Multicomponent Seismic Data Antialiasing Interpolation and Its Application for Sparse OBN Data Processing

     During the acquisition of seismic data offshore, due to the complex seabed terrain, ocean currents, and the costs of instruments, the multicomponent seismic data recorded by OBS or OBN are usually sparsely and irregularly sampled. Sparse sampling and irregular missing seismic traces generate severe spatial aliasing, which disturbs subsequent seismic migration processing. Furthermore, the data collected by both OBS and OBN are multicomponent in nature. Regarding the reconstruction of multicomponent data, the current methods are mainly scalar-based, treating the multicomponent data as several independent components and interpolating each component separately. These component-wise approaches ignore the internal mutual relationships among the different components, thereby damaging the vector-field nature of the seismic elastic wavefield. Given this, we propose a vector Project onto Convex Sets (POCS) reconstruction method based on the complexified quaternion Fourier transform, which achieves joint vector reconstruction of the three-component (3C) OBN data. This proposed method not only reconstructs data for the three components with their respective missing patterns, but also preserves the vector polarization characteristics of the subsurface particles.

     For sparse 3C OBN sampling data, we propose a new vector anti-aliasing POCS interpolation method based on a dip angle scanning strategy. There are two innovative points for this method: Firstly, we utilize the first L maximum values of the negative second derivative of the dip scanning energy spectrum to pinpoint the position of effective wave dips, enhancing the accuracy of dip identification. Secondly, we adopt a 2D Gaussian tapered window function instead of the tradational 2D rectangular tapered window function to mitigate the Gibbs oscillation phenomenon and suppress the energy tailing effect at the edges of recovered seismic events. Finally, several sparse OBN field data reconstruction test results demonstrate the effectiveness of the proposed anti-aliasing vector POCS reconstruction method.