American Oil and Gas Reporter - September 2016 - 48

SpecialReport: Reservoir Optimization & Modeling
Summary volumes of P-impedance, Vp/Vs, density, Young's
modulus and Poisson's ratio, as well as probability volumes for
porosity, individual and most probable lithofacies, were generated. These volumes were tested against blind wells and subsequently used for interpretation and well planning.
Figure 3 displays a cross-section flattened to the top of the UpFIGURE 4
S5 Geobody in 3-D Reservoir View

FIGURE 5
Thickness Maps Overlaid with Structure Contours
(Top of Upper Cretaceous Sand/Seismic
Facies S4 and S5)

per Cretaceous Sand horizon. It compares the level of detail seen
in the well logs with the original seismic dataset, and shows results from deterministic inversion along with results from geostatistical inversion. Figure 3A shows the lateral variability of the
well lithofacies (overlaid on seismic), and how limited seismic
resolution was over the target zone.
Figure 3B shows the well lithofacies overlaid on the P-impedance derived from deterministic inversion. P-impedance is
smeared, showing only relative lateral and vertical interwell
changes for the tight sands, which are represented by high P-impedance and are displayed in yellow. It is evident that vertical resolution still is limited.
Figure 3C shows filtered well P-impedance logs overlaid on
the P-impedance derived from geostatistical inversion, while Figure 3D shows the well log lithofacies overlaid on the lithofacies
volume derived from geostatistical inversion. The characterization of the reservoir produced by geostatistical inversion provided a highly detailed, seismically constrained and accurate subsurface model calibrated to well control.
The well data correlated well with the seismically derived subsurface geomodel for both lithofacies and associated reservoir properties, and the results provided a reliable indicator of interwell
subsurface reservoir conditions.
Again, the subsurface model honored all well data, including
wells subsequently used as blind tests, verifying the value of the
model as an accurate and predictive tool for reservoir development and well planning.
Stratal slices and cross-section views of the five lithofacies (seismic facies) were generated from the geostatistical inversion results. The cross-sections and stratal slices describe the variability of the heterogeneous reservoir and appear geological in nature. The analysis showed that reservoir quality increases from
seismic facies S1 to S5, with the S4 and S5 facies as the pay reservoirs.
In addition, effective porosity was generated by geostatistical simulation using the inversion outputs as secondary trends (cosimulation). The 3-D elastic property and lithologic volumes produced highly detailed 3-D models of effective porosity, exploiting the relationships between the reservoir and elastic properties.
Reservoir Characterization

FIGURE 6
Lithofacies Volume from Geostatistical Inversion
And Well Treatment Plots
Zones of average reservoir quality
require higher frac pressure

Zones of better reservoir quality
require less frac pressure

80 ft
2

2
Breakdown pressure
Stage 29: 7,500 psi

Breakdown pressure
Stage 28: 6,000 psi

Breakdown pressure
Stage 27: 5,500 psi
1`

48 THE AMERICAN OIL & GAS REPORTER

Increasing Reservoir Quality

A series of highly detailed lithologic and elastic rock property 3-D volumes were created through a Markov chain Monte
Carlo and Bayesian inference method to systematically incorporate various sources of prior information in an unbiased, rigorous and consistent manner. The final synergized datasets accurately characterized the 3-D reservoir distribution as tested against
blind wells.
The 3-D reservoir geobody in Figure 4 shows lithofacies S5
(displayed in red) for the Upper Cretaceous Sand where the probability of encountering the S5 facies type is between 50 and 100
percent. The percentage probability cutoff can be investigated easily by changing the cutoff.
The three panels in Figure 5 display thickness maps overlaid
with structure contours for the top Upper Cretaceous Sand (seismic facies S4 and S5), including from left to right, gross thickness, probability thickness and porosity-thickness (Phi-H).
A horizontal well penetration relative to the seismic-derived

American Oil and Gas Reporter - September 2016

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