Finite-frequency Fresnel Volume Tomography And Application

Seismic exploration has always been one of the most important technical methods of existing geophysical methods and plays an important role in petroleum exploration. The exploration of oil and gas in plain areas in China has been deeply developed and land exploration has turned to the northeast, southeast, and other regions. The main problems encountered in these regions are the dramatically changing landforms and low signal-to-noise ratio of seismic data. The traditional ray tomography technology assumes seismic waves as infinite frequency zero-volume rays, which inevitably leads to sampling rays along the high speed areas. Thus, the accuracy of inversion is not satisfying, which would cause it impossible to restore the true structures of these regions. Wave equation tomography is dependent on the initial model, seismic wavelet, data quality, and other harsh conditions. In addition, the requirement of storage and computation capacity for wave equation tomography is very high, which makes it impossible to apply it on practical data. The theory of finite frequency tomography takes into account of the characteristics of seismic wave band and the effects caused by the media of the first Fresnel zones to the observation information. Therefore, its inversion accuracy is higher, which is of importance to study this theory.In this paper, the development history of the seismic tomography technology is briefly introduced. The generation, development, and the research status both at home and abroad of the finite frequency theory are introduced. In the study of the finite frequency Fresnel volume tomography technology, a high accurate and efficient method of forward modeling is required. The forward modeling method adopted in this paper includes wave front travel-time calculation, ray tracing, and finite frequency Fresnel volume travel-time computation. First, the wave front travel-time of the entire model must be confirmed, and then the ray path is determined by the positions of the receiving points and the known wave front travel-time. After the ray path is known, the travel-time is then calculated according to the sensitivity function of the kernel of the network nodes. The calculation of the sensitivity function is key to the finite frequency Fresnel volume tomography.Starting with the wave equation of Born and Rytov approximation, the calculation of finite frequency sensitivity kernel was derived in detail. According to the travel-time, ray path, and sensitivity kernel function calculated, we establish the inversion equation by using the back projection algorithm (ART) and the simultaneous iterative reconstruction technique (BRT) and then solve the equation. The sensitivity kernel function of velocity is an important part of this paper. After the theory research, the approximate Fresnel volume tomography and the finite frequency Fresnel volume tomography’s sensitivity kernel are compared through three different models. After completing the theoretical research, due to some problems encountered in the pick-up process of the actual data and to the rough terrain of the fields, the initial data preprocessing, modeling, and terrain digitalization are studied in detail in the fourth chapter. The imaging effect and the applicability of the finite frequency Fresnel volume tomography are tested and compared with that of the approximate Fresnel volume tomography inversion. Finally, real seismic data are processed and stack section is obtained based on the corresponding inversion result.A summary of some acknowledgements of the research of the finite frequency Fresnel volume tomography and suggestions on future works are given at the end of this thesis.