SIMULTANEOUS SEPARATION, INTERPOLATION AND TUBE WAVE SUPPRESSION OF VERTICAL SEISMIC PROFILING USING MATCHING PURSUIT BASED SPARSE BEAM FORMING

摘要

Since the vertical seismic profiling acquires seismic data with high resolution and superior S/N ratio, it is essential for reservoir monitoring in seismic exploration. The vertical seismic profiling data is usually separated into down-going and up-going wavefields in its processing procedure. The wavefield separation is mainly achieved by frequency-wavenumber method and plane wave decomposition method. However, when the vertical seismic profiling data is aliased, these methods perform poorly. Besides, to preserve the high frequency component during the wavefield separation, the data interpolation is required to avoid the aliasing effect of the signal. Furthermore, there is usually strongly aliased tube wave noise in the vertical seismic profiling data which poses challenge for the wavefield separation and subsequent processing steps. In this paper, we propose a solution using matching pursuit based sparse beam forming and it can simultaneously separate the wavefield, interpolate the data and suppress the tube wave noise. The matching pursuit based sparse beam forming is guided by the un-aliased information of the data. 'Thus, it can handle the aliasing effect of both signal and tube wave noise. Since the sparse beam forming method obtains beam forming result with superior resolution of the ray parameter, it can separate the wavefield and suppress the tube wave noise based on their differences in the ray parameter domain. Then the filtered data can be backward transformed to a denser grid for a better preservation of the high frequency component. Numerical examples on both synthetic and real data show that the proposed method works well in terms of wavefield separation, tube wave suppression and data interpolation.