Key points are not available for this paper at this time.
Modern marine-streamer time-lapse (4D) seismic projects generally use broadband processing flows that compensate source and receiver ghost effects. They are mostly very successful, but it can occasionally be difficult to achieve an accurate match between the surveys at low frequencies. A test project using data from the Heidrun field, offshore Norway, suggested that broadband processing can exacerbate minor issues that are typically insignificant for non-broadband flows. Seismic surveys acquired with multisensor marine streamer systems tend to undergo wavefield separation and other processing during acquisition, with different acquisition contractors using differing workflows. This can cause discrepancies between the surveys that the 4D processing contractor cannot compensate. Similarly, ancillary datasets, such as farfield signatures, may be generated inconsistently. We recommend that time-lapse seismic processing begin with raw measurements wherever possible so that identical processing flows may be used for each survey. Even then, it may not be possible to completely compensate the impact of differing acquisition systems.
Smith et al. (Mon,) studied this question.