Joint Exploration Of Deep Seismic Reflection And Deep Seismic Sounding With Its Application In Crust Structure Research

Structural difference, chemical and physical changes of the crust and interaction with mantle profoundly affect surface geology, geomorphologic processes and human life. Thus, research on crustal structure and evolution always focus geoscientists’attention. Two active seismic methods:deep seismic sounding and deep seismic reflection, which provided detail structure information of crust and uppermost mantle, are important channels for explorating of reservoir, mineral resources and understanding of the earthquakes, volcanic activity and geodynamic processes. The former could provide reliable crustal velocity structure, and the latter could provide fine crustal structural configuration. Based on a summary of research progress of domestic and international, this thesis investigates some problems about joint exploration of deep seismic sounding and deep seismic reflection with its application in crust structure research. And its research contents could be departed into3portions below:1、 Structure information recognition of deep seismic reflection profilesBecause of the attenuation effect of long distance propagation and complex geologic body in deep crust, the seismic events from the middle and lower crust in deep seismic reflection section are always displayed as energy weak, discontinuous, zonary or interlaced. These make the interpretation became difficult. In the second portion, we summarized pattern recognition method, artificial neural network method and x2distribution method, and developed two new methods to recognize structure information from deep seismic reflection profiles. With considering the character of deep seismic reflection data and problems encountered in interpretation, the first method is based on digital image processing theory. First order differential Sobel edge detection operator was employed to extract the reflected framework of the section. According to the histogram of amplitude distribution, density slicing algorithm was used to acquire false color image which improved the sections’display effect. The second method based on skeletonization technique and digital image processing theory is composed of the following module:data pretreatment, amplitude analysis, median filtering, object recognition, continuity calculating and continuity filtering. We save the objects’identification number, number of elements, elements’position, object length, angle and other information in a relational database. Besides to recognize structure information from deep seismic profiles, this method is also used to analyze the seismic events angle distribution of structure complex area. This method gets better recognized effect than the image processing method. Because we removed waveform characterization and iterative steps, this method is more efficient than the pattern recognition method. 2、 Crustal structure imaging of deep seismic sounding and joint exploration methods with deep seismic reflection data.The first portion reviewed basic theory of first arrivals finite difference tomography and seismic travel time inversion methods. Several noticeable problems and modeling strategies during processing and interpretation were discussed. We considered that:according to the convergence of travel time RMS error, variable parameters utilised from coarse to fine will effectively avoid appearance of local minima during topographic inversion. Forward modeling with the traditional "trial-and-error" way is necessary. When travel time RMS error reaches a certain standard, then inversion using damped least squares method could prevents generation of artificial information.If the deep seismic sounding data was collected along the same profile with deep seismic reflection, their non-uniqueness will be constrained by two different kinds of data through reciprocal constraint processing or joint inversion. The third Chapter introduces two joint exploration methods with wide angle reflection\refraction data and near vertical reflection data. In case both data precision are close, merging the reflection velocity analysis result and wide angle reflection\refraction velocity structure will improve the limitation of near vertical reflection velocity analysis, and optimize the depth migrated section. In addition, picking zero-offset travel time from two way travel time stacked reflection profile and joint ray tracing inversion with wide-angle reflection and refraction data will obtain more reliable velocity structure. If both data’s precision and anisotropy are different, we can use information provided by coincident deep seismic reflection section profiles, such as interface morphology, surface velocity etc., as priori information to constrain the deep seismic sounding modeling.3、 Application in crustal structure research of the northern margin of the north China craton and the southwestern margin of the Ordos block.Following the constraint modeling approach, Hole’s first arrival tomography program was used to obtain the upper crustal velocity structure and Zelt’s2D travel time ray tracing and inversion program was used to obtain the full crustal structure of the northern margin of the north China craton Huailai-Sonid Youqi profile and the northeastern margin of the Tibet plateau Ordos-Liupanshan profile.2D P-wave velocity modeling was done layer by layer using the top to bottom approach. The velocity model was altered by trial and error, and the forward model was updated by damped least-squares inversion. Finally, we draw the following conclusions respectively: The flat and relatively shallow Moho (-40km) of the central Asian orogenic belt may be attributed to the extension; thicker crust appears beneath the Yinshan-Yanshan belt and was probably generated by compression in the Jurassic-Cretaceous and modified during the latter extension. Relatively high Velocities in the upper crust of the middle port of the profile may represent the outcrop of large area granites. Strong velocity variations beneath the Bainaimiao arc and Ondor Sum subduction accretion complex indicate that multiple pulses of magmatism occurred during the complex tectonic evolution of this area. The velocity structure varies significantly from the north China craton to the central Asian orogenic belt, and the boundary between them appears at the Chifeng-Bayan Obo fault.The upper crust of the Ordos basin has relatively lower velocities, and the velocity contour is sub-horizontal which represents stable sediments. The upper crust of the Qilian orogen has strong velocity variation. It should be a result of northeastward compression of the Tibet plateau during the Cenozoic. Different characteristics in the first arrival topographic model indicated that the Haiyuan-Liupanshan fault is the boundary between the Ordos basin and the Qilian orogen. The Qilian orogen has relatively lower velocity in the middle and lower crust, while the Ordos block has higher velocity and relatively simple crust which indicated it is a stable old craton. The crustal thickness varies from～50km in the west to～42km in the east, and double Moho appears beneath the Liupanshan area. It suggests that the thin and rigid Ordos crust was pushed into the thick and soft Qilian crust with the far field effect of the collision between the Indian plate and the Eurasian plate. Transverse compression leads to all layers above Moho upswept in this region. We inferred that strong earthquake activity should be related with the upswept rigid crust’s elastic rebound and release of stress.