Optimal stacking by utilizing pre-migration deghosting technique

摘要

Conventionally acquired marine data is composed of both primary and ghost components; when using the traditional stacking flow, the data, therefore, suffers from destructive interference. With the development of new pre-migration deghosting techniques, we have an opportunity to optimize the traditional flow in hopes of better attenuating incoherent noise and migration artifacts. Our idea involves using pre-migration deghosting to segregate the recorded data into four separate data sets: primary, source ghost, receiver ghost, and source-receiver ghost data sets. Because the four individual data sets avoid the destructive interference, they each contain improved low- and high-frequency responses, as well as improved signal-to-noise ratios (S/N) over the input data. Furthermore, they individually produce consistent migration images. By selectively stacking these four migration images in a coherent fashion, we produce final stack images which have wider bandwidth, sharper wavelets, and fewer migration artifacts compared to the conventional stacking flow. We Validate this optimal stacking flow using both Reverse Time Migration (RTM) and Kirchhoff migration for a narrow azimuth towed streamer data set from offshore Brazil.