SPE 166201

Geomechanics Coupled Reservoir Flow Simulation for Diagnostic Fracture

Injection Test Design and Interpretation in Shale Reservoirs
A.S.Padmakar, Chevron

Copyright 2013, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Annual Technical Conference and Exhibition held in New Orleans, Louisiana, USA, 30 September–2 October 2013.

This paper was selected for presentation by an SPE program committee following revi ew of information contained in an abstract submitted by the author(s). Contents of the paper have not been
reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessar ily reflect any position of the Society of Petroleum Engineers, its
officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission to
reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuo us acknowledgment of SPE copyright.

Abstract
Diagnostic fracture injection testing (DFIT) is an extremely popular technique to generate direct estimates of some of the key
characteristics of shale reservoirs. DFIT analysis is generally done using analytical models traditionally developed for high-
permeability reservoirs. The use of DFIT to analyze tight shale reservoirs introduces additional issues regarding applicability
of these techniques for tight systems as well as operational issues such as long well shut-in times to obtain reasonable reservoir
parameters. In this work we conduct geomechanics coupled reservoir flow simulations in order to forward model and analyze
DFITs for tight shale gas reservoirs. This capability greatly enhances our ability to efficiently design the DFITs for tight shale
reservoirs. In this work we simulate a typical DFIT using a geomechanics coupled reservoir flow simulator and generate the
pre-closure and after-closure pressure response from the flow simulation model. This response is then analyzed using a
standard Nolte pre-closure and after-closure analysis technique with an objective to evaluate the reservoir properties. We not
only we show the validity of the Nolte analysis technique for tight rocks but we also provide guidelines on the amount of shut-
in time required to generate a reasonable estimate of the reservoir properties from DFIT pressure response. We also show that
the geomechanics coupled flow simulation of DFIT can provide estimates on fracture dimensions which compare reasonably
with those given by more traditional fracture design tools. We demonstrate that the geomechanics coupled reservoir flow
simulation provides an additional advantage over traditional fracture design tools in that it can numerically model the system
response even after fracture closure.

Introduction
Shale gas reservoirs are typically developed using long horizontal wells with multistage fractures. It is generally accepted that
the nature and extent of the completions in addition to the inherent in-situ formation properties contribute significantly to the
overall performance of multifractured horizontal wells in shale reservoirs. The proper evaluation of formation geomechanical
properties is thus critical for the optimal design of a fracture stimulation job during the completions process. A diagnostic
fracture injection test (DFIT), or mini-fracture test, is an important test that has been used historically to evaluate
geomechanical properties of the formation such as fracture closure pressure and formation leakoff characteristics, which are
valuable input parameters for designing the main fracture stimulation job. Further, when conducted for a long period of time,
DFITs can often yield formation properties such as initial reservoir pressure and effective bulk permeability. However, due to
the tightness of the rocks these tests must be conducted for an extended duration in shales to obtain reliable reservoir
properties, which is often difficult because of the equipment and personnel waiting on-site for the main fracture stimulation
job. In addition, predicting the falloff time required for meaningful estimates of formation pressure and permeability is
difficult, as it depends on having prior knowledge of the permeability in addition to the geomechanical properties of the
formation. In this work we attempt to provide a process to better estimate the time required to monitor the falloff period by
performing geomechanics coupled reservoir flow simulation in order to forward model and analyze DFITs prior to the an
actual field test. This capability greatly enhances our ability to efficiently design and interpret the test and to provide
guidelines on the duration of the test.
2 SPE 166

Background

Diagnostic fracture injection testing involves a short-duration, small volume fracture operation in which a small amount (< 10
barrels) of fluid is pumped into a formation until fracture initiation. The well is then shut in, and the well’s pressure is
monitored to fall off naturally over a prolonged time. The intent is to break down the formation to create a short fracture
during the injection period and then to observe closure of the fracture system during the ensuing falloff period. It is also
referred to as a mini falloff test, injection falloff test or mini-frac test. Fig. 1 shows typical pressure response to a DFIT.

Breakdown Pressure

Bottomhole Pressure
ISIP After-Closure

Fracture Closure
Pressure

Pre-Closure

Linear Flow

Injection Rate

Radial Flow

Fig. 1: Typical pressure response to a DFIT

The typical properties obtained from a field DFIT test are as follows:

Pre-Closure: Fracture closure pressure and closure time

Pre-Closure: Leakoff characteristics (normal/pressure dependence/tip extension)
After-Closure: Initial reservoir pressure
After-Closure: Reservoir permeability (late-time analysis)

In his pioneering work Nolte (1979) presented a detailed analysis of the fracturing process that enabled evaluation of some of
the key parameters that determine the fracture dimensions and the parameters critical to the fracturing process based on the
pressure decline following the fracturing process. The parameters Nolte was able to quantify from the pressure decline are the
fluid loss coefficient, the fracture length and aperture, and the time for the fracture to close. Nolte et al. (1993) showed that for
normal leakoff conditions the pressure decline is a linear function with respect to the G-function. The G-function time is a
dimensionless time function relating the shut-in time to the total pumping period. The G-function analysis provides an estimate
of the fracture closure pressure and the fracture closure time as shown in the G-function transformation plot in Fig. 2.

P Vs. G
Derivatives
Pressure

Fracture
Closure

G dP/dG Vs. G

dP/dG Vs. G

G-Function Time
Fig. 2: G-Function transformation plot

Talley and Nolte et al. (1999) have shown the feasibility of determining transmissibility from after-closure radial flow analysis
and determining reservoir pressure from after-closure linear flow analysis. Barree et al. (1996,1998) extended Nolte’s work to
diagnose other phenomena such as pressure-dependent leakoff (PDL), fracture tip extension, and fracture height recession
based on the detailed analysis of the three pressure functions (P, dP/dG and GdP/dG) plotted against the G-function to
uniquely determine the existence and magnitude of these behaviors. Barree et al. (2009) have further extended this analysis
SPE 166  3

through comprehensive comparison with other diagnostic plots such as the sqrt(t) plot, the log-log plot, the after-closure
analysis (ACA) linear flow, and the radial flow plot. This enables a consistent interpretation of closure pressure and time as
well as proper identification of pre-closure and post-closure flow regimes and reservoir properties across different
interpretation techniques. In addition Craig et al. (2006) and Soliman et al. (2005) have extended traditional pressure transient
analysis for falloff testing for interpretation of long-term pseudolinear and pseudoradial response of DFITs for determination
of reservoir parameters. These techniques complement traditional Nolte after-closure analysis and enable consistent and
comprehensive evaluation of late-time data over different analysis methods.
In addition to the analytical analysis of the fracturing process, a number of two-dimensional and three-dimensional fracture
modeling simulators account for many of the complex phenomena affecting the fracturing process. Veatch et al. (1983a,
1983b) provide a comprehensive review of the hydraulic fracturing design and technology. Meyer et al. (1989) proposed a
criterion for fracture propagation based on the concept of critical stress intensity fracture (KIC). This is based on the
relationship for the stress intensity factor (KI) for a discontinuous displacement on the crack border provided by Bouteca et al.
(1988). Meyer et al. (1989) have combined this fracture propagation criterion with the usual mass conservation, continuity
equation, momentum conservation equation, and elastic width opening pressure condition and obtained the fracture
propagation solution by a numerical integral solution methodology. They have also done extensive comparisons and
benchmarking of their numerical simulator with other similar numerical tools as well as with corresponding analytical models
where such models are applicable. In this work we have used the MFrac™ three-dimensional hydraulic fracture modeling tool
by Meyer Associates, which can model the fracture generation process based on injection schedule, stress profile and a few
other rock properties. This model accounts for variable confining stress contrast, variable fracture toughness, moduli, rock
deformation, time-dependent fluid rheology, fluid leakoff into formation and proppant transport with settling and closure.
However, a major limitation of this model is that it is a pseudo-3D model, not a fully coupled 3D model that is more robust in
determining the fracture dimension and its propagation in vertical and linear dimension (Abou-Sayed et al. 1990).

Description of the Geomechanics Coupled Reservoir Flow Simulator used in this Study

In this work we have examined the utility of two major classes of modeling tools to aid with the design of the DFIT. The first
class of tools, essentially the MFrac tool mentioned in the previous section, is primarily focused on fracture initiation,
propagation and subsequent closure. This tool has a very rigorous capability for fracture initiation, propagation, fluid leakoff
and closure but does not handle reservoir flow behavior in the formation in a very rigorous manner. The second class of tools
is standard reservoir flow simulators. These tools are designed to handle the flow in the reservoir in a rigorous manner but do
not typically include the underlying geomechanical effects of fracture creation and the corresponding stress effects on the
formation and their impact on the formation properties. Most of them have pressure dependent permeability and porosity
function at best. However there are several comprehensive numerical flow simulators which incorporate full geomechanics
capabilities. Dean et al. (2006) have presented various different schemes for coupling the geomechanics calculations with the
flow calculations and compared the results from these different techniques for several sample problems. One advantage of
these comprehensive tools is that they can be quickly turned around for flow simulation and have all the related geomechanics
effects captured during the flowback stage of the hydraulic-fracturing operation and during subsequent production stage. In
addition, they do not require explicitly incorporating any new physics (such as fracture growth or stress shadow effects) as the
underlying coupled geomechanics model is fully equipped to handle these scenarios.
In this work we have used the GEM® software tool provided by the Computer Modeling Group (CMG). GEM is a numerical
reservoir flow simulation tool with a coupled geomechanics feature that can model fracture initiation, propagation, closure,
and falloff behavior of a typical DFIT. These models are discussed in detail and have been applied for geomechanical
evaluation of a saline aquifer during CO2 sequestration by Tran et al. (2005, and 2009). A major feature of numerical
simulators is grid-scale specification of reservoir and fracture properties.
In this work the Barton-Bandis model (Tran et al. 2009) is used to specify the relationship between the fracture opening and
the permeability of the fracture system. In this model a secondary fracture system is defined in the grid via the standard dual-
permeability formulation. As pressure increases in the regular grid, the normal stresses on the fractures increase. Eventually
the stress breaks past the failure envelope of the rock, causing a fracture to propogate and allow fluid to flow through the
fracture system in addition to the underlying matrix system. The Barton-Bandis stress permeability relationship model is
shown in Fig. 3.

MFrac is a trademark of Meyer & Associates, Inc.

GEM is a federally registered trademark of the Computer Modeling Group, Ltd.
4 SPE 166

Stress Path
khf

Fracture Permeability
kccf

Subsequent Stress Path

krcf
Initial Stress Path

FRS Normal Fracture Effective Stress (σn’ )

Fig. 3: Barton-Bandis stress permeability relationship model

In the Barton-Bandis model the fracture permeability is computed by the following equations:

(1)

- (2)

(3)

- (4)

Test Problem and 3D Fracture Modeling

In this work we set up a test case scenario representative of a typical deep overpressured tight shale system that had the
properties are shown in Table 1.

Top Depth (ft) 10940

Thickness (ft)
(Overburden/Formation/Underburden) 40 / 210 / 50
Fracture Injection Depth (ft) 198 from top of overburden
Initial Reservoir Pressure (Psia) 9500
Reservoir Temperature (F) 280
Total Porosity (%) 8.5
Total Water Saturation (%) 35
Permeability Variable
Formation Compressibility (1/psi) 5e-6
Fracture Compressibility (1/psi) 5e-5
Fracture Toughness (psi-in^1/2) 1580
Young’s Modulus (psi)
(Overburden/Formation/Underburden) 1.1E7 / 3.4E6 / 1.1E7
Poisson’s Ratio
(Overburden/Formation/Underburden) 0.275 / 0.19 / 0.275
 SPE 166  5

Vertical Stress Gradient (psi/ft) 1.075

Horizontal Stress Gradient (psi/ft) 0.879
Injection Fluid Water
Injection Rate (bpm) 2
Injection Duration (min) 5
Table 1: Description of the test case scenario

The permeability in the test problem has been varied over a few orders of magnitude by keeping all the other parameters
constant so as to estimate the effect of the permeability on the key variables that drive the DFIT design and analysis in the 3D
fracture model. The formation permeabilities studied in this work ranged from 1E-5 mD to 1E-2 mD, with a base case
permeability of 1E-4 mD. One of the motivations for this study was to determine the fracture characteristics from a typical
DFIT, especially the vertical height growth, and to estimate the time required to conduct these tests to obtain reasonable
reservoir properties. Knowing the expected height of the fracture would help in planning the tests across a vertical wellbore in
such a manner that interference of the fractures from multiple tests can be eliminated by allowing for sufficient distance
between multiple DFITS along the same wellbore. The method to perform multiple DFITs in a single well bore stems from the
need to characterize different formations simultaneously, and Martin et al. (2013) have described a method to conduct the
same in mature, depleted formations. The vertical extent of the fractures can be estimated a priori from the 3D fracture model,
and the possibility for overlap between two adjacent fractures can be avoided by spacing the DFITs appropriately along the
wellbore. In addition to the fracture characteristics, the 3D fracture models also provide the time to fracture closure. The
results from the fracture models are provided in Table 2. The leakoff coefficient is assumed to be constant for each of the
cases and has been calculated assuming no filtrate invasion and with a constant pressure drop of 150 psi across the fracture
plane into the formation.

Leakoff Fracture Fracture Closure

Perm Coefficient Length Height Fracture Tip Time
(mD) ft/sqrt(min) (ft) (ft) Width (in) Extension (min)
1.00E-05 6.68E-06 122 200 0.023 Yes 3200
1.00E-04 2.11E-05 107 178 0.023 Yes 564
1.00E-03 6.68E-05 94 162 0.0233 Yes 100
1.00E-02 2.11E-04 80 138 0.026 No 29
Table 2: Summary of parameters from 3D fracture models

The fracture model suggests a very high fracture height growth even in the high leakoff (high permeability) scenarios. The
fracture model seems to exhibit a much larger fracture height in comparison to the fracture length. This could be because the
fracture model accounts for the minimum horizontal stress only and does not include the over-burden stress in the fracture
propagation calculation. As expected, the low permeability scenarios show significant fracture tip extension both vertically and
horizontally for a significant period after the end of the shut-in period. A typical example of this for the 1E-4 mD case is
shown in Fig. 4, which indicates increase in the fracture length and height even after shut-in. As expected, the tip extension is
more prominent in the low permeability cases. This is significant as it has been reported that, for tight formations, the fracture
tip extension can often be misinterpreted as pressure dependent leakoff on the traditional Nolte G-function analysis plot
(Barree et al. 2009).
 6 SPE 166

250 10000

200 9900
Fracture Length (ft)

Fracture Length, Height (ft)

Fracture Height (ft)
Pressure (Psia)

Pressure (Psia)
150 9800

100 9700

50 9600

0 9500
0.001 0.01 0.1 1 10 100
Time (Hours)
Fig. 4: Summary of fracture modeling results for the 1E-4 mD rock

Fig. 5 shows the fracture width contours and the width profiles for the 1E-4 mD rock. In this case the fractures grow
asymmetrically in the vertical direction due to the stress contrast between the reservoir and the underlying formation.

Fig. 5: Width contours from fracture modeling for the 1E-4 mD rock

Geomechanics Coupled Flow Simulation

As discussed earlier in this work we have also used the GEM numerical reservoir flow simulation tool with a coupled
geomechanics feature to model the DFIT case scenario described in the previous section. Table 3 shows the x-direction and y-
direction grid in the geomechanical coupled flow models used in the simulation of the DFIT for the test problem (the z-
direction grids are 1 ft thick each). The minimum horizontal stress and the overburden stress are set at 9616 psi and 11768 psi,
respectively, at the top of the formation with an overburden gradient of 1.075 psi/ft in the vertical direction. In general, the
stresses can be specified at each individual grid block in the numerical simulator such that any variation in the stresses in the
horizontal and the vertical direction can be accounted for in a rigorous manner. In this work we have varied the the overburden
stress but kept the minimum horizontal stress constant across the different layers so as to be consistent with the 3D fracture
models discussed in the previous section.
 SPE 166  7

X-Direction (ft) 4*0.01 1 2 2 5 4*10 50 100 200

Y-Direction (ft) 5*1 2 2 25*10 100 200
Table 3: The grid dimension used in the reservoir flow simulation in this study

The Barton-Bandis parameters used in this study are provided in Table 4. The fracture permeability corresponds roughly with
the value obtained for slit flow for a maximum fracture width (e0) of 3E-6 ft. The corresponding fracture stiffness has been
adjusted such that the fracture permeability approaches the closure permeability (krcf) of 1E-6 mD when the stress in the
fracture approaches the fracture closure pressure according to the Barton-Bandis relationship given in the previous section.

e0 (ft) 3E-6
Fracture Stiffness (Kni) 4E5 psi/ft
Fracture opening stress (FRS) -150 psi
khf 85 mD
kccf 85 mD
krcf 1E-6 mD
Table 4: Barton-Bandis parameters in this study

Fig. 6 shows the fracture profile from the geomechanics coupled reservoir flow simulation for the 1E-4 mD rock. The fracture
dimensions from the geomechanics coupled flow simulation are in general shorter than those observed in the 3D fracture
models discussed in the previous section. This could be because the actual leakoff into the formation is much higher than in
the 3D fracture model which assumes a constant pressure drop and a constant leakoff coefficient across the fracture. Also the
fracture width in the simulation model is larger than that calculated by the 3D fracture model. Similar to the 3D fracture
model, the flow simulation model also exhibits fracture length and height extension for some time after shut-in, as shown in
Fig. 7. As mentioned earlier, this has significance when interpreting the closure time from the Nolte G-function analysis.

Fig. 6: Fracture profile from geomechanics coupled flow simulation for 1E-4 mD rock (Orange color denotes crack opening
permeability)
 8 SPE 166

125 10000

Fracture Length (ft)

100 Fracture Height (ft)
9900
Fracture Length, Height (ft)

Pressure (Psia)

Pressure (Psia)
75 9800

50 9700

25 9600

0 9500
0.001 0.01 0.1 1 10 100 1000 10000
Time (Hours)

Fig. 7: Pressure response and fracture dimensions of 1E-4 mD rock from geomechanics coupled flow simulation of DFIT

Analytical Interpretation of the Geomechanics Coupled Flow Simulation Results

As mentioned in a previous section, we cannot obtain any reservoir related parameters from the 3D fracture models as they do
not explicitly model flow into the reservoir after fracture closure. However, the pressure response from the geomechanics
coupled flow simulation is amenable for both pre-closure and after-closure analyses as it capture both the fracture propagation
as well as the reservoir falloff regimes of the DFIT. Fig. 8 shows the Nolte pre-closure analysis of the synthetic pressure
signature obtained from the geomechanics coupled flow modeling simulation for the 1E-4 mD case. The Nolte G-function
analysis gives us a closure time of 1152 min and the fracture closure pressure of 9565 psia. The closure time is roughly double
the closure time obtained from the fracture models in Table 2 for this case. The closure pressure, however, is lower than the
input minimum horizontal stress of 9617 psia in Table 1.
 SPE 166  9

G-Function
120 70000 0.40
Analysis 1 Inj. Volume 9.94 bbl
ISIP 35296.3 psi(a) 68000
115 Ddatum 10940.000 ft 0.38
Frac grad 3.226 psi/ft 66000
G-Function Alpha 1.000
110
64000
0.36
105 62000

60000 0.34
100
58000
0.32
95
56000

90 54000 0.30
Fracture Closure 52000
85 Gc 37.038
0.28
tc 1152.00 min 50000
pc 9565.4 psi(a)
80 48000
0.26
46000
Semilog Derivative G dp/dG (psi(a))

75

First Derivative dp/dG (psi(a))

44000 0.24
70
42000
0.22
65 40000

p (psi(a))
38000
60 0.20
36000
55
34000 0.18

50 32000
0.16
30000
45
28000 0.14
40
26000
0.12
35 24000

22000
30 0.10
20000
25
18000 0.08
Fracture Closure
20 16000
Gc 37.038 0.06
tc 1152.00 min 14000
15 pc 9565.4 psi(a)
12000
0.04
10
10000

Semilog Derivative 8000 0.02

5
p data
First Derivative 6000
0 0.00
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 125 130 135 140 145 150 155 160 165 170 175 180 185 190 195 200

G-function time

Fig. 8: Nolte G-function analysis of the geomechanics coupled flow simulation response for 1E-4 mD rock

Fig. 9 shows the semi-log derivative plot that identifies the -3/4 slope bilinear flow regime and the -1 slope pseudoradial flow
regime used in the Nolte after-closure analysis. We could not identify a -1/2 slope pseudolinear flow regime from the Nolte
after-closure analysis for this rock (or for the tighter 1E-5 mD rock). We see from the semi-log derivative plot that, for the 1E-
4 mD rock, the time it takes to reach bilinear flow and pseudoradial flow is roughly 480 and 6500 hours respectively. As is
expected, the time to reach pseudoradial flow is prohibitively long to be able to estimate the formation permeability reasonably
accurately from DFITs, especially for rocks that are 1E-4 mD and tighter. The corresponding reservoir pressure estimated from
the Nolte after-closure analysis is 9500 psia same as the input reservoir pressure in the numerical simulator. The corresponding
reservoir permeability estimated from the Nolte after-closure analysis is 5.4E-4 mD, which is higher than the input value of
1E-4 mD. Fig. 10 shows the pressure profiles at 600 and 8000 hours obtained from the flow simulation. The two profiles show
distinct linear flow and pseudoradial pressure propagation for the two different time periods, which confirms the flow regimes
identified by the semi-log derivative plot.
 10 SPE 166

Derivative
105
Inj. Volume 9.94 bbl
ISIP 35296.3 psi(a)
5
Ddatum 10940.000 ft
3 Frac grad 3.226 psi/ft
2 Fracture Closure
tc 1152.00 min
104
pc 9565.4 psi(a)

3
2

103

3
p, Semilog Derivative (t)dp/d(t) (psi(a))

Fracture Closure
102
tc 1152.00 min
5 pc 9565.4 psi(a)

3
2

101

3 Slope -3/4
2
Slope -1
1.0

3
2

10-1

3
2

10-2

3
2 pdata
Derivative data
10-3
10-4 2 3 4 5 6 7 8 10-3 2 3 4 5 6 7 8 10-2 2 3 4 5 6 7 8 10-1 2 3 4 5 6 7 8 1.0 2 3 4 5 6 7 8 101 2 3 4 5 6 7 8 102 2 3 4 5 6 7 8 103 2 3 4 5 6 7 8 104 2 3 4 5 6 7 8 105

t (h)

Fig. 9: Semi-log derivative plot for the 1E-4 mD rock

Fig. 10: Pressure contours for 1E-4 mD rock at 600 and 8000 hours respectively

Fig. 11 shows the Soliman-Craig impulse derivative plot that identifies the bilinear (1/4 slope ) flow regime, pseudolinear (1/2
slope) flow regime and the pseudoradial (zero slope) flow regime. The Soliman-Craig analysis for the 1E-4 mD rock shows
that the time to reach bi-linear, pseudolinear flow and pseudoradial flow is roughly 580, 3500 and 7000 hours, respectively. As
mentioned earlier, the time to reach pseudoradial flow is prohibitively long. The corresponding reservoir pressure estimated
from the Soliman-Craig analysis is 9497 psia, which is reasonable, but the permeability estimate is 4E-4 mD, which is higher
than the input permeability of 1E-4 mD similar to the Nolte after-closure analysis. One advantage of the Soliman-Craig falloff
analysis over the Nolte after-closure analysis is that it is not dependent on the closure time, which can be subjective and
difficult to ascertain from the G-function plot and can affect the subsequent Nolte after-closure analysis.
 SPE 166  11

Derivative
105

104
Bilinear 1/4

6 kfW(sqrt(k)) 0.00 md 3/2 ft

k md
4 kfW md.ft Radial 0

3 k 4.3964e-04 md
Impulse Derivative t (tp + t) dp/dt ((psi/cP) hr)

Linear 1/2
103 Xf(sqrt(k)) 1.40 md 1/2 ft
k md
Xf ft
6
sXf

102

101

Derivative data
1.0
10-4 2 3 4 5 6 7 8 10-3 2 3 4 5 6 7 8 10-2 2 3 4 5 6 7 8 10-1 2 3 4 5 6 7 8 1.0 2 3 4 5 6 7 8 101 2 3 4 5 6 7 8 102 2 3 4 5 6 7 8 103 2 3 4 5 6 7 8 104 2 3 4 5 6 7 8 105

t (h)

Fig. 11: Soliman-Craig analysis of synthetic pressure response for the 1E-4 mD rock

Table 5 provides a summary of the fracture dimensions obtained from the geomechanical coupled reservoir flow simulation
and the results from the subsequent DFIT analysis of the pressure response generated from these simulations. As mentioned
earlier, the fracture dimensions obtained from the flow simulations are similar to those from the 3D fracture models. However,
the actual length and height are smaller than those from the 3D fracture model for all the different permeability scenarios. The
closure time obtained from the Nolte G-function analysis seems to be shorter than that from the 3D fracture models for the
higher permeability scenarios and lower for the lower permeability scenario. This could be because it is difficult to ascertain
the peak in the derivative of the semi-log G-function plot especially for the lower permeability 1E-5 mD rock. The much
shallower peak in the semi-log derivative G-function plot has been observed by Nojabaei and Kabir et al. (2013) as well for
tight rocks. This can in turn lead to improper diagnosis of the closure time for tight rocks. The summary of Nolte after-closure
analysis as well as Soliman-Craig analysis shows an increase in time to linear flow as well as radial flow, with decreasing
input permeability as expected. However, the actual time to linear flow and radial flow differ between the two analyses, which
could be due to subjective interpretation inherent in both the techniques. The permeability estimates from the DFIT analyses
are generally lower for the higher permeability rocks, which might also be due to the subjective nature of the interpretation. In
contrast, the permeability estimate for the lower permeability rock, especially the 1E-5 mD rock, is about an order of
magnitude higher from both Nolte after-closure analysis and Soliman-Craig analysis. This could be due to insufficient time for
radial flow to fully develop even with an extremely long test duration for this tight rock.

Geo-mechanical
Perm Nolte G-Function
Model Fracture Nolte After Closure Analysis Soliman-Craig Analysis
(mD) Analysis
Parameters

Initial Time to Time to Initial

Closure Time to Bi- Time to Time to Perm Time to Perm
Closure Pressure Bi-linear Linear Pressure
Length (ft) Height (ft) Pressure linear flow Linear Radial Estimate Radial Estimate
Time (min) Estimate Flow Flow Estimate
(Psia) (hrs) Flow (hrs) Flow(hrs) (mD) Flow(hrs) (mD)
(Psia) (hrs) (hrs) (Psia)

1.00E-05 154 126 108 9935 264 - 10000 9500 4.43E-04 600 - 12000 9500 4,34e04
1.00E-04 84 110 1152 9565 480 - 6500 9502 4.67E-04 580 3500 7000 9499 4.39E-04
1.00E-03 54 97 360 9600.3 - 46 900 9500 1.10E-03 - 40 700 9498 8.29E-04
1.00E-02 44 71 265 9613 - 10 95 9499 4.37E-03 - 15 45 9494 3.40E-03

Table 5: Summary of properties from analytical interpretation of geomechanical coupled flow simulation pressure response for
DFIT for different permeablities
 12 SPE 166

Conclusions

1) Geomechanical coupled reservoir flow simulations allow the modeling of the pressure response of DFIT before
fracture closure as well as the falloff period after closure of the fracture.
2) Traditional 3D fracture modeling tools provide an estimate of the fracture dimensions as well as the time to fracture
closure. However, they cannot provide the pressure response into the reservoir during the falloff regime.
3) 3D fracture modeling tools and geomechanics coupled reservoir flow simulation show similar trends in fracture
geometry with permeability. However, actual fracture dimensions differ, which could be due to different assumptions
between the two models.
4) Nolte pre-closure analysis, Nolte after-closure analysis and Soliman-Craig falloff analysis are extremely valuable
techniques in analyzing the synthetic pressure response generated from geomechanics coupled flow simulation in
order to estimate time to fracture, time to linear flow and time to pseudoradial flow. This provides an a priori estimate
of expected duration to conduct the actual DFIT in the field to obtain meaningful fracture and reservoir properties.
5) Finally, caution should be exercised when interpreting reservoir properties (especially permeability) in a tight
formation, particularly when tests are of a very short duration.

Acknowledgments

The author acknowledges Chevron management for granting permission to publish this work. The author also thanks Kurt
Wilson, Simeon Eburi and Andrea Farr for their valuable suggestions on the manuscript.

Nomenclature

e = Current fracture aperture (ft)

e0 = Initial fracture aperture (ft)
FRS = Fracture Opening Stress (Psi)
Kni = Fracture Stiffness (Psi/ft)
kccf = Fracture permeability at zero stress (mD)
krcf = Fracture closure permeability (mD)
khf = Fracture opening permeability (mD)
Vj = Stress to fracture stiffness ratio (ft)
Vm = Minimum fracture aperture correlated to closure permeability (ft)
σn’ = Normal fracture effective stress (psi)

References

Abou-Sayed, A.S. and Sinha, K.P. 1984. Evaluation of the Influence of In-Situ Reservoir Conditions on the Geometry of Hydraulic
Fractures Using a 3-D Simulator: Part 1 – Technical Approach. Paper SPE 12877 presented at the SPE/DOE/GRI Unconventional Gas
Recovery Symposium, Pittsburgh, Pennsylvania, USA, 13–15 May.
Barree, R.D. 1998. Applications of Pre-Frac Injection/Falloff Tests in Fissured Reservoirs – Field Examples. Paper SPE 39932 presented at
the SPE Rocky Mountain Regional/Low Permeability Reservoir Symposium, Denver, Colorado, USA, 5–8 April.
Barree, R.D., Barree, V.L. and Craig, D.P. 2009. Holistic Fracture Diagnostics: Consistent Interpretation of Prefrac Injection Tests Using
Multiple Analysis Methods. SPE Prod & Oper 24 (3): 396–406. SPE-107877-PA. doi: 10.2118/107877-PA.
Barree, R.D. and Mukherjee, H. 1996. Determination of Pressure-Dependent Leakoff and Its Effect on Fracture Geometry. Paper SPE 36424
presented at the SPE Annual Technical Conference and Exhibition, Denver, Colorado, USA, 6–9 September.
Bouteca, M.J. 1988. Hydraulic Fracturing Model Based on a 3-D Closed Form: Tests and Analysis of Fracture Geometry and Containment.
SPE Prod Eng 3 (4): 445–454. SPE-16432-PA. doi: 10.2118/16432-PA.
Craig, D.P, and Blasingame, T.A. 2006. Application of a New Fracture-Injection/Falloff Model Accounting for Propagating, Dilated, and
Closing Hydraulic Fractures. Paper SPE 100578-MS presented at the SPE Gas Technology Symposium, Calgary, Alberta, Canada, 15–
17 May. doi: 10.2118/100578-MS.
Dean, R.H., Gai, X., Stone, C.M, and Minkoff, S.E. 2006. A Comparison of Techniques for Coupling Porous Flow and Geomechanics. SPE
J. 11 (1): 132–140. SPE-79709-PA. doi: 10.2118/79709-PA.
Martin, A.R., Cramer, D.D., Nunez, O. and Roberts, N.R. 2013. A Method to Perform Multiple Diagnostic Fracture Injection Tests
Simultaneously in a Single Wellbore. SPE Prod & Oper 28 (2): 191–200. SPE-152019-PA. doi: 10.2118/152019-PA.
Meyer, B.R. 1989. Three-Dimensional Hydraulic Fracturing Simulation on Personal Computers: Theory and Comparison Studies. Paper SPE
19329 presented at the SPE Eastern Regional Meeting, Morgantown, West Virginia, USA, 24–27 October.
Nojabaei, B. and Kabir, C.S. 2012. Establishing Key Reservoir Parameters with Diagnostic Fracture Injection Testing. SPE Res Eval & Eng
15 (5): 563–570. SPE-153979-PA. doi: 10.2118/153979-PA.
Nolte, K.G. 1979. Determination of Fracture Parameters from Fracturing Pressure Decline. Paper SPE 8341 presented at the SPE Annual
Technical Conference and Exhibition, Las Vegas, Nevada, USA, 23–26 September.
Nolte, K.G., Mack, M.G. and Lie, W.L. 1993. A systematic method for applying fracturing pressure decline: Part 1. Paper SPE 25845
presented at the SPE Rocky Mountain Regional/Low Permeability Reservoir Symposium, Denver, Colorado, USA, 12–14 April.
SPE 166  13

Soliman, M.Y., Craig, D., Barko, K., Rahim, Z., Ansah, J. and Adams, D. 2005. After-Closure Analysis to Determine Formation
Permeability, Reservoir Pressure, and Residual Fracture Properties. Paper SPE 93419-MS presented at the SPE Middle East Oil & Gas
Show and Conference, Kingdom of Bahrain, 12–15 March. doi: 10.2118/93419-MS.
Talley, G.R., Swindell, T.M., Waters, G.A. and Nole, K.G. 1999. Field Application of After-Closure Analysis of Fracture Calibration Tests.
Paper SPE 52220 presented at the SPE Mid-Continent Operations Symposium, Oklahoma City, Oklahoma, USA, 28–31 March.
Tran, D., Nghiem, L. and Buchanan, L. 2005. An Overview of Iterative Coupling between Geomechanical Deformation and Reservoir Flow.
Paper SPE 97879 presented at the SPE International Thermal Operation and Heavy Oil Symposium, Calgary, Alberta, Canada, 1–3
November.
Tran, D., Shrivastava, V., Nghiem, L. and Kohse, B. 2009. Geomechanical Risk Mitigation for CO2 Sequestration in Saline Aquifers. Paper
SPE 125167 presented at the SPE Annual Technical Conference and Exhibition, New Orleans, Louisiana, USA, 4–7 October.
Veatch, R.W. 1983. Overview of Current Hydraulic Fracturing Design and Treatment Technology – Part 1. J Pet Technol 35 (4): 677–687.
SPE-10039-PA. doi: 10.2118/10039-PA.
Veatch, R.W. 1983. Overview of Current Hydraulic Fracturing Design and Treatment Technology – Part 2. J Pet Technol 35 (5): 853–864.
SPE-11922-PA. doi: 10.2118/11922-PA.