1Movement Observation Network of China (CMONOC-I), and the other isTectonic and Environmental Observation Network of Mainland China (CMONOC-II). Thiswork collects data from the projects above, combining with data from Himalaya GPSnetwork implemented by California Institute of Technology (Caltech), and from southernTibet GPS network by Institute of Tibetan Plateau, Chinese Academy of Sciences(ITPCAS), with a total number of GPS stations as858. Useing the software GIPSY/OASIS(Version6.0) from JPL, NASA and the PPP (precise point positioning) model, this workmakes sophisticated processing on daily data and obtains the daily loosely constrainedsolutions. Then, using the software QOCA of JPL, this study performs joint adjustment fordaily loosely constrained solutions of all stations, yielding time series of their coordinatechanges.(2) The GPS position time series analysis shows that vertical deformation is moreprone to non-tectonic deformation noise. As to continuous GPS stations with enoughobservation span and data, the noise has little effect on the secular term deformationestimated, but more on the reference frame. As to non-continuous GPS stations, it is hardto separate the non-tectonic deformation noise from the secular term with limited observation data. In general, there are two ways to remove the noise: one uses geophysicalmodel, and the other employs mathematical function. This work attempts to combine thesetwo approaches for noise removal. In the first step, non-tectonic deformation noise, such asvariations of land water loading and atmospheric non-tidal loading, are eliminated from thetime series analysis by using published geophysical models. The next step chooses thecontinuous GPS stations, which have been model-corrected, to construct a Delaunaytriangle network. Then, for a given non-continuous GPS station, this work calculates itscorrect term by using the3continuous GPS stations of its corresponding Delaunay triangleand the algorithm of inverse-distance weighted average. Next, it calculates the correctionvalues of vertical non-tectonic deformation for this station in the days when GPSobservations were carried out. The spectral analysis shows that in current continuous GPSstations distribution, the method can enhance the deformation signal.(3) This work uses the QOCA software to perform joint adjustment on the dailyloosely constrained solutions of858GPS stations in Tibet, together with271IGS referencestations in the world. Considering several great earthquakes that occurred around theTibetan plateau, the effects of these great events on the relevant GPS stations are removedby using the coseismic module in the software QOCA based on a fault dislocation model ofelastic half-space, and real coseimic displacements of these events. Finally, on theassumption that the3D coordinates of all stations change linearly with time, this workestimates their coordinates, optimal3D velocities, and corresponding standard deviationsin ITRF2008reference. From the horizontal velocity field relative to the stable Eurasia, itis clear that the Indian plate is moving in NNE direction at a rate~40mm along theHimalayas, and this motion slows down toward north, causing strongly shortening betweenIndia and Eurasia and extruding of Tibetan crust. The velocities also present a divergenceoutward from the Himalaya; in the eastern plateau explicitly the velocities exhibit eastwardand southeastward motions and clockwise rotation around the Eastern Himalayan syntaxis.Such a pattern of horizontal GPS velocities can be explained by shortening in theIndia-Eurasia convergence and lateral extrusion of upper crust as described in the literature.The vertical velocity field relative to its stable north surroundings indicates that the Tibetplateau is continuing to rise up as a whole relative to its stable north neighborhood. However, some areas show insignificant uplift or even decline, with details as follows:①The Tibetan plateau is continuing to rise up with an average rate between1to2mm/a. The declining areas generally correspond to the low reliefs of Cenozoic basins.②The range of Himalaya with average altitude of~6000m is continuing to upliftwith the highest rate of~3mm/a in the whole plateau, and the uplift rate is~6mm/a withrespect to the south foot of the Himalayan range, which means that the Himalayan range isstill rising intensively with the underthrust of the Indian plate. In the mid-southern plateau,there is a region of a size of~500km×1500km, showing obvious decline with the ratesbetween0to-2mm/a. From North Qaidm fault to the North Qilian fault, it demonstratesapparent uplift, with typical rates between1to2mm/a. Whereas the basin regions in themid-southern plateau, north margin of the Qaidm basin and Hoxil corridors presentsignificant or insignificant decline.③In the northeastern plateau, along a NNE direction, the east part of the northeasternplateau has a gentle trend of decline at a large scale of~1000km. At the southeasterncorner of the plateau, from the profile along SSE direction, it can be seen that the upliftrate is gently decreasing from0~1.5mm/a in the inner plateau to0~-1.5mm/a outside theplate, with the decreasing of terrain height.(4) There are two quite different models about the Tibetan plateau deformation--block model and continuous model. But none can completely explain current deformationin the Tibetan plateau. Perhaps, a mixture of a variety of models might be able to accountfor tectonic deformation characteristics of the whole plateau, while with a dominant modelfor a specific place. This work attempts to explain3D GPS velocity fields in thenortheastern Tibetan plateau by the block model based on Okada’s work. It uses557horizontal velocity vectors and337vertical ones as the constraints for the model. Throughrational parameters adjustment, a best-fit model of velocity fields and tectonic slip rates arederived from the model. The results show that the horizontal velocity can be explained bythe slip on the major faults, but vertical ones rather unconvincing.(5) The comparison of vertical velocity and long-term topography and K-meansclustering analysis demonstrate that present-day rising and sinking of sub-regions generallycorrespond well to the Cenozoic orogenic belts and basins, respectively. In the mid-southern plateau, the crust shows a weak uplift or even subsidence, implying that thisregion is undergoing a retreat after a rapid supplemental rise, in agreement with the modelof convective removal of mantle lithosphere. |
PDF Full Text Request