A Multi-Disciplinary Workflow to Achieve the Largest, Seamless, High-Quality Presalt Image in Santos Basin, Brazil
Abstract
The Industry’s largest, seamless, high-quality Reverse Time Migration (RTM) seismic image over a
massive 34000 sq. km of Santos Basin, Brazil, has been produced by applying advanced imaging
techniques on various vintages of data. With the ever-shrinking turnaround time requirements for
seismic processing, having an effective cross-disciplinary cooperation is necessary to efficiently and
accurately image these large datasets. Understanding the response of a seismic image to velocity or
geological models, or imaging algorithms is critical for imaging the reservoir targets accurately.
Furthermore, minimizing the uncertainty associated with the resulting image is a prerequisite.
Additional challenges involved with the processing of large datasets are the High Performance
Computing (HPC) and Quality Control (QC) requirements. To ensure production of quality images in a
timely manner, real time analytics are needed to manage HPC throughput, disk utilization, and QC of
seismic images. This results in the need for workflows across multi-disciplinary teams efficiently
contributing to and creating the required seismic images. The use of such a workflow on a large seismic
dataset over the Santos Basin is outlined here.
Reservoir targets in the Santos Basin are mainly localized in presalt carbonate build-ups underneath the
Layered Evaporite Sequences (LES). A geological understanding of LES complexity and the differing
character of the post-salt carbonates layers, along with available well data, must be incorporated into
the geophysical velocity model building flow to produce a clear presalt image with a consistent reservoir
geometry. The decision-making process benefits from efficiently analyzing a large number of volumetric
datasets, generated for QC and constrained by geological and geophysical metrics. The goal throughout
the workflow was to use all available tools in a manner that promotes efficiency, improves quality and
reduces turnaround time. For example, to reduce the time required for an enormous amount of Base of
Salt (BOS) interpretation, a workflow using Full Waveform Inversion (FWI) was developed for a data-
driven, computer-based approach to salt velocity model building. The result of this multi-disciplinary
approach is evident in a very large-scale seismic dataset with enhanced presalt imaging, which can be
used for reservoir architecture analysis and exploration risks mitigation.
AAPG Datapages/Search and Discovery Article #90350 © 2019 AAPG Annual Convention and Exhibition, San Antonio, Texas, May 19-22, 2019