American Oil and Gas Reporter - January 2016 - 64

FIGURE 3
Petrophysical Example

12,800

12,850

12,900

FIGURE 4
DeWitt Facies Maps and Arbitrary Cross-Sections
Across Lavaca and DeWitt Models
Facies Key:
Organic Marl
Bioclastic Marl
Bioturbated Marl
Clayey Mudstone
Recrystallized Limes
Chalk

DeWitt Dip Line

64 THE AMERICAN OIL & GAS REPORTER

Lavaca Dip Line

expulsion occurs during core retrieval in
low-permeability, overpressured reservoirs.
In addition, some swelling of the core
plugs may have occurred in the laboratory.
Generally speaking, computed porosity
mimicked the trend seen in bulk density,
resulting in lower computed porosity values than measured in core.
Water saturation was defined as bulk
volume water (BVW)/porosity. The model
was calibrated by adjusting saturation
parameters and aligning the resulting
BVW to core measured BVW.
This process then was executed iteratively to derive the best possible mineral
estimation, while matching TOC, porosity,
BVW, and resulting water saturation. An
example of the petrophysical output calibrated to core is presented in Figure 3.
Track 1 shows gamma ray (green)
and caliper (red) data. The green shading
in track 3 is TOC from petrophysical
analysis, whereas the black dots are TOC
from core. Similarly, the solid line in
track 4 is log RHOB, whereas the black
dots are core RHOB. In track 5, dark
gray with patterns denotes clay, yellow
is quartz minerals, light blue is limestone,
light gray is pyrite, black is kerogen, and
white is porosity. The solid line is petrophysical analysis dry weight clay, and
the black dots are core x-ray diffraction
dry weight clay. In track 6, the light blue
curve is petrophysical analysis water saturation, light blue dots are core water
saturation, red curve/dots are porosity,
and dark blue curve/dots are BVW.
In the process of calibrating and developing the petrophysical model, it was
observed that formation water resistivity
(Rw) was not constant across the modeled
area. Two Rw trends were discovered,
and the transition between them appears
abrupt. We believe influence from clastic
Woodbine sedimentation from the northeast is the main factor in the Rw change.
The areas where the different trends were
utilized are geographically consistent with
this interpretation.
Modeling Results
In both models, core facies descriptions
drove neural net facies training and estimation, and variograms by facies per unit
(66 total variograms per model property)
drove the geostatistical mapping. The
neural net training set varied in accordance
with paleogeography and core availability.
The estimated facies logs were used as a
bias filter for upscaling the petrophysical
property logs and as a conditioning filter
for data analysis, variograms and petrophysical property simulations. Figure 4
shows a map view of a surface within the
Eagle Ford in the Dewitt model, and
cross-section views of the facies models
American Oil and Gas Reporter - January 2016

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