GEOPHONE NOISE ATTENUATION IN THE COMPLEX WAVELET DOMAIN FOR WAVEFIELD SEPARATION OF OCEAN BOTTOM SEISMIC DATA

For ocean bottom seismic data processing, hydrophone and vertical geophone are combined to separate wavefields into upgoing and downgoing components. However, vertical geophone usually records strong shear noise (also known as V-z noise), which can significantly contaminate the results of wavefield separation and degrade image quality. In this paper, we propose a workflow to attenuate geophone noise for wavefield separation of ocean bottom seismic data. The workflow includes three steps: (1) calibration of geophone data against hydrophone data; (2) combination of calibrated geophone and hydrophone data to perform initial decomposition of upgoing and downgoing waves; (3) using hydrophone data and the downgoing wavefields from initial decomposition to perform enhanced wavefield separation in the dual-tree complex wavelet domain. The enhanced wavefield separation is achieved by thresholding the amplitude ratio of the downgoing wavefields from initial decomposition and hydrophone data in the complex wavelet domain. The proposed method is demonstrated on both synthetic and field data.