Multiple Suppression and Imaging of Marine Seismic Data from The Shallow Water Area in Southern East China Sea Shelf Basin

Abstract:

 Neither surface-related multiple elimination(SRME) nor predictive de-convolution method is effective to suppress the multiple of marine seismic data from the shallow water area. The former method needs the accurate reflection of seafloor, which is mixed with the direct wave in the near offset range. The other one could probably lose the primary wave when applied to the shallow water seismic data. We introduced the new method: deterministic water-layer de-multiple method (DWD)，which is capable for the poor extrapolate result of near-offset traces. Firstly, the data shifts as downward continuation in tau-p domain with a water-layer period and the multiple model will be obtained. Then, the original seismic subtracts adaptively with the multiple model. Finally, we would get the de-multiple data after inverse tau-p transform. Marine seismic real data is from southern part of East China Sea Shelf Basin. This area has become the potential target for marine hydrocarbon exploration, it is located in the junction of the Eurasian plate，pacific plate and Indian plate. Because the average water depth is less than 100 meters, seismic data contains abundant of multiple, especially the surface-related multiple. As a result，it is difficult to distinguish the strata structure clearly. We used DWD approach to remove the water-layer multiple, cut off the seafloor reflection events and then suppressed the residual surface-related multiple by the traditional SRME. At last , the radon transform was applied to eliminate the multiple with long period . With these steps, we suppressed the multiple of marine seismic data from this area effectively. After multiple is removed , we acquired more accurate velocity to build the velocity model of migration. With the pre-stack migration technique, reflections from each geological period are shown clearly in the seismic section.

This work was supported by the National Science Foundation of China(grant no. 41476053).