Multiple attenuation for variable‐depth streamer data: From deep to shallow water

Abstract

Variable-depth streamer acquisition is becoming a key technique for providing wide bandwidth seismic data. Varying the receiver depth creates wide receiver ghost diversity and produces a spectacular increase in the frequency bandwidth. However, compared to conventional data, this variable-depth streamer data implies a major challenge in processing: how to deal with various receiver ghosts. The ghosts have to be preserved up to the deghosting step. Here we present the implication for the following de-multiple methods: Shallow-Water Demultiple, Tau-P deconvolution and Surface-related multiples elimination in deep and shallow water environments.

Challenges of Pre-Salt Imaging in Brazil’s Santos Basin - A Case Study on a Variable-Depth Streamer Data Set

“…By convolving traces with different wavelets, the traditional SRME method produces multiple models substantially different from the input. Sablon et al (2011) presented an adjusted 3D SRME procedure for variable-depth streamer data. The workflow involves deghosting the input data, and subsequently re-ghosting the "ghost free" multiple model.…”

Velocity model building for variable-depth streamer data: Brunei case study

“…Our workflow contains the following stages:  special pre-conditioning of data by applying denoise and demultiple (Sablon et al, 2011) tools  high-density RMO joint picking on migrated and mirror-migrated gathers using cross-semblance criteria  dip picking and skeleton creation on deghosted migrated stack (Soubaras, 2010) Figure 1: RMO picking of migrated and mirror-migrated gathers. The semblance panel corresponding to the deghosted gather (Soubaras, 2010) is shown on the bottom, while the semblance panel computed for the symmetrized gather is indicated on top.…”

“…We have illustrated the capability of our proposed highdensity/high-order cross-semblance RMO picking algorithm before pre-stack CIGs deghosting by applying the new methodology to a variable-depth streamer 2D dataset from Brunei. Dedicated denoising and SRME demultiple processes were applied to the data prior to RMO picking to increase the signal-to-noise ratio (Sablon et al, 2011). Figure 2 shows that only the primary is picked as the red RMO curves follow the primaries rather than the ghost.…”

“…Since an accurate deghosting cannot be done at an early step, it is performed through a joint deconvolution of a migration and a mirror migration (Soubaras, 2010). Several steps of the processing sequence have been already published: namely, multiple removal (Sablon et al, 2011) and 4D processing (Charrier et al, 2012). In this paper, we will address the complex issue of velocity model building (Depagne et al, 2012).…”