Shallow Water Demultiple Using Hybrid Multichannel Prediction

In this paper, we present a hybrid approach combining the strengths of Shallow Water Demultiple (SWD) and model-based methods. We show that the multichannel prediction operator estimated by SWD can be used as an accurate kinematic representation of the seafloor reflection with a higher signal-to-noise (S/N) ratio, allowing the travel time of the seafloor to be automatically estimated. With this information the Green’s function of water-layer primary reflections can be modeled, and form a hybrid operator such that it compensates the missing high dips of the multichannel prediction operator and contains more than the seafloor event as represented by the Green’s function only. The hybrid operator is then used for multiple prediction and removal. We first present the theory of the method and then demonstrate its effectiveness on synthetics and field data acquired offshore Australia.