Preconditioning input data for full waveform inversion using dipole sparse tau-p inversion

Low-frequency seismic data are crucial for the success of full waveform inversion (FWI) used in velocity model building. Marine acquisition technologies have evolved to increase low-frequency seismic signals. However, significant low-frequency noise, often caused by water flowing across the streamers, still alters the signal-to-noise ratio. We propose to use an algorithm based on a dipole sparse tau-p inversion to attenuate low-frequency and high-dip noise. When applied to synthetic and real data sets, this method provides clean low-frequency input data to enhance the performance of FWI.