Study&Application On The Common-Reflection-Surface Stack

Common-Reflection-Surface Stack was first proposed by Karlsruhe University ProfessorHubral in1990s, which is a special kind of zero-offset (ZO), or common offset (CO) imagingmethod. Besides, this theory has two features that it is dependent with the macro-speed modeland it fully implemented by data-driven way. Using CRS stack, not only a high-qualitystacked profile map, but also three groups useful parameters of the wave field can be obtained.It is the concept of the reflective surface introduced into CRS theory and the contribution tothe reflected wave made by all the scattering points within the first Fresnel zone on thereflective surface which can help to enhance the signal-to-noise ratio of the stacked sectionand maintain a certain resolution by increasing folds. So CRS stack is more adapted to theseismic data with low signal-to-noise ratio and superimposed times. Compared with theconventional stack method, CRS stack neither needs statics corrections for prestack data, nora necessary macro-speed model.Based on studying and analysing the theory of CRS stack and correlative literature aboutnew technique of CRS stack in this paper, the author designed and completed the followingthree programmes independently: conventional CRS stack, CRS stack based on dipdecomposition method and confined exit angle filtering based CRS stack. The author verifythe correctness and practicality of this method by testing model data and the actual data. Onthis basis, the conclusion that the superposition aperture is the core parameter in obtaining ahigh quality stacked profile map can be made. And it only has theoretical significance that ifthe radius of the first Fresnel zone was chosen as CRS stack aperture. In the selection of stackaperture, the observation system, geological environment and signal to noise ratio of seismicdata in each time period must be taken into account.Though a high signal-to-noise ratio stacked profile map can be acquired through theconventional CRS stack, because of the discrimination of inclination, there is an obviousdislocation in the cross point of reflection events in stack section, resulting in that deepdiffraction can not be protected effectively. As for this problem, the CRS stack algorithmbased on dip decomposition proposed in the paper has solved the problem of dipdiscrimination and improved the corresponding stack section. In addition, based on the consideration of enhancing the information of reflected wave from some special target, aprocess of filtering the angle of departure based on dip decomposition is applied to stacksection to enhance the reflected signal of the angle of departure from the targets. Though acertain effect can be obtained through the above processing in the model testing, because theconstrained force against exit angle from correlation spectrum is low, to further improve theprogram, ultimately realize the method of confined exit angle filtering.