Solving fault-related imaging issues: a case study comparing conventional processing and multiparameter FWI imaging

Seismic imaging of faulted reservoirs presents a challenging task. Faults cause diffractions which need to be preserved, but their multiples (diffracted multiples) must be attenuated without damaging primaries to avoid a noisy image. Additionally, conventional tomography may not adequately model the complex velocity structure caused by faulting. These fault-related imaging issues were addressed in an offshore Western Australian field covered by a narrow-azimuth streamer dataset, using two strategies. First, conventional seismic data reprocessing was performed with careful noise reduction, and fault-guided tomography. The noise reduction process included detailed surface-related multiple elimination (SRME) model generation and mild pre-migration Radon demultiple. Fault-guided tomography was performed using interpreted faults as a guide for tomography velocity updates. The results showed significant improvements over existing seismic images. Second, a test was performed to leverage multiparameter full waveform inversion (FWI)’s use of the full unprocessed wavefield, including refractions, multiples and intact diffractions, to derive a velocity model and FWI-reflectivity for seismic imaging of faulted reservoirs. The results of multiparameter FWI imaging were comparable, if not superior, to the results from the conventional processing using tomography-based velocity and migration, achieving comparable results in a shorter timeframe. These studies suggest a potential for multiparameter FWI imaging as an alternative, quicker turnaround time solution to conventional processing, for imaging faulted reservoirs accurately.