American Oil and Gas Reporter - January 2015 - 169

indicates a high degree of variability in
well performance.
Identifying Correlations
Two methods were applied to investigate relationships between the modelderived parameters and reservoir/well
completion properties. The first method
investigated the existence of a linear relationship by computing the correlation
coefficient between the variables (a value
of +1 implies a strong positive linear relationship, -1 indicates a strong negative
linear relationship, and 0 implies no linear
relationship between two variables). The
second method investigated nonlinear relationships by making cross plots of
model parameters (including transforms)
and reservoir/completion properties, and
then evaluating recognizable functional
relationships.
Table 2 summarizes the computed correlation coefficients for the parallel-flow
model and reservoir/completion properties,
with the shaded areas denoting values
greater than 0.2. The initial production
rates show some level of correlation with
most reservoir and completion properties.
The time constants do not show the same
level of correlation.
Table 3 shows the same information
for the logistic-growth model. Based on
these criteria, the carrying capacity possibly has a linear relationship with well
spacing, porosity, average injection pressure, total fluid injected, and mass of
sand injected. That the carrying capacity
correlates with well spacing and porosity
suggests a relationship between carrying

capacity and the drainage volume of the
well. The hyperbolic constant n and constant a have no linear relationship with
reservoir/completion properties. Neither
was there an obvious nonlinear relationship
from the cross plots.
The results of the statistical analysis
show that the model parameters correlate
to some degree to reservoir and completion
properties. The next step was to use design-of-experiment, numerical reservoir
simulation, and response surface modeling
to develop functional relationships between
model parameters and reservoir/completion
properties.

Functional relationships were developed by generating data from numerical
reservoir simulation models based on the
results of a fractional-factorial DOE. This
process resulted in 256 numerical reservoir
simulation runs. Reservoir and well properties used for this experiment included
fracture half-length, fracture permeability,
initial oil saturation, initial reservoir pressure, wellbore pressure, matrix permeability, number of fracture stages, number
of clusters per stage, porosity, reservoir
thickness, fracture width, fracture spacing,
and oil viscosity
After generating the synthetic data,

TABLE 2
Summary of Computed Correlation Coefficient
Between Model Parameters
And Reservoir/Completion Properties (Parallel-Flow Model)
qi1
(bbl/d)

qi2
(bbl/d)

qi3
(bbl/d)

T1
(days)

T2
(days)

T3
(days)

Number of stages

0.47

0.43

0.31

-0.07

-0.10

-0.06

Lateral length (ft)

-0.19

-0.21

-0.30

-0.13

0.15

0.03

Spacing (acres)

0.32

0.35

0.22

-0.34

0.10

0.09

Initial water saturation
(fraction)

-0.35

-0.54

-0.30

0.18

-0.18

-0.02

LAFAYETTE, LA.-The Thunder Bayou Prospect well in Vermillion Parish,
La., logged 490 gross feet of high-quality
pay within the primary Cris R2 object,
says PetroQuest Energy Inc. The company
has a 50 percent working interest and 37
percent net revenue interest in the well.
For comparison, PetroQuest says it
has produced in excess of 70 billion cubic
feet equivalent gross from Cris R2 sand
since 2012 at its La Cantera project two
miles south of Thunder Bayou.
The company says it expects to begin
production from the Thunder Bayou discovery well in the second quarter, with a
projected gross daily rate of 25 million30 million cubic feet plus associated oil
and natural gas liquids. It adds Thunder
Bayou's land location will allow production
to be established quickly, with significantly
lower facility costs compared with La
Cantera's wet marshland location.
r

0.28

0.21

0.27

-0.09

-0.12

0.22

TVT (ft)

-0.40

-0.59

-0.40

0.29

-0.03

-0.11

Net to gross (fraction)

0.15

0.21

0.29

-0.14

-0.18

0.32

Overpressure (psi)

0.18

0.33

0.19

-0.02

0.07

-0.12

Pressure (psi)

0.18

0.33

0.19

-0.02

0.07

-0.12

Depth (ft)

0.15

0.30

0.17

-0.05

0.14

-0.14

Average injection
pressure (psig)

0.35

0.36

0.16

-0.21

0.09

0.02

Total fluid injected

0.46

0.44

0.30

-0.37

0.08

0.12

Sand (lbs)

Thunder Bayou Well Set
For Production In 2015

Porosity (fraction)

0.48

0.40

0.36

0.02

-0.12

-0.06

Shaded cells = ||  0.2

TABLE 3
Summary of Computed Correlation Coefficient
Between Model Parameters and Reservoir/Completion Properties
(Logistic-Growth Model)
N (bbl)

n

a

Number of stages

0.14

0.02

-0.02

Lateral length (ft)

0.00

-0.13

-0.09

Spacing (acres)

0.25

0.02

0.04

Initial water saturation (fraction)

-0.11

-0.09

-0.01

Porosity (fraction)

0.46

-0.12

-0.05

TVT (ft)

-0.05

-0.09

0.02

Net to gross (fraction)

0.19

0.02

0.01

Overpressure (psi)

-0.04

0.10

0.00

Pressure (psi)

-0.04

0.10

0.00

Depth (ft)

-0.03

0.10

0.02

Average injection pressure (psig)

0.23

-0.02

0.09

Total fluid injected

0.39

-0.02

-0.02

Sand injected (lbs)

0.26

0.02

0.01

Shaded cells = ||  0.2

JANUARY 2015 169
American Oil and Gas Reporter - January 2015

Table of Contents for the Digital Edition of American Oil and Gas Reporter - January 2015
