American Oil and Gas Reporter - October 2016 - 60

SpecialReport: Gulf of Mexico Update
approximately 15,000 foot-thick canopy
over the structure. Below the salt canopy
is a 9,000 foot-thick section (from the
base of salt to the base of the reservoir
sands) that is composed of deepwater
shale and turbidite sand packages in the
upper/middle Miocene reservoir. As shown

in the dual-coil seismic image in Figure
2, the three key pay intervals in the
faulted, four-way structure are the middle
Miocene (Serravallian) Gff10s, Gff40s
and Sh20s.
Four wells with several sidetracks
were drilled to discover and appraise the

FIGURE 3
Multidisciplinary Workflow for Building the Simulation Model
Structural Framework

Grid Design

Compartments
Log Data

EOD Map

Reservoir
Property
Population

Training Image

Analysis

STOIIP +
Simulation

PHIE

MPS-Facies

FIGURE 4
Current Depth Structure Map of Gff40s Reservoir Zone
Using Dual-Coil 3-D Data

C.I - 200'
0

2,000 4,000 6,000 8,000 10,000

60 THE AMERICAN OIL & GAS REPORTER

20,000 feet

field, but none of the wells penetrated
oil/water contact. Multiple fluid contacts
were inferred from pressure data in the
wells, implying that reservoir compartmentalization was difficult to support
with faults mapped from existing seismic
data.
The low-resolution seismic data (including both narrow- and wide-azimuth
3-D) also made it difficult to discern
stratigraphic changes that could create
flow barriers or compartmentalization.
As a result, engineered faults from previous
geological models were used to explain
pressure differences between wells.
However, with the latest 3-D dualcoil seismic data, more small offset faults
could be mapped, which helped explain
some of the pressure differences observed.
Fault seal analysis was performed to understand the flow barrier capability of
faults, and this information was incorporated into the simulation model.
Filling the geological model with reservoir facies had its own challenges; the
size and extent of the depositional fill
and rock types in the fill were problematic.
In addition, a process was needed to populate the geological model representing
the fill, which also proved challenging.
The depositional setting for the reservoir pay zones varied from unconfined
to weakly confined deepwater channel
systems in a salt withdrawal basin. With
the help of outcrop analogs, we interrogated seismic data to define the depositional environment boundaries. Despite
several penetrations in the reservoir zones,
there still was uncertainty in facies typing
and fluid contacts. Consequently, core
data, logs, fluid parameters and regional
analogs were used as part of a multidisciplinary approach to reduce facies type
and fluid contact uncertainties.
This process of integrating data and
information from a variety of sources,
tools and methods in the geological and
flow simulation model greatly increased
our understanding of the geology and
petrophysics behind the workflow.

American Oil and Gas Reporter - October 2016

Table of Contents for the Digital Edition of American Oil and Gas Reporter - October 2016