Abstract:

The elimination of multiple wave is  a key issue in seismic data processing and interpretation, and will directly affects the quality of seismic data processing. Surface-related multiple wave elimination(SRME)theory discussed is a data-driven concept, and the multiple wave can be predicted more precisely than traditional method by the convolution of seismic data without the knowledge or assumption about the subsurface. Facing more and more 3D marine seismic exploration data, traditional 2D SRME exposes its disadvantages in algorithm, it is necessary to extend 2D SRME to 3D. Compared with traditional 2D SRME, 3D SRME considered the multiple wave contribution more comprehensively, calculated each of the multiple wave contribution more efficiently, predicted multiple waves more precisely. We estabished 3D inclined layer velocity model and simulated seismic data contained multiples in order to use 3D SRME technology to predict multiple waves. Comparing the results predicted by 3D and 2D,we found that the predicted result of 3D SRME is more consistent with the theoretical multiple position. Finally, we used the least square matching to achieve the elimination of multiples successfully and suppressed the multiple information effectively.

Key words: multiple wave, 3D SRME, multiples prediction, data-driven