3D Methodology for OBC Pre-Processing

It is more and more observed that OBC PZ images show better quality than the streamer. This is due to better multiple attenuation, multi-azimuth and high fold coverage. This leads to a wider bandwidth and a better signal to noise ratio. However, while 3D OBC designs are considered as multi-2D acquisitions, the pre-processing will remain far from the ideals. In the following paper, a fully 3D OBC pre-processing methodology is presented. This methodology is based on the 3D(τ,p) transform. Due to the fact that a spatial transform is used, aliasing becomes a real issue. The interpolation solution to overcome such a problem is briefly described. Then, the generalization of the PZ summation to 3D is discussed and compared to the standard 1D approach. Finally, a method for linear noise attenuation is described and its application on a real dataset shown.