SPE-177027-MS

Increasing Production and Reserves in a Mature Field with Hydraulic

Fracturing by Combining Fracture Pressure Analysis, Pressure Transient
Analysis, and Rate Transient Analysis
E. Murminacho, H. Sanchez, and M. Lopez, Operaciones Rio Napo C.E.M; R. G. Rachid, J. Maniere,
and A. Milne, Schlumberger

Copyright 2015, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Latin American and Caribbean Petroleum Engineering Conference held in Quito, Ecuador, 18 –20 November
2015.

This paper was selected for presentation by an SPE program committee following review 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 necessarily 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 conspicuous acknowledgment of SPE copyright.

Abstract
Even as new discoveries are developed and produced, substantial additional reserves can be secured by
enhancing recovery from mature fields. It is known that one of the most cost-effective ways to revitalize
mature fields is to institute an efficient workover program. These programs can be more effective because
of the implementation of fracturing/refracturing techniques for unlocking new pay intervals or putting
back on production existing, nonproducing stages. Fracture stimulation is commonly used to increase the
production and improve the economics of oil and gas fields worldwide. In many cases, however, if the
average production increase justifies the hydraulic fracture, very little effort is made to optimize the
treatments. Further economic gains can be achieved through the optimization of the hydraulic fracturing
process by improving the fracture model calibration and better understanding of reservoir parameters. This
is particularly true in mature fields where the fracture geometry understanding is critical to avoid
contacting a water zone or a gas cap in some cases. Also, sometimes there is an impact on the reserves
determined by decline curve analysis (DCA) after the fracturing treatment.
The analysis of the fracturing model through pressure interpretation or fracture pressure analysis (FPA)
provides powerful tools not only for understanding and improving the fracturing design parameters, but
also for better understanding the reservoir parameters. On the other hand, reservoir transmissibility, a
variable of significant importance in hydraulic fracturing design and production evaluation (PE), is
frequently unknown because candidate wells either do not flow or pretreatment pressure transient tests are
not available. The after-closure analysis (ACA), the component of the FPA in the after-closure region,
allows for the identification of linear and radial flow and the determination of reservoir transmissibility
and pressure from injection testing. The reservoir parameters determined with this method can be verified
by comparison with results obtained from conventional well testing or pressure transient analysis (PTA)
and rate transient analysis (RTA).
This paper presents a detailed analysis and comparison of the reservoir parameters obtained by FPA,
PTA, and RTA in block 60 (Sacha field), a mature field located in the Oriente basin operated by
Operaciones Rio Napo CEM. The three techniques have been integrated for reservoir modeling in a
2 SPE-177027-MS

moderately permeable, underpressure, black oil sandstone reservoir allowing not only increasing the
production but also the reserves in the hydraulically fractured wells.
Introduction
Fracture pressure analysis (FPA) comprises a number of specific tests and their interpretation developed
by Nolte (1979) for fracturing pressure pumping and decline analysis and then extended to after-closure
analysis (Nolte et al. 1997). FPA explains not only the primary fracture propagation behavior but also
provides the chance to have a direct reservoir characterization (transmissibility and pressure). The FPA
methodology is based on determining consistency between the dynamic closure tests—step rate test
(SRT), equilibrium test (Weng et al. 2002), and flowback rebound test (FBRT)— net pressure trend from
the Nolte-Smith plot, pressure decline analysis before the fracture closes from Nolte’s G-function plot,
and the period after the fracture closes (ACA).
The development of increasingly lower permeability reservoirs where FPA cannot be directly applied
has led to a significant increase in the prestimulation injection testing operations for reservoir character-
ization, generally referred as the ⬙diagnostic fracture injection test⬙ (DFIT) in the literature. These tests,
almost always consisting of a single small injection, are carried out in the most challenging geological
environments (extremely low-permeability and high-stress formations, such as shale plays) in which most
of the obtained information is often less than satisfactory, which is understandable in view of the
geological complexity. Most of the injected fluid is believed to be slowly transferred from the dominant
fractures into a network of dynamically activated fissures long after the fluid injection stops, and any
diffusion into the matrix itself (with permeability ranging from 10 to 100s nanodarcies) would require a
well to be shut-in for an unrealistically long period of time to see any recognizable reservoir response.
Although it may be possible to have more confidence in the evaluation of in-situ stresses by combining
classical FPA with repeated cycles of injection and depressurization (Fidan et al. 2014), proper charac-
terization of the reservoir transmissibility and pressure is still hindered by the inerrant complexity of the
fracture/fissure interactions and the theoretical time required to reach a transient flow regime in a tight
reservoir.
However, in conventional reservoirs, the FPA methodology is often ignored. The purpose of this paper
is to illustrate the benefits that can be obtained by applying the FPA methodology to improve the
management of conventional reservoirs that are still the primary source of hydrocarbons worldwide. FPA
is not only an important tool for production enhancement through optimization of hydraulic fracturing, it
is also an excellent means of determining reservoir pressure and transmissibility when well test data or
pressure transient analysis (PTA) is not available.
In Ecuador, there are mature subhydrostatic sandstone reservoirs that have lateral waterdrive or that are
located close to the oil/water contact (OWC). With time, the water saturation has increased and water
breakthrough is common in the high-permeability streaks resulting in significant water cut increase. In
these fields, the challenge is to increase oil production and recoverable reserves without increasing water
production. For this reason, until recently, hydraulic fracturing was not considered a viable option due to
the proximity to the OWC.
The Sacha field, which is geographically located in the Oriente basin in Ecuador, is an example of a
mature field with an active aquifer. The main productive reservoir is the Hollin formation, which is
composed of the Upper Hollin (UH) and Lower Hollin (LH), which are separated by a 10- to 30-ft shale
break. In some wells only the UH produces since the LH is invaded by a strong aquifer. For this reason,
matrix treatments were historically performed to selectively treat the UH, with mixed results. Sometimes,
the matrix acidizing treatments were not sufficient to fully remove the damage because of the high clay
content, whereas in other cases, the radial extent of the damage was excessive, making the matrix
treatments uneconomic. To provide more treatment penetration, hydraulic fracturing was proposed as a
recompletion strategy to fully bypass the damaged zone, but there was concern regarding fracture
SPE-177027-MS 3

proximity to the OWC because of the small shale break. To mitigate this risk of producing with high water
cut after fracturing, low-viscosity fracturing fluids based on viscous disproportionate permeability
modifiers (VDPM) were selected for use. After combining the VDPM-based fracturing fluids and the
engineering application of the FPA, a new opportunity was opened for this mature field since several
marginal wells could be hydraulically fractured and oil production was increased without increasing the
water cut since the wet zone was not contacted with the fracture. The ACA provided information related
to reservoir parameters (transmissibility and pressure) that was not available in most of the wells from
PTA. In other words, the correct understanding of fracturing parameters and modeling through FPA were
the key factor to release this mature reservoir’s potential.

Fracture Pressure Analysis: Theoretical Background

Fracture pressure analysis (FPA), during and after fracturing treatments, has been used extensively to
estimate fracture and reservoir parameters. The technique is commonly employed on location to estimate
fracture design parameters and the pressure drop (tortuosity) due to the orientation of the hydraulic
fracture in the formation with respect to the perforations at the wellbore (Behrmann and Nolte 1998).
When the true value of formation transmissibility (or permeability) is known, the fracture geometry can
be optimized in terms of productivity. However, accurate formation permeability estimates from well tests
are rarely available.
During a hydraulic fracture treatment, the fracture growth and propagation are followed by a pressure
decline when the pumps are stopped, that ultimately could approach the reservoir pressure in some cases.
Hence, the pressure interpretation during the treatment can be divided into two periods: fracture treatment
and pressure decline, represented in Fig. 1. The pressure response during the fracture treatment period is
used to evaluate the geometry/growth of the hydraulic fracture into the formation (Nolte and Smith 1981).
The data from the pressure decline as the hydraulic fracture closes are used to quantify fluid efficiency,
nonideal events (such as post-injection fracture propagation, pressure dependent leakoff to fissures, and
height recession during closure) and also the hydraulic fracture geometry (Nolte 1979). Finally, after the
fracture has closed, the pressure decline can reach pseudolinear and pseudoradial flow, which can be
analyzed in a similar manner to traditional well test to determine transmissibility (kh/␮) and reservoir
pressure (Gu et al. 1993; Abousleiman et al. 1994). The correct interpretation of the pressure response
during a hydraulic fracturing treatment depends on identifying the closure pressure (pc) (Fig. 1). However,
in most nonideal cases, pc can only be correctly defined using dynamic closure pressure tests—step rate
tests (SRT) and flowback rebound tests (FBRT)— to obtain an independent estimate. Meanwhile, an
extensive review of FPA has shown that many characteristics of G-function plots, which are described in
the literature as being the signature of a unique event, are not, in fact, limited to a unique event. In
particular, the S-shape decline curve, which is conventionally considered to be the unique signature of
height growth, has been found to reflect other events, such as a tight reservoir with natural fissures,
invasion of crosslinked gel in highly permeable reservoirs (creeping internal filter cake), and misalign-
ment of far-field fractures with the wellbore (tortuosity).
4 SPE-177027-MS

Figure 1—Hydraulic fracture cycle schematic showing the primary parameters involved affecting the bottomhole pressure behavior

It is theoretically possible to obtain a similar estimate of the fracture length (or leakoff) from both the
preclosure analysis and after-closure analysis (ACA), and this shows ⬙synergy⬙ between them, by
confirming the fracture length from two perspectives: a mechanical closure analysis (calibration of the
fracture’s model geometry) and pressure transient analysis using ACA (Nolte et al. 1997).
However, PTA of a hydraulically fractured well requires that both linear and radial flow regimes to be
present to get a unique match. In the case of ACA, no bilinear flow is expected until the stress on the
unpropped fracture is 2,000 psi below closure pressure (Fredd et al. 2000) and fracture conductivity below
1 md-ft. This limits the communication between the wellbore (where the measurement is taken) and the
reservoir itself.
In most cases, because of the time required with the well shut-in to see well-defined linear and radial
flow regimes, a single injection test is unlikely to provide the information required. For this reason, an
alternative approach is to create a small hydraulic fracture with the first injection test using brine to obtain
the after-closure radial flow regime. Then the after-closure linear flow regime can be determined from the
calibration-injection test performed prior to the main fracturing treatment.
Within the stimulation community, different injection tests are performed to determine reservoir and
fracturing fluid parameters to optimize the treatment design. Although the theoretical basis of these tests
are identical, the commercial names differ, which unfortunately has created some confusion in terms of
their correct deployment and analysis. However, in every case, the pressure transient period of any
injection test into the reservoir can be used to determine the reservoir pressure and transmissibility
(Economides and Nolte 2000).
A mini falloff (MFO) test is a short injection test that creates a fracture (or not) with a Newtonian fluid,
using standard high-pressure pumping equipment and sensors, that results in a transient pressure buildup
and falloff in the reservoir. Similar to a production buildup test, the analysis of the MFO test provides the
reservoir transmissibility (kh/␮) and pressure, when the pressure decay is solely governed by pressure
diffusion in the porous medium (reservoir). This is in marked contrast with the modified Mayerhoffer
method, which uses the closure pressure period, in which the results are highly dependent on a correct
estimate of the fracture geometry (Ispas et al. 1998). Performing an MFO test that creates a hydraulic
SPE-177027-MS 5

fracture has the advantage of bypassing any near-wellbore damage and connecting laminations in the
formation that may not be connected to the wellbore, which maximizes vertical coverage of the reservoir.
For this reason, an MFO test is preferred to a closed chamber test in formations where fluid inflow is
severely restricted by formation damage or a limited number of perforations. In addition, as part of the
MFO ACA, an inversion of the reservoir flux shut-in time permits an independent determination of the
linear flow initiation which is associated to the closure time. The method of estimating closure time (Tc)
proposed by Nolte is described in Appendix A.
A diagnostic fracture injection test (DFIT) describes an injection test similar to an MFO, although this
test is carried out with a linear or crosslinked gel fluid in some cases (similar to a minifrac or calibration
injection test). A DFIT may start with an SRT, the pump rate increasing in steps, and, at the end of a period
of constant injection, finish with a stepdown test (SDT) when the pump rate is stepped down. Combining
an SRT and SDT in the same injection sequence creates a deviation from the condition of constant rate
that is required for a valid G-function analysis (the fluid-loss function used for a G-plot is based on
constant injection rate). Similarly, the after-closure pressure response of a hydraulic fracture in a reservoir
requires a constant injection rate to use a simple pressure transient superposition simulation. The industry
developed the pressure analysis based on constant injection rate as the rate can perfectly be controlled
when injecting an incompressible fluid such as brine. Therefore, when required, the SRT and SDT should
be run as separate tests.
A minifrac or calibration injection/decline tests are performed using a viscous fracturing fluid
immediately prior to the main fracture treatment to calibrate the design parameters, fracture closure
pressure, fracture gradient, fluid leakoff coefficient, and fluid efficiency. The correct value for fluid
efficiency is critical when the design is for a tip screenout (TSO) to create a short highly conductive
fracture in a highly permeable reservoir.
The recommended FPA test sequence is to first break down the formation, preferably immediately after
perforating, to create a short fracture, use the pressure falloff to determine the reservoir properties, and
then perform a dynamic closure test, depending on the reservoir conditions. In a low-permeability but not
overly depleted reservoir, use an FBRT whereas in a highly permeable or depleted reservoir, perform an
SRT and/or equilibrium test. Verify the value of the closure pressure with the reopening pressure at the
start of each injection. Then pump the calibration treatment and confirm the closure pressure during the
falloff period based on deviation of G-plot and the after-closure period. In reservoirs with high perme-
ability or low fluid efficiency, the SRT must be pumped after the calibration test. When bottomhole
pressure gauges cannot be run, the SRT should be pumped while recording instantaneous shut-in pressure
(ISIP), and the SRT plot becomes a plot of ISIP versus rate. With this approach, the total friction at each
pump rate is recorded, which then provides a better estimate of the excess pressure near the wellbore. It
is not necessary for an SRT to reach the maximum pumped rate planned for the fracturing treatment.
Whenever possible, brine should be used for MFO tests, SRT, and FBRT because the computed friction
pressure of brine or water is always more reliable than when friction reducers or polymers are added. For
example, the friction pressure of a linear gel depends on a number of different factors, such as mixing
conditions, hydration time, water composition, water temperature, and the polymer used.

After-Closure Analysis (ACA)

Well testing has been used for decades to determine essential formation properties and to assess wellbore
condition. There are many different types of tests that can be utilized to collect this information depending
on when the test is conducted, the well location, the well type, and the formation type. However, under
certain conditions, traditional well test methods are not feasible for various reasons, for example, depleted
reservoirs that do not flow before being completed.
The period during pressure decline following the fracture closure and preceding the pseudoradial flow
can exhibit also reservoir pseudolinear flow. This period can be used to determine closure time and,
6 SPE-177027-MS

eventually, spurt loss and fracture length (Nolte et al. 1997). If the decline period is sufficient to reach
radial flow, the transmissibility (kh/␮) can also be obtained in the same injection test. By combining the
ACA of standard injection tests with a conventional decline analysis, engineers can attain a more complete
understanding of the fracturing process. It is not the objective of this work to provide a review of ACA;
Nolte et al. (1997) give a review and the procedures. In the after-closure period of an unpropped fracture
injection, there is potential for the reservoir pseudolinear and pseudoradial flow periods. It is known that,
after the fracture closure, the radial flow period is a function of the injected volume, reservoir pressure,
formation transmissibility, and closure time. The relationship of these parameters is provided in the
following equations using the radial-flow time function, FR.
(1)

(2)

Here tc is the time to closure with time zero set as the beginning of pumping (since diffusion is started),
Pr is the initial reservoir pressure, and mR is functionally equivalent to the Horner slope for conventional
well testing.
Thus, a Cartesian plot of pressure versus the radial-flow time function yields reservoir pressure from
the y-intercept and the slope (mR) that permits determination of transmissibility:
(3)

In Eq. 3, k, h, and m are expressed in oil field units, tc is in minutes, and Vi is the injected volume in
bbl (all other equations are either dimensionless or in consistent units). Before the radial-flow period
occurs, the after-closure pseudolinear flow can sometimes be identified. The after-closure pseudolinear
flow analysis was adapted from the heat transfer analysis of Carslaw and Jaeger (1959). They considered
a semi-infinite body whose surface was held at a constant temperature relative to its surroundings,
followed by insulation of the surface and then thermal decay ensuing. For heat transfer and reservoir
transient behavior, temperature and pressure are analogous. If one assumes the pressure in the fracture is
essentially constant during injection (as is typically observed), the pressure decline after closure behaves
as the thermal decay and, in the absence of spurt loss, can be expressed by the reservoir analog of the heat
transfer problem. The linear-flow time function is FL.
(4)

(5)

Then the coefficient or Cartesian-plot slope is

(6)

where Ct is total leakoff coefficient and ct is total compressibility. From Eqs. 5 and 6, the pressure
decline during the linear-flow period can be written in terms of reservoir diffusivity, storage, total fluid
loss coefficient, and closure time as follows:
(7)

When spurt loss exists, the initial linear-flow behavior follows a different time behavior but eventually
conforms to the behavior of FL (t, tc). After this behavior is established, the linear-flow slope analysis can
SPE-177027-MS 7

be used, in conjunction with reservoir parameters and the preclosure decline analysis to determine the
magnitude of spurt loss (Nolte et al. 1997).
Fracture half-length is determined from the time of transition from linear to radial flow (Nolte et al.
1997). The fracture length determined from this method can be compared to that determined from the
conventional preclosure analysis. Thus the ACA can be used as a quality check on the fracturing and
reservoir parameters used in the preclosure analysis (synergy between the pre- and post-closure periods).
Fig. 2 shows the ACA plot with the different flow periods that can be identified (linear, transition, and
radial) when plotting ⌬P (P(t) - Pr) against the 1/FL2 function. Tp is the dimensionless parameter time that
will be explained further in this paper.

Figure 2—Plot for ACA with the different flow regimes based on Nolte’s FL(tc/t) function for Tp 0.05

After-Closure Analysis Simulation and Pressure Matching Process

After-closure analysis (ACA) is modeled by using a cell-based simulator, which superposes in space and
time at the wellbore all the diffusive pressure response created by each elemental volume of leakoff in
each cell along the fracture length from the start of the propagation until closure and thereafter. The
simulator response depends on the fluid efficiency, spurt, and a dimensionless time Tp for ACA (Nolte et
al. 1997), similar to the dimensionless time used in PTA for a hydraulically fractured well, which
characterizes the ratio between the diffusive front distance from the propagating fracture and the fracture
length (Eq. 8).
(8)

The pressure disturbances induced by fluid loss disperse or diffuse into the reservoir at a rate governed
by the ratio of the mobility k/␮ and storage A ct (i.e., the reservoir diffusivity k/␸␮ ct) expressed in terms
of the permeability k, porosity A, viscosity ␮and total compressibility ct. The resulting pressure patterns
created by the disturbances depend also on the rate of fracture propagation, which is equivalent to a
diffusion rate of L2/ t, being L the propagating fracture length at shut-in. The relative magnitude or ratio
of the diffusion rates determines the shape of the pressure patterns and defines the dimensionless time Tp.
Tp is a constant for low-efficiency fluid (Economides and Nolte 2000). The set of solutions for different
Tp values is incorporated in proprietary software that enables the analyst to match after-closure real data
and eliminates the need to have a uniquely well-defined flow regime. Real pressure data can be matched
8 SPE-177027-MS

with the solutions type curves from the late linear-flow period to a time before the true radial-flow regime
is fully developed (transition period between linear and radial flow). This is the case for the data presented
further in this paper. The type curve matching method also provides the ratio between the reservoir leakoff
coefficient to the total leakoff coefficient and, therefore, an insight into the leakoff mode that can be
classified into three types: reservoir control, filter cake control, or enhanced leakoff possibly from
activation of fissures. The plot in Fig. 2 shows the ACA plot for the pressure decline with Tp ⫽ 0.05 and
negligible efficiency that was created by using the ACA simulator mentioned before. The type curve for
this simulation allows computing the transition period between linear and radial flows. It must be noticed
that if the after-closure data show radial flow, then classical PTA methods (injection falloff, reciprocal of
buildup) or the impulse test solution are sufficient to obtain the correct reservoir parameters (Gu et al.
1993). Nevertheless the shut-in time must be taken at closure when a hydraulic fracture is created and not
at the end of pumping, although in most cases when the efficiency is very low (⬍ 10% is the ideal
condition for ACA), pumping time (tp) and closure time (tc) can indeed be considered equivalent.

Case Study
The area of study in this paper is the Sacha field that is geographically located in the Oriente in Ecuador
(Fig. 3).

Figure 3—a) The Oriente basin and Sacha field geographical location in Ecuador. b) Stratigraphic column

Geology and Reservoir Description

The Hollin and Napo structures are the principal productive reservoirs. Depths range from 9,000 to 11,000
ft true vertical depth (TVD), reservoir pressures from 2000 to 4,300 psi, oil gravity is from 20° to 30° API,
and the gas/oil ratio (GOR) varies from 50 to 100 scf/STB. Fig. 3 also shows the general stratigraphy of
the Oriente basin.
SPE-177027-MS 9

The Sacha field produces oil mainly from the Hollin formation. It is composed of fluvial and deltaic
sedimentary deposits and is subdivided into the Lower Hollin (LH) and Upper Hollin (UH). The UH is
a thinner formation with poorer petrophysical properties than the LH, and both formations are connected
in some areas (pressure support from the LH aquifer).
The uppermost section of UH has permeability from 10 to 100 md with reservoir pressure varying from
2000 to 4300 psi. The advance of the aquifer into the LH formation has forced the operator to abandon
the LH reservoir in many wells where the two formations are separated by a 10- to 30-ft shale interval.
Although the UH formation is still able to produce oil with relatively low water cut, the relatively low
permeability and formation damage makes it marginal for oil production in many cases and stimulation
is required. Stimulating the UH reservoir requires fracturing without growing into the LH (aquifer). Fig.
4 shows a typical log profile for these two formations and a preliminary fracturing geometry in the UH
for the Sacha-258 well. From the geomechanic standpoint, the Oriente basin can be considered a relaxed
or ⬙normal faulting⬙ environment. Fig. 5 shows the stress regime; a normal fracture gradient is expected
in this area.

Figure 4 —Sacha-258 well log data and preliminary fracturing geometry simulation
10 SPE-177027-MS

Figure 5—Stress regime of the Oriente basin; normal faulting is the predominant regime

The first step for production enhancement through hydraulic fracturing was the candidate selection
process. The workflow for this process includes verification of reservoir continuity, a petrophysical
evaluation, and estimation of recoverable reserves. The most typical workover operation consisted of
isolating the LH formation (aquifer) with a squeeze cement job and then stimulating the UH reservoir. The
stress profile for hydraulic fracturing treatments was calculated based on petrophysical data because sonic
logs or geomechanical models were not available. Low-viscosity fracturing fluids were used for the
fracturing treatments. The unique fluid rheology contained the fracture growth, and there was no growth
into the lower water zone (LH formation).

Fracturing Fluid Laboratory Testing and Rheology

Two different fracturing fluids were used for the fracturing campaign in Sacha field. The conventional
guar-based fluids and the fluids based on viscous disproportionate permeability modifiers (VDPM).
Historically, fracturing has not been considered to enhance production where the pay zone is close to the
oil-water contact (OWC) or a lateral aquifer is suspected. But with the increased application of hydraulic
fracturing in mature fields, interest has grown for including a disproportionate permeability modifier
(DPM) or relative permeability modifier (RPM) in these treatments to mitigate the water production after
fracturing. By adjusting the concentration and ratio of the polymers and the concentration of the enhancer,
engineers can optimize the fluid viscosity for any given application. Also, fluid-loss additives can be used
during the fracturing pad to improve fluid efficiency. The fracturing fluids used in the case study had low
concentrations of an RPM and a guar polymer that creates a VDPM. The RPM reduces the effective
permeability to water in the fracture faces, and the relatively low viscosity (adjustable) of the fluid limits
the net pressure and the risk of fracture height growth without compromising proppant transport. Fig. 6
and Table 1 show the typical rheology of the VDPM-based fracturing fluids used in the Sacha field. Fig.
7 shows the rheology of the VDPM fluid for a typical ramp-up or ramp-down with a laboratory standard
rheometer.
SPE-177027-MS 11

Figure 6 —VDPM fracturing fluid rheology at 100 sec-1 (shear rate) and 225 (F bottomhole static temperature (BHST). Average viscosity
varies from 200 to 100 cP. Bottom scale in plot is time (hh:mm:ss)

Table 1—VDPM– based fracturing fluid rheology data from laboratory rheometer

Figure 7—Plot from Table 1 data. a) Cartesian scale. b) Log-log scale

From the rheology experiments, shear stress is seen as almost independent of the rotational speed
(shear rate). It can be considered as having a Bingham plastic behavior with a yield stress equal to the
mean level of stress measured but with a very low plastic viscosity. This behavior is not common although
some mixed metal hydroxide (MMH) drilling fluids exhibit similar behavior. Finally, Fig. 8 shows the
typical rheology of the typical guar-based fracturing fluid.
12 SPE-177027-MS

Figure 8 —Guar-based fracturing fluid rheology at 100 sec-1 (shear rate) and 220(F (BHST). Average viscosity varies from 200 to 100
cP. Bottom scale in plot is time (hh:mm:ss)

Fracture Pressure Analysis, Pressure Transient Analysis, and Rate Transient Analysis
A detailed reservoir and production analysis was performed for the Upper Hollin (UH) formation on well
Sacha-258. Then hydraulic fracturing, which had not been performed before in Sacha field, was
successfully implemented, resulting in a four-fold increase of the well’s productivity index after the
treatment (Fig. 9). This was the most effective stimulation performed in Sacha field from the ⬙sustained
production⬙ standpoint (Fig. 10).

Figure 9 —Productivity index before and after fracturing for Sacha-258

SPE-177027-MS 13

Figure 10 —Sacha-258 well production before and after fracturing

Fracture Pressure Analysis (FPA) A detailed FPA of all the treatments performed in Sacha-258 well
is provided in the Appendix B. Fig. 11 shows the calibration injection/decline and fracturing treatment
performed in Sacha-258 well, both of them with wellhead pressure (WHP) sensors. From this test
performed with the fracturing fluid itself fluid efficiency and reservoir transmissibility were determined.
The fluid efficiency was based on the G-plot and the reservoir transmissibility was obtained from the
after-closure analysis (ACA) plot taking into account the average reservoir pressure in this area from
offset wells (Fig. 12) to perform the pressure match in the ACA plot.
14 SPE-177027-MS

Figure 11—Sacha-258.fracturing charts. a) Calibration injection and decline treatment. b) Hydraulic fracturing treatment
SPE-177027-MS 15

Figure 12—Pressure decline analysis with G-function plot. b) ACA plot with permeability calculation from transmissibility estimate for
Sacha-258 well

The G-function plot shows an almost ideal behavior during pressure decline (Fig. 12). No height
growth is observed in this plot, which is important since we do not want to invade the Lower Hollin
formation (aquifer) with the fracture. Pcl is closure pressure (pc) and Eff is fracturing fluid efficiency in
Fig. 12a. This is not surprising as this depleted reservoir shows a decrease of in-situ stress proportional
to 2␩ ⌬ pR (where ⌬pR is the variation in reservoir pressure) and 2␩ is a constant (poroelastic coefficient).
The initial reservoir pressure in this reservoir was 4,300 psi whereas current average reservoir pressure is
about 2,500 psi from the PTA further in this paper. Fig. 12b shows the ACA, a more detailed analysis is
presented in Appendix B.
On the other hand, the pc was also determined by applying the ⬙Extract Tc⬙ algorithm (Fig. 13) that uses
post shut-in data to determine the closure time from the reservoir’s transient behavior after the fracture
closes (Appendix A). The values of pc obtained by the two methods are very close each other: 5,680 psi
from G-plot and 5,725 psi from the extract Tc algorithm.
16 SPE-177027-MS

Figure 13—Extract Tc plot. This is another algorithm for independent closure time estimate(and pc) based on after-closure linear flow
evaluation. For reference, see Appendix A

Pressure Transient Analysis (PTA) A buildup test was performed on the Sacha-258 well after the
fracturing treatment. (Fig. 14). Permeability was estimated in 60 md from this buildup (using the same pay
height and viscosity data as in the ACA).

Figure 14 —Pressure buildup for the Sacha-258 well

SPE-177027-MS 17

Rate Transient Analysis (RTA) Reserves by decline curve analysis (DCA) were compared for the cases
before and after the treatment by using the dynamic material balance time (DMBT) concept (Mattar and
Anderson 2005). Fig. 15 shows that after fracturing, the reserves increased more than 10 times, which
represented a remarkable opportunity for the field development plan in UH formation.

Figure 15—PDA with DMBT for Sacha-258

Apart from the DCA with DMBT, the RTA was performed with Agarwal-Gardner type curves
(Agarwal et al. 1999) and the production history match was done by using a gridless reservoir evaluation
software tool (GREST) involving semi-analytical methods (Zhou et al. 2013) for the approximation of the
deconvolved pressure-rate function (Fig. 16).
18 SPE-177027-MS

Figure 16 —RTA and history match with the GREST for the Sacha-258 well

Closing the loop among FPA, PTA, and RTA FPA (ACA), PTA, RTA were compared for this well
obtaining a reliable value of reservoir’s transmissibility and pressure. Fig. 17 shows a summary of the
permeability calculation based on the three methods.

Figure 17—Transmissibility estimation comparison between ACA, PTA, and RTA for the Sacha-258 well

Fracturing Campaign Summary

The initial three-well fracturing campaign in Sacha field applying FPA methodology showed outstanding
results and created a new opportunity for development of the UH formation. On average, the oil
production increased threefold (or 300 BOPD per well on average), and the water cut was not substantially
modified after the fracturing treatment (Fig. 18).
SPE-177027-MS 19

Figure 18 —Production results of the initial campaign in Sacha field before and after the fracturing treatment

Conclusions
● Fracture pressure analysis (FPA) added value in Sacha field unlocking reservoir’s potential
through the implementation of hydraulic fracturing in UH formation with low-viscosity fluids in
spite of the proximity to the aquifer.
X Production increased on average 300 BOPD per fractured well in the first campaign without
contacting the lowermost aquifer separated by a 10- to 30-ft shale from the pay zone.
X No height growth was detected in the lower wet zone (hydraulically fractured wells did not
increase water cut after the fracture job).
X Another algorithm, independent from G-plot and based on the after-closure linear flow regime
identification through pressure’s second order derivatives, was implemented for closure time
(and closure pressure) estimation with consistent results.
● Reservoir parameters were compared among FPA, pressure transient analysis (PTA), and rate
transient analysis (RTA) and they were shown to provide reliable and similar values for trans-
missibility and pressure in UH formation.
● A remarkable increase in the reserves by decline curve analysis (DCA) was identified after
hydraulic fracturing (up to more than 10 times in some cases).
● A bottomhole pressure (BHP) sensor would give more information on the Nolte–Smith plot for
improving the tip screenout technique in future treatments to improve fracture conductivity and
production.

Acknoledgments
The authors would like to thank Operaciones Rio Napo CEM and Schlumberger for giving permission to
publish this paper. The authors would also like to thank K.G.Nolte for the clarification of the Extract Tc
method.
20 SPE-177027-MS

SI Metric Conversion Factors

bbl ⫻ 1.589 873 E – 01 ⫽ m3
ft ⫻ 3.048 E – 01 ⫽m
°F (°F – 32)/1.8 ⫽ °C
in. ⫻2.54 E ⫹ 00 ⫽ cm
cp ⫻ 1.0 E – 03 ⫽ Pa•s
gal ⫻ 3.785 412 E – 03 ⫽ m3
md ⫻ 9.869 233 E – 04 ⫽ m2
psi ⫻ 6.894 757 E ⫹ 00 ⫽ kPa

References
Abousleiman, Y., Cheng, A.H-D. and Gu, H. 1994. Formation Permeability Determination by Micro
or Mini-Hydraulic Fracturing, J. Energy Resour. Technol 116 (2), 104 –114. 01– June. doi:
http//doi:10.1115/1.2906014
Agarwal, R. G., Gardner D. C., Kleinsteiber, S. W. et al. 1999. Analyzing Well Production Data Using
Combined-Type-Curve and Decline-Curve Analysis Concepts. SPE Res Eval & Eng 2 (5):
478 –486. SPE-57916-PA. http://dx.doi.org/10.2118/57916-PA.
Behrmann L.A. and Nolte K.G. 1998. Perforating Requirements for Fracture Stimulations. Presented
at the SPE Formation Damage Control Conference. Lafayette, Louisiana, USA, 18 –19 February.
SPE-39453-MS. http://dx.doi.org/10.2118/39453-MS.
Carslaw, H.S. and Jaeger, J.C. 1959. Conduction of Heat in Solids, second edition. Great Britain:
Oxford University Press.
Economides, M. and Nolte, K.G. 2000. Reservoir Stimulation, third edition. s.l.: John Wiley & Sons,
Chapter 9.
Fidan, E., San Prasad, P., Darous, C. et al. 2014. Comprehensive Multiple Fracture Pressure Analysis
Tests to Appraise and Develop Unconventional Organic-Rich Carbonaceous Shales in North
Kuwait. Presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands,
Texas, USA, 4 – 6 February. SPE-168585-MS. http://dx.doi.org/10.2118/168585-MS.
Fredd, C.N., McConnell, S.B., Boney, C.L. et al. 2000. Experimental Study of Hydraulic Fracture
Conductivity. Presented at the SPE Rocky Mountain Regional/Low-Permeability Reservoirs
Symposium and Exhibition, Denver, Colorado, USA. 12–15 March. SPE-60326-MS. http://
dx.doi.org/10.2118/60326-MS.
Gu, H., Elbel, J.L., Nolte, K.G. et al. 1993. Formation Permeability Determination Using Impulse
Fracture Injection. Presented at the SPE Production Operations Symposium, Oklahoma City,
Oklahoma, USA, 21–23 March. SPE-25425-MS. http://dx.doi.org/10.2118/25425-MS.
Ispas, I.N, Britt, L.K., Tiab, D. et al. 1998. Methodology of Fluid Leakoff Analysis in High-
Permeability Fracturing. Presented at the SPE Formation Damage Control Conference, Lafayette,
Louisiana. 18 –19 February. SPE-39476-MS. http://dx.doi.org/10.2118/39476-MS.
Mattar, L. and Anderson, D. 2005. Dynamic Material Balance (Oil or Gas-in-Place Without Shut-ins).
Presented at the Canadian International Petroleum Conference, Calgary, Alberta, Canada, 7–9
June. PETSOC-2005-113. http://dx.doi.org/10.2118/2005-113.
Nolte, K G. 1979. Determination of Fracture Parameters from Fracturing Pressure Decline. Presented
at the SPE Annual Technical Conference and Exhibition, Las Vegas, Nevada, USA, 23–26
September. SPE-8341-MS. http://dx.doi.org/10.2118/8341-MS.
Nolte K.G. and Smith M. B. 1981. Interpretation of Fracturing Pressures. J Pet Technol 33:
1767–1775. SPE-8297-PA. http://dx.doi.org/10.2118/8297-PA.
SPE-177027-MS 21

Nolte K.G., Maniere J.L., and Owens K.A. 1997. After-Closure Analysis of Fracture Calibration
Tests. Presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas,
USA, 5– 8 October. SPE-38676-MS. http://dx.doi.org/10.2118/38676-MS.
Weng X., Pandey V., and Nolte K.G. 2002. Equilibrium Test—A Method for Closure Pressure
Determination. Presented at the SPE/ISRM Rock Mechanics Conference, Irving, Texas, USA,
20 –23 October. SPE-78173-MS. http://dx.doi.org/10.2118/78173-MS.
Rozo R., Paez J., Rojas A., Milne A., Soler D. 2007. Combining Acid- and Hydraulic-Fracturing
Technologies Is the Key to Successfully Stimulating the Orito Formation. Presented at the SPE
Hydraulic Fracturing Technology Conference, College Station, Texas, U.S.A, 29 –31 January.
SPE-104610-MS. http://dx.doi.org/10.2118/104610-MS.
Zhou W., Samson B., Krishnamurthy S, et al. 2013. Analytical Reservoir Simulation and Its
Applications to Conventional and Unconventional Resources. Presented at the EAGE Annual
Conference & Exhibition incorporating SPE Europec, London, UK, 10 –13 June. SPE-164882-
MS. http://dx.doi.org/10.2118/164882-MS.
22 SPE-177027-MS

Appendix A
Background for Closure Time (Tc) Estimate or ⴖExtract Tcⴖ

⬙Extra tc⬙ is the name of the algorithm, which extracts the time of the linear-flow beginning from pressure decline data. This
analysis and its results are completely independent of any preclosure analysis (e.g., using G-plot). The after-closure linear-flow
regime initiation is associated to the time of the fracture closure.
This algorithm uses post-shut-in data to determine or ⬙extract⬙ the closure time (tc) from the reservoir’s transient behavior
after the fracture closes. When the fracture closes, the fluid-loss stops and this ⬙reservoir shut-in⬙ event becomes embedded in
the data during the early-after-closure data. This embedment time is approximately the period between starting injection (t ⫽
0) to fracture closure (t ⫽ tc), that is
(A-1)

The basis for the extract Tc algorithm is that, for an injection test consistent with the mini falloff (MFO) design
recommendations, the early-after-closure pressure data are in pseudolinear flow that can be expressed (Economides and Nolte
2000) as
(A-2)

This after-closure linear-flow behavior occurs even when the flow regime during fracture propagation is beyond pseudo-
linear flow (e.g., into transitional flow conditions as desired for a mini falloff test to characterize the reservoir properties). The
algorithm uses second-order derivatives (i.e., the derivative of a derivative) to remove the dependence on the constant and Pi,
while isolating (for extracting) the a priori unknown value of tc within Eq. A-2. It computes at every time step the ratio of
second derivative d2p/d2t to the first derivative dp/dt of recorded data, which provides a value of the reference time (tc)
embedded in the linear flow function. Note that the equation, and hence the algorithm, does not depend on the shut-in time.
Application of second-order derivatives of pressure, as for the extract Tc algorithm, will generally require extensive and
intelligent filtering. This is particularly the case for the low-quality pressure data that is typically obtained from standard
high-pressure gauges used on treating lines.
The after-closure linear-flow regime, assumed by the algorithm begins with the ⬙reservoir shut-in⬙ event as the fracture
closes. However this regime’s clock-time duration, for extracting closure, decreases inversely to the dimensionless time Tp
because the duration of this Tp for the regime is a fixed value ~ 0.02 (Economides and Nolte 2000, Fig 9.38). As a result the
clock-duration of this after-closure regime (⌬tc, lf) required to extract closure, can be expressed as
(A-3)

and with Eq. 1,

(A-4)

for the prospective, post-closure, clock-time to extract the closure time.

As an example, consider a pumping time of tp ⫽ 10 min having a dimensionless time of Tp ⫽ 0.2 for the fluid-loss-induced
reservoir transient, and a closure time of tc ⫽ 12 min (i.e., closure 2 min after SI):
● Eq. A-1: ⌬tc, emb ~ 12 min
● Eq. A-3: ⌬tc, lf ~ (0.02 / 0.2) (10 min ⫽ 1 min
● Eq. A-4: ⌬tc,extr ⫽ min (1, 12) ⫽ 1 min
This example indicates that extracting tc for propagation in radial flow (Tp ⬎ 1) is practically impossible (e.g., with Tp ~
1, the above example would result in ⌬tc,extr ⫽ ⌬tc, lf ⫽ 0.2 min, or about 10 sec. However, for propagation in linear flow, Tp
⬍ 0.02, (as for a crosslinked gel drilling fluid and with tp ⫽ 10 min) ⌬tc, lf ⬎ 10 min, or ⌬tc,extr becomes ~ tc.
SPE-177027-MS 23

Appendix B
Fracture Pressure Analysis (FPA) for Sacha-258 well

This appendix describes the method to confirm the consistency of the analysis using all the recorded treatments (or cycles) for
Sacha-258 well (Fig B-1 and Table B-1). Knowledge of fracture mechanics and pressure diffusion are required to interpret and
explain deviations from ideal behaviors, which are often observed in real cases.

Figure B-1—Pumping Treatments (cycles) performed on Sacha -258 well

Table B-1—Treatments and fluids related to Figure B-1

Cycle 1 MFO (brine)
Cycle 2 Viscous acid treatment
Cycle 3 Injection test (brine)
Cycle 4 Injection test (brine)
Cycle 5 Calibration injection/decline treatment (Crosslinked gel)
Cycle 6 Propped fracturing treatment (Crosslinked gel)

The cycle 1 (Fig. B-2 and Fig. B-3) did not have enough pressure decline time to get a reliable match because of the
falling-liquid level effect (depleted reservoir). Then, the after-closure analysis (ACA) match is selected to be consistent with
data from the buildup test available and, closure pressure (pc) was obtained from later data at 5,700 psi (cycle 5).
24 SPE-177027-MS

Figure B-2—MFO (Cycle 1)

Figure B-3—a) G-plot, b) ACA-plot for Cycle 1

In practice, the MFO test should be designed so that the dimensionless time at shut-in Tp is greater than 0.5 to 1 (pure radial
flow occurs when Tp ⬎2). In other words, the diffusion front perpendicular to the fracture reached rapidly the length of the
fracture and the boundary geometry results quasi-radial. Now, after a calibration treatment with crosslinked gel, even in a
highly-diffusive case, the fracture length is considerably larger than the one created during the MFO test (typically done with
brine) and therefore a period of after-closure reservoir linear flow is common: At shut-in, the pressure front is highly elliptical
or close to linear and allows to compute an estimated of the start of linear flow, which corresponds to the fracture closure
(Appendix A).
In this case, this reservoir’s mobility is highly enough to satisfy the condition for a MFO test but a short injection test can
only provide correct reservoir pressure (Pr) in a virgin layer unaffected by the nearby wells (producers or injectors). On the
other hand, in a depleted reservoir under active production, the Pr is not always uniform. Due to this effect, differences between
Pr from short-term injection tests (ACA) and long-term buildup tests (PTA) are expected. Reservoir pressure transient effect
are reduced on the first cycle. In order to improve the test quality and decline time, a lower rate and larger volume treatment
SPE-177027-MS 25

would be preferable by filling up and pumping at 5 bpm for at least 5 min. The rate could also be reduced to 3 bpm to decrease
the hydraulic fracture length that would produce a shorter linear-flow period and then an earlier radial-flow regime for
reservoir’s transmissibility evaluation. In addition, to increase the recording time, a lower density would be required in the
wellbore fluid if possible (diesel). A bottomhole pressure (BHP) gage would not solve this drawback since, after the pressure
vanishes on surface, a falling-liquid level starts and more fluid is injected into the matrix. Only a bottomhole shut-in will
provide the correct information for a complete ACA under this depletion condition.
The cycle 2 corresponds to a viscous acid treatment that was pumped as a prepad ahead of the main propped fracturing
treatment (Rozo et al 2007). This treatment gives the opportunity to investigate how to adjust the range to use for a nonconstant
injection rate case. Then, two scenarios can be considered:
1. Using the full pumping treatment range for the analysis (Fig. B-4). In this case the ACA match provides a Pr ⫽ 2410
psi and a transmissibility ⫽ 400 md.ft/cp. This scenario results in a Pr lower than during the first injection and is less
consistent. (Fig. B-5)

Figure B-4 —Viscuos acid system (cycle2)

Figure B-5—a) G-plot, b) ACA-plot for Cycle 2 in Fig B-4

26 SPE-177027-MS

2. Using only the last pumping period at fracturing rate (above 8 pm) after a short shut-in (Fig. B-6). This scenario
would provide a pressure more consistent with other estimates. Pr ⫽ 2690 psi and transmissibility ⫽ 400 md.ft/cp
(Fig B-7).

Figure B-6 —Viscous acid system (cycle2) with the last pumping period selected for ACA

Figure B-7—a) G-plot, b) ACA-plot for Cycle 2 with the last pumping period selected for ACA

The cycles 3 and 4 correspond to additional injection tests that are not essential for the FPA. As for cycle 1, the volumes
and rates are not well-adapted to the reservoir’s condition in order to obtain a unique ACA. Only a few points can be used for
the ACA which makes the results uncertain. On the other hand, these tests are affected by prior injections.
In the cycle 5 (Fig. 11a), both the G-plot and the extract Tc algorithm provide almost the same value of pc ~ 5700 psi (Fig
12 and Fig. 13). Indeed, the G-plot is almost ideal and the extract Tc algorithm indicates the presence of a strong after-closure
linear to transitional flow regime. The value of pc is confirmed later from cycle 6 pressure decline. If the transmissibility is set
to 400 md.ft/cp (from buildup test data), the corresponding Pr is 2666 psi which is consistent with previous cycles (Fig. 12b).
Nevertheless a more accurate visual match provides a transmissibility of 554 md ft /cP and Pr of 2,971 psi. One potential
explanation for this scenario would be that the cycle 5 opened a thicker the pay zone (Fig. B-8).
SPE-177027-MS 27

Figure B-8 —ACA-plot with a different match for Cycle 5

The cycle 6 (Fig. 11b)correspond to a small propped fracture treatment of 11,000 lbs of proppant. The G-plot (Fig. B-9a)
shows some height growth behavior and pc (top of G-slope curve hump) corresponds almost exactly to the same pc obtained
during the calibration injection/decline (cycle5).

Figure B-9 —a) ACA match using buildup data. b)ACA-plot with a different match for cycle 6

The ACA match presented is nonunique but provides a consistent estimate of the reservoir properties: Pr ⫽ 2656 psi and
transmissibility ⫽ 404 md.ft/cp (Fig. B-9b). Fig. B-10 shows the extract Tc analysis for cycle 6.
28 SPE-177027-MS

Figure B-10 —Extract Tc analysis for cycle 6

A summary of the ACA results is presented in Table B-10 showing the robustness of the method despite of the deviations
from ideal conditions. Unfortunately, as none of the pressure declines showed radial flow, external information must be used
to perform a consistency analysis. This post-mortem analysis used the result from the buildup test (transmissibility of 400
md.ft/cp) to verify the corresponding reservoir pressure estimate after each cycle. In general, the after-closure data match was
nonunique due to the short recording time. All of the cycles did not show a clear pc estimate on G-Plot (except the cycle 5)
and, the after-closure data was very short for a reliable analysis although consistent with a transitional to radial behavior.

Table B-2—Summary of the ACA, PTA and RTA. Permeability has been calculated based on transmissibility using the same viscos-
ity and pay height for all the cases