| The investigation of the earth structure, especially the crust and upper mantle structure from seismic travel time and waveform data is one of basic topics of seismology. The exploration of the fine lithospheric structure is not only a main project of the continental dynamics and earthquake dynamics, but also for the prospection of natural resource and seismotectonic environment. Since the 1980s, with the quick development of the global and regional seismic networks in the world, especially movable broadband seismic observation technique, the teleseismic receiver function (RF) method has become the most important method of exploring the crust and upper mantle structure.In light of some problems in theoretical and applied investigation of the RF method, the theory about the RF method, including the technique of the RF isolation, the nonlinear RF inversion and the grid-search method of estimating the crustal thickness and Poisson's ratio as well as the RF migration technique, has been discussed systematically in this dissertation. Our discussion is divided into two parts: in the first part are reviewed systematically the main advances on the RF method and proposed our improvements in light of the theoretical problems and method shortcomings. The second part deals with several applications of the RF method in the data interpretations of the passive broadband seismic array observations. Our results demonstrate the actual usefulness of the RF method in the investigations of the crust and upper mantle velocity structure from passive seismic array observations. Since all of our research areas are the hot-point areas in the Chinese continental dynamics, our results will facilitate to settle scientific problems related to these areas.In the following are outlined our main results given in this dissertation:1. Theoretical study of receiver function method(1)Two methods of isolating the RF from the teleseismic P waveform data, the maximal likelihood deconvolution method in frequency domain and the maximal entropy de-convolution method in time domain, are discussed, respectively. Their differences and relationship as well as the advantage and disadvantage in applications are reviewed in detail.(2)To deal with the sediment effects, an improved grid-search method of considering the shallow structure for estimating the Poisson's ratio and crustal thickness is proposed.(3)New advances of the RF migration techniques are reviewed systematically and two commonly used migration methods, the common conversion point stacking method and the Kirchhoff scattering method, are summarized in the unified theoretical frame.(4)Travel times of crust reverberations are very close to that of Ps converted phase in the crust and top of the upper mantle, causing difficulty in imagining discontinuities in these depth range. To solve this problem, a new wave equation migration method for crust discontinuities is proposed, which utilizes the 1-D velocity structure obtained from receiver function inversion as the reference model. In this method, the Green's function for the reference model is synthesized by the reflectivity method and its depth derivative is calculated by a numerical method. The final migration imaging is formed by the cross-correlation of the receiver function observed at the surface and the calculated depth derivative of Green's function. Both theoretical and real data calibrations show that it is a feasible procedure to reduce the effect of reverberations and can significantly improve the migration result for crust discontinuities. Though it takes the 1-D model as reference, when using receiver functions from multiple back-azimuths, we can still obtain the inhomogeneous structure. Meanwhile, this method foretells the importance of 3-component receiver functions in RF migration.2. Crust and upper mantle velocity structure beneath the northeastern Tibetan Plateau and OrdosFrom teleseismic P-waveform data recorded at the movable broadband seismic array across the north-eastern Tibetan Plateau and Ordos block, the fine shear wave velocity structure and the first-order discontinuities of the crust and upper mantle down to 660km deep along the profile about 1000km long are investigated by using the non-linear RF inversion method and the migration technique, respectively. Our results manifest that(1) The crust structure from the northeastern Tibetan Plateau to Ordos block has a clear blocked characteristics. Totally, there are five major tectonic units along the passive seismic profile. The boundaries between different blocks are the Qingtongxia -Guyuan fault, the Zhuanglanghe fault, the west Qinling fault, the Riyueshan fault as well as the Maqen fault, respectively. Among them, the crustal structures inside of the Ordos, Qilian as well as the part to the west of the Riyueshan fault are uniform and stable, relatively, and the crustal structure within the transition zone between the Ordos and Qilian block as well as that between the west Qinling and Riyueshan fault has obviously lateral variations and a complicated crust-mantle boundary. All of the boundary faults between the different blocks from the northeastern Tibetan Plateau to Ordos block have been extended to the Moho depth, suggesting that the interactions between the Tibetan Plateau and Ordos block reach to the upper mantle.(2)The seismicity within the North-South seismic zone is related closely to the crustal structure. Especially, the earthquake-prone area corresponds to the region with a compli- cated crustal structure. However, the domination of the crustal structure to the seismicity is different for different blocks. The seismicity in the Haiyuan earthquake region takes place within the high-velocity medium around the low-velocity medium, suggesting that the earthquakes in these regions are caused mainly by the local stress heterogeneity between high- and low-velocity media. The distribution of the earthquakes in the Qilian block is related mainly to the active faults, suggesting the seismicity in this area is caused mainly by the eastward compression of the Tibetan plateau to the Qinling faults.(3)The results by using the RF migration technique reveal the lateral variations of the 410km and 660km discontinuity of the upper mantle underneath the northeastern Tibetan Plateau and Ordos block. The depths of the 410km and 660km discontinuity are increased by 10km and 5km, respectively, underneath the western part of the Qsaidam block. This leads to a 5km thickness decrease of the upper mantle transition zone, suggesting a higher environment temperature under the western part of the profile than the eastern part. This observation verifies the inference about the crust and upper mantle structure according to the results by seismic tomography.(4)The RF inversion method can provide good results consistent with those by using the reflection/refraction method in the crustal structure studies. Due to that the Shear wave is much more sensitive to some specific structures than the P-wave, however, our results obtained using the RF method provide some important information in some regions (e.g. in the Qilianshan block), which is not given by the refraction/reflection method.3. Crust and upper mantle structure beneath the central Tien ShanFrom the waveform data recorded at 38 permanent and portable seismic stations distributed in the central Tien Shan area, we investigated the crustal Poisson's ratio, crustal thickness and the upper mantle discontinuities by using the RF grid-search method and the RF migration technique, respectively. Our results manifest that(1)The crustal thickness obtained by using the RF grid-search method is consistent with that obtained by using the RF inversion, suggesting that the RF grid-search method, as an independent technique, can be used well for estimating the crustal thickness. However, this method can be used for estimating the averaged crustal Poisson's ratio, which is hard to be obtained by using other methods.(2)The imaging of the upper mantle discontinuities by using the RF migration technique manifests that the imaging quality is dominated directly by the station interval and the data quantity. The increase of observational period is unable to improve the imaging quality. A high density of stations will be necessary not only from the theoretic point of view, but also in practice, if the high quality imaging of the upper mantle discontinuities is needed. The first Fresnel diffraction radius can be used for well estimating the station interval in the array observations.(3)The crustal thickness estimated using the RF grid-search method is consistent with that given by former studies in the same area. The crustal thickness reaches to 60km in the central Tien Shan, and it has variations from 46km to 55km in the Tarim basin and Kazakh Shield. The minimal crustal thickness reaches to 37.5km in the Naryn basin.(4)The averaged crustal Poisson's ratio is related closely to the tectonic uplift in the Tien Shan and adjacent regions and it has an obviously blocked distribution along the direction perpendicular to the Tien Shan range. The averaged crustal Poisson's ratio is larger at the north edge of the Tarim basin than that in the normal continental crust. A very low Poisson's ratio can be found in the southern Tien Shan. The Poisson's ratio near the normal value in the continent has been found in the region between the southern and northern Tien Shan. A Poisson's ratio obviously larger than the normal value has been found in the Kazakh Shield and northern Tien Shan. The abnormal low Poisson's ratio in the southern Tien Shan could indicate that its crustal compositions are different from those in other regions. It could be caused also by a higher content of void water or the hot crust and upper mantle. Meanwhile, the Poisson's ratios between the northern Tien Shan and the Kazakh Shield could indicate that the crustal rock compositions in this region are similar each other.(5)The image of the first-order discontinuities in the upper mantle along the NS-direction profile across the central Tien Shan by using the RF migration technique offer an observational proof about the thermal structure of the upper mantle underneath the central Tien Shan, which proves the low-velocity anomaly nearby the 410km discontinuity underneath the Kazakh shield (Chen et al., 1997). Meanwhile, the first-order discontinuities in the upper mantle along the mountain strike reveal that the upper mantle transition zone is thinner in the west part and thicker in the east part, suggesting that the upper mantle is warmer in the western part and cooler in the eastern part underneath the central Tien Shan.4. Crust and upper mantle structure underneath the Chinese Tien ShanIn terms of the RF grid-search method and migration technique, the stratified structures of the crust and upper mantle underneath the Chinese Tien Shan are investigated from the waveform data recorded by the movable seismic array, which is composed of 51 stations along the Kuqa-Kuytun. Our results manifest that:(1)Based on the results given by using the non-linear RF inversion method (Li, et al., 2007), the crust-mantle boundary is investigated in terms of the RF grid-search and migration method. Our results show a clear asymmetric crustal structure in the north and south side and further confirm the depth variation and dislocation structure of the Moho boundary along the profile, which was given by the RF inversion study.(2)The average crustal Poisson's ratio along the profile has been obtained in terms of the RF grid-search method. Our results show that the average crustal Poisson's ratio is lower than its normal value beneath the Tien Shan range. This implies that a higher content of silicon and quartz inside of the mountain range and a relatively weak crust. However, an obviously higher Poisson's ratio is found for the crust beneath the Junggar basin, suggesting that it has the obvious feature of the ocean crust. The relatively weak crust should be an intrinsic factor for the quick uplift of the Tien Shan under the external pressure.(3)Based on the results given by the non-linear RF inversion, the discontinuities within the crust are imaged using the RF migration technique. Our results also reveal the blocked structure and deformation of the Tien Shan crust along the profile, which are similar with those given by the RF inversion. Among them, there are obvious differences between The Tarim basin and south Tien Shan. The boundary between these two blocks should be located to the south of the frontal fault of the southern Tien Shan and to the north of Kuqa. According to the crustal velocity discontinuities and relocated event locations, it can be found that the northern marginal fault of the central Tien Shan is the most significant boundary fault separating the Tien Shan crust. The lower-crust structure is similar each to other underneath the northern Tien Shan and Junggar basin. The upper crust underneath the northern Tien Shan has a complicated nappe structure.(4)Taking the northern marginal fault of the central Tien Shan as the border, the spacial distribution of the faults that the main faults in the northern Tian Shan have a clear south dipping, while those in the southern Tian Shan have a clear north dipping, suggesting a clear bilateral compressional deformation of the mountain range. This implies that under the pressure of the Tarim block and the resistance of the Junggar block, the deformation of the Tien Shan range is increased from the surface to the deep part. This implies that the interaction between the Tien Shan and basins in both sides should not be confined only to the crust.(5)0ur results show that the Tarim block thrusts down to the southern Tien Shan at the top of the upper mantle beneath the Tarim and southern Tien Shan. The S-wave velocity structure of the crust and upper mantle underneath the Tien Shan does not provide any evidence about the partial melting within the crust. Therefore, our results manifest that the dynamic process of the mountain building should deal with the whole lithosphere. The Tien Shan orogenic dynamics should be coincident with the discontinuous compressional thrust model.(6)The imaging of the upper mantle discontinuities clearly shows the lateral variations of the 410 and 660 km discontinuity. Along the whole profile, the 410 km discontinuity is located at the depth of 400-410km. Underneath the central and northern Tien Shan, however, it becomes shallow. The depth of the 660km discontinuity varies between 660km and 680km and it becomes larger underneath the southern Tien Shan, suggesting a low-temperature anomaly dipping southward from the northern to southern Tien Shan, which reaches to 660km underneath the southern Tien Shan. Although this low-temperature anomaly covers the whole upper mantle transitional zone, its lateral size is less than 200km. This observation is consistent with the results given by the seismic tomography of the upper mantle and shows the possibility of the small-scale mantle convection underneath the Chinese Tien Shan.(7)In comparison with the Chinese and the central Tien Shan, it can be found that their dynamic process of the mountain building is quite different. Beneath the central Tien Shan to the east of the Talas-Fergana fault exists the warm upper mantle. The crust in its southern side also has a higher temperature. However, there is no evidence of the thermal anomaly in the crust underneath Chinese Tien Shan and it has a strong vertical divergent movement between different blocks. Therefore, the single or bilateral subduction may play the main role in the quick uplift of the Chinese Tien Shan.
|
PDF Full Text Request