CSUG/SPE 136565

Integrated Analysis Combining Microseismic Mapping and Fiber-Optic

Distributed Temperature Sensing (DTS)
E.H. Holley, SPE, U. Zimmer, SPE, M.J. Mayerhofer, SPE, E. Samson, SPE, Pinnacle Technologies

Copyright 2010, Society of Petroleum Engineers

This paper was prepared for presentation at the Canadian Unconventional Resources & International Petroleum Conference held in Calgary, Alberta, Canada, 19–21 October 2010.

This paper was selected for presentation by a CSUG/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
Mapping the microseismic activity during a hydraulic treatment is widely used to determine the geometry of the stimulated
fracture network. Microseismic maps provide reliable information on the development of fracture symmetry, half-length, azimuth,
width and height, and their dependence on the treatment parameters and reservoir characteristics. Beyond that, these fracture
geometries are used to better understand fracture modeling and even production characteristics.
Fiber-optic-based distributed temperature sensing (DTS) arrays provide almost immediate updates of the near-wellbore
temperature distribution in approximately one-meter intervals. In injection treatments, the near-wellbore temperature distribution
can be used to determine isolation effectiveness, the relative amount of fluid each perforation cluster takes, fracture initiation
points, and effective fluid diversion. In production analysis, DTS measurements can quantify production rates from each
perforation interval, crossflow rates while shut-in, and fluid types recovered from each perforation interval.
The detailed near-wellbore results available through DTS coupled with the far-field geometry acquired through microseismic
mapping provide an accurate picture of the completion effectiveness. Microseismic mapping results often show adequate
resolution over a large area but lack the fine resolution that would allow it to identify near-wellbore effects in the meter range.
When modeling and interpreting the treatment geometry obtained by the microseismic-event distribution, it is important to include
the correct near-wellbore effects, which are readily accessible through DTS measurements. Combining the two diagnostic tools is
valuable for real-time decision making, post-treatment analysis, and production analysis to assess the completion effectiveness.
Incorrect assumptions about perforation breakdown, fracture-initiation points, interval isolation, or limited-entry effectiveness
can lead to misinterpretations of the microseismic results. Using both diagnostic tools provides firm answers to the overall
completion effectiveness. This paper focuses on three distinct aspects of combining the analysis of microseismic mapping and
DTS. The first is the real-time aspect, wherein real-time decisions and adjustments are made during the fracture treatment with the
objective of manipulating the results towards the desired outcome. The second is using both tools to perform more accurate post-
fracture analysis, including calibrated fracture modeling, entry effectiveness, correct interval spacing, and stimulated reservoir
volume (SRV) analysis. The third area covered is combining these diagnostic tools with a production analysis, which is acquired
through analysis of the temperature data.

Introduction
Microseismic mapping has become a standard method to determine the geometry of the fracture network created during hydraulic
treatments. Although these results have greatly contributed to a better understanding of the propagation of fracture networks and
the effectiveness of changes in the treatment schedule, microseismic monitoring faces some challenges when it comes to the
detailed interpretation of fluid movements in the immediate vicinity of the treatment well. Depending on the relative position of
the observation well, the resolution of the microseismic results might be a few meters, which is not sufficient to answer detailed
questions regarding points of fracture initiation from the wellbore. The generation, processing, and visualization of microseismic
events also might be delayed a few minutes, which results in a delayed interpretation as well. Fiber-optic-based temperature
sensing provides a constant 1-m resolution along the full length of the wellbore, and because only minimal processing is involved,
2 CSUG/SPE 136565

the results are available every few seconds, making it an excellent choice for assessing the near-wellbore fluid movement in almost
real time.
An important factor for the correct interpretation of these results is the accurate interpretation of the fracture treatment results.
It is important to consult the fracture treatment charts when observing real-time DTS changes during the job. This would allow for
correlating DTS results to fracture responses. Often times surface pressure responses or changes in proppant concentrations will
reveal changes in the temperature signature along the wellbore. Relating pressure responses, fluid conditions, as well as rate and
proppant responses to the diagnostic responses of DTS and microseismic-event distributions is especially critical for correct real-
time changes of the treatment schedule.

Microseismic Monitoring
During hydraulic-fracture treatments, new fractures are created in the rock. Where existing critically stressed fractures already
exist in the rock, the injections facilitate slippage of these fractures, which in turn creates seismic waves. These seismic signals can
be recorded using geophone arrays deployed in the vicinity (Fig. 1). By identifying the P- and S-wave travel times and using an
underlying velocity model, the location of the event can be calculated. Over the course of a hydraulic treatment, often thousands of
individual events are created and located on a map. The event distribution describes the general fracture geometry, azimuth, half-
length, width, and height. This in turn can be interpreted in terms of zonal isolation, stage overlap, and formation containment.
Another analysis that can be completed using these maps is the determination of fracture geometry to constrain and calibrate
fracture models. The created fracture geometry can also be directly correlated to production results, which assists in better
interpretation of the completion effectiveness of the well.

Fig. 1—Microseismic mapping of hydraulic treatments.

Distributed Temperature Sensing (DTS)

Fiber-optic-based DTS is a relatively new technique that has been applied to hydraulic-treatment monitoring only in recent years.
It is based on the fact that the local temperature in the wellbore is changed at positions where fluid enters the formation (Fig. 2).
The temperature in the immediate vicinity of the borehole is measured using a customized fiber-optic cable that is installed based
on the completion design. These fiber-optic cables can operate successfully in ambient temperatures up to 300oC for several years
and have the capacity to provide a temperature reading every 1- m along their length within a few seconds. This temperature
profile can be used to determine the fluid movement within the wellbore and the points of entry into the formation as the fluid exits
the wellbore.
CSUG/SPE 136565 3

In this case (Fig. 2) the DTS measurement represents a trace after pumping has stopped. The trace below signifies warm back
data and not injection. It has become a common interpretation (Huckabee 2009) that warm back data can indicate where major
fractures have been created with respect to fluid volume. A certain area of the reservoir that returns to geothermal temperature at a
slower rate after injection most likely took a majority of the treatment fluid. This will be discussed in greater depth during the case
study example. It is seen below that all intervals initially warm back at a consistent rate indicating that all cooler intervals took a
significant amount of fluid.

Fig. 2—Correlating DTS-temperature readings with local fluid flow.

Integrated Scenarios
Although microseismic mapping and DTS readings provide valuable information themselves, the combination of both methods
provides a much better description of the completion effectiveness. The DTS measurements provide very detailed near-wellbore
information, whereas the microseismic map provides information about the fluid movements farther into the formation. The
following few scenarios and their respective microseismic and DTS response and interpretation will show the combined
interpretation. Because public data for demonstrating these scenarios is currently extremely limited, drawings are used to illustrate
the different effects. More real-data case studies showing these scenarios will become available to the public soon. These drawings
are based on extensive experiences in using both technologies individually and the developing understanding of using these
technologies in simultaneous operations.
The drawings represent a complete microseismic map of the geometry on the left with the DTS trace during injection on the
right. The final example provided is actual data that was provided from a well in which simultaneous microseismic results and
DTS results were acquired. The drawing scenarios below are only exaggerated traces of temperature responses that would
typically be procured. Through our real-time experience thus far the fluid temperature exiting the wellbore is typically 3˚-5˚ C
cooler from the wellbore temperature during fluid injection. This difference allows clear visibility of fluid exit from the wellbore.
Clearly the underlying assumption being that the geothermal reservoir temperature is much warmer than our injected fluid
temperature.

Scenario 1: Geometry and Fluid-entry Points Behave as Expected

The example in Fig. 3 highlights a case in which all the perforated intervals are taking fluid and the far-field geometry follows
expectations. In this case, the geometry does not appear to be affected by any irregularities in the reservoir. The geometry in this
case also represents a highly connected complex fracture network typically seen in naturally fractured shale plays. Fig. 3 represents
a multiple-perforation completion method in a cased and cemented wellbore. The DTS data shows that breakdown was achieved
across all four perforated intervals, and all four intervals are taking treatment fluid. Isolation in this case is good because there is
no indication of flow behind cement or fluid flow down hole of the packer. During real time, no adjustments would be
recommended because the completion, according to the diagnostic tools, is running according to schedule. The final diagnostic
tool, not represented here, would be to consult the fracture-treatment chart.
4 CSUG/SPE 136565

Fig. 3—Drawing shows fluid entry through all perforations, no isolation issues, and a broad complex fracture network.

The second step in this diagnostic approach is to perform post-fracture analysis using both diagnostic tools. Because a highly
complex fracture network was created in this scenario, stimulated reservoir volume (SRV) analysis would be performed to assess
completion effectiveness. The SRV value uses the microseismic-event distribution to calculate a total volume of rock that was
stimulated. The DTS data provides the evidence that this particular fracture volume was created by all intervals taking treatment
fluid, which might not be the case for other perforation clusters, even in the same well. If different individual stage geometries are
created within the same well, the knowledge of the number and position of the entry points can help clarify if the created geometry
is a function of the reservoir, treatment parameters, or the points of fracture initiation itself. This will help in determining the best
method to create the desired fracture geometry and if the network geometry is a function of treatment parameters, perforation
strategy, proppant concentrations, etc. These finals results could not likely be concluded if only one of the diagnostic tools was
present.
The third and final step of the combined diagnostic analysis discussed in this paper is the addition of the production results
procured from the fiber-optic DTS cable. Having the ability to accurately determine production contribution from each individual
perforation interval within a treatment has clear value. Combining the SRV results obtained from the microseismic mapping with
the near-wellbore injection and individual stage production results procured from the fiber-optic cable provides a concise guideline
to the overall completion effectiveness. The relevance of SRV, fluid-initiation points, treatment parameters, and perforation-
strategy effectiveness can all be directly compared to production results per treatment stage. This provides an excellent foundation
for improving future completions in the same reservoir.
Typically, a case study is developed on a well-by-well basis. Using both of these diagnostic tools in conjunction provides the
ability to have multiple case-study designs per stage within a single well.

Scenario 2: Flow Behind Casing

In this case (Fig. 4), it is noticed that a near-wellbore anomaly might be present in the form of poor cement in the lateral section
outside of the plug-side perforation cluster. Using the microseismic data, it is easy to see that the geometry is not occurring directly
across a perforated interval. This occasionally happens in mapping geometries, and the troubleshooting begins with examining the
location accuracy and reviewing the velocity-model calibration. However, with the presence of the DTS data during the job, it is
easily seen that the fluid is traveling behind casing toward the plug and that fluid is exiting from the perforation cluster nearest the
plug. This knowledge greatly improves confidence in the microseismic-event locations and velocity model. Having a consistent
temperature throughout, as seen in the DTS image in Fig. 4, means there is direct fluid contact with the fiber-optic cable across this
entire lateral distance. The most likely reason for this behavior is an annular space of poor cement in which the fluid travelled.
CSUG/SPE 136565 5

Fig. 4—Fracture initiation during injection away from perforation caused by poor isolation.

In this scenario, the microseismic events clearly show a single point of fracture initiation. The stimulation in this particular case
is creating planar fracture geometry away from any of the perforated intervals. The confidence in the microseismic event-location
accuracy is enhanced by the presence of the DTS fiber-optic cable. Clearly, only one of the perforated intervals in this case is
taking treatment fluid. This provides valuable insight into the completion strategy for this particular stage. Having the
microseismic-mapping data allows the visualization of the unstimulated virgin reservoir area that exists outside of the four
perforated clusters in this stage. With high confidence in the microseismic-event locations, the reasons for the poor entry
effectiveness can be assessed because three of the perforation clusters were not stimulated. Understanding event magnitudes and
event density in this area coupled with treatment information provides some insight into a possible solution to prevent
unstimulated areas moving forward. The knowledge that the fluid clearly accepted a tortuous path through cement instead of a
direct path to the formation also highlights some treatment-chart information. Often, a tortuous path as seen above will not permit
large proppant concentrations to pass to the formation. When larger proppant concentrations get on depth in this environment,
severe pressure response will often occur and the job will be cut short instead of other perforated intervals breaking down to take
fluid. Understanding the reservoir in this way helps to make remedial solutions in real time as both diagnostic tools are monitored.
An interesting assessment to perform on a stage such as the one above is coupling real-time information with the production
analysis. If the entire frac job is put away in a single perforated interval, the fluid volume was not spaced evenly. Comparing
production results from a stage such as this to a stage in which fluid diversion was created more evenly across all perforation sets
is very helpful. It can certainly assist in providing insight into future fracture treatments in this particular reservoir. Placing an
entire designed treatment through a single point of initiation away from the actual perforation is very unlikely unless a nearby fault
is accessed by the fluid. Jobs like this often do not get pumped to completion, given the tortuous nature of the fluid travel
immediately outside of the wellbore. Thus, the final conclusion can be made that the ability to place the treatment was not a
function of the completion strategy, but was a function of the poor isolation at the near-wellbore area.

Scenario 3: Formation Irregularities Dictating Fluid Travel

In the third scenario (Fig. 5), the DTS results indicate good fluid entry through at least three out of the four perforations. From
these results, it would appear that the treatment was successful, given most of the perforation intervals took fluid. In contrast, the
microseismic results provide the far-field picture of the created fracture geometry. Although the event distribution covers most of
the perforations, indicating an agreement with the DTS measurements that these perforations take fluid, the event distribution
farther away from the wellbore clearly shows a high concentration of events in a small linear area. This event distribution is often
associated with a highly naturally fractured, faulted, or depleted part of the reservoir. Often, neighboring wells that have been on
long-term production will affect the direction the fluid will travel in the reservoir because of depletion. The defining factor, with
respect to microseismic events when looking for a naturally fractured or faulted area, is an increase in overall event magnitude or
relative event density.
6 CSUG/SPE 136565

Fig. 5—Fault reactivation during injection is clearly visible in the microseismic results but not viewable with DTS results.

This case highlights results in which the geometry and fluid path are not influenced by poor perforation strategy or treatment
design, but by the reservoir itself. If a highly naturally fractured or faulted area is present, it can dictate the path of fluid migration
and the geometry created. This heterogeneity in the reservoir is often encountered, and unfortunately remedial solutions are few.
Microseismic mapping and DTS can also be used in combination in such a case when it is expected that the fracture growth is
being dictated by a previously treated or produced area of the reservoir. The fracture mapping provides the geometric evidence that
the fracture grew toward a previous well, but the DTS will provide information with regards to the effect this will have on
production in that particular interval. Comparing production results in this way will allow for an improved understanding of
correct well spacing moving forward.

Scenario 4: Stage Isolation

This scenario (Fig. 6) shows diagnostic responses that indicate poor stage isolation. Poor stage isolation can be a function of a
leaking plug or packer, poor cement, or a leak in the casing. Occasionally, it will be recognized that microseismic events will be
produced in a previously treated interval; however, it is difficult to discern the exact reason for the event occurrence. If poor
isolation is suspected, it is difficult to determine if the fluid traveled through a poorly set plug or through a microannulus caused by
poor cement. Depending on event quantities located in a previously treated interval and time between stages, it can also be difficult
to judge if the events are a function of leakoff-slippage events or newly created fracture extension. Having the DTS in place allows
full understanding of the direction of fluid travel in the near wellbore. If this problem is observed in real-time, remedial solutions
aimed to divert fluid to the interval of interest for that particular stage can be assessed.

Fig. 6—Both methods show a potential stage isolation issue. It is much easier to discern with the DTS data.
CSUG/SPE 136565 7

The next step in the diagnostic process is analyzing the information after the fracture has been completed. If fluid placement of
the treatment is diverted to multiple intervals because of poor isolation, the created fracture geometry will most likely be affected
for all intervals involved. Understanding how this change in fluid placement alters far-field geometry, and in turn production of
those intervals, is a valuable asset.
Analyzing the individual stage production with respect to fracture geometry is a key benefit to the entire process. Historically,
production results were acquired on a well-by-well basis and were compared wholly to the created fracture geometry garnered
from microseismic data. With the advent of the DTS technology for real-time and production results, it is now possible to
accurately analyze datasets on a stage-by-stage basis within a single well (Huckabee 2009). Having this ability significantly
improves the case studies available to understand a particular reservoir or completion strategy. Observing each individual stage
within a well as its own case study, with respect to created fracture geometry, points of fracture initiation, individual production,
and treating parameters, provides a significantly better understanding of a given well, as opposed to grouping all this information
together on a well-by-well basis.

Case Study Example

The following examples depict an actual dataset in which microseismic mapping and DTS data were simultaneously acquired
during the fracture treatments. No decisions based on the available diagnostics data were made to deviate from the original job
design while the fracture treatment was being completed. However, when the diagnostic information was reviewed after the
treatment had been completed, the conclusion based only on the microseismic data, with respect to the completion strategy and
reservoir properties, changed drastically when the real-time DTS data was added to the dataset.
This well was completed as a cemented and cased lateral with valve isolation between treatment intervals. There was an
adjacent vertical well to the SW of the Stage 3 perforations that had previously been on production. This depleted region of the
reservoir directly impacted the fracture geometry on that side of the lateral. Poor cement behind the casing played a major role in
effective fluid placement during the fracture. Each stage was analyzed individually with both diagnostic tools.
The images below are all structured in the same format. The image on the left of each figure is a map view of the treatment
lateral with the microseismic-event locations for each particular stage. The image on the right of each figure is the DTS trace taken
during the stimulation treatment. The stars in each image represent the perforation locations for each individual stage.
Unfortunately no production data will be discussed in depth on a stage by stage basis for this particular case study. Actual data
examples with associated production values will be assessed when they are released. The target formation for these stimulation
stages was the Barnett Shale in Texas. This type of environment will typically create highly complex networked fracture
geometries that will reveal some geometric overlap through the microseismic data. Quantifying that overlap in a shale
environment and correlating that to points of fracture initiation at the wellbore will be observed throughout this case study.

Stage 1
Reviewing the microseismic-treatment information for Stage 1 (Fig. 7) first reveals that a complex fracture network was created in
this case. The interpretation of the microseismic mapping data concludes that the fracture appears asymmetric. Looking at the
microseismic and DTS data combined reveals that this entire complex network was created from a very narrow point of fracture
initiation. The total network-fracture width is 280 m, and the fracture fluid is exiting the wellbore with an initiation width of only
15 m. This helps to provide evidence of a relationship between far-field geometric growth as a function of fracture initiation.
Comparisons to the SRV and widths of fracture initiation at the near wellbore can be explored as other stages are reviewed in this
particular well.
8 CSUG/SPE 136565

Stage 1 perf interval

Fig. 7—Stage 1 combined real-time results

For this particular stage, the isolation and point of fracture initiation is fairly well contained within the desired perforation
interval. The treatment chart and DTS information indicate that the completion of this stage was achieved as designed. When
looking at the DTS information on the right of these case-study figures, it can be seen that the entire wellbore up hole from the
desired perforation interval cools. The cooling in the area across from the perforation is caused by direct fluid contact with the
fiber-optic cable. The temperature difference is only a few degrees, but it is still easy to recognize. For Stage 2, the same treatment
design was pumped; however, the same result was not achieved according to our DTS and microseismic information.

Stage 2
The treatment charts for Stages 1 and 2 are essentially identical (Fig. 8). Wellhead rates and surface-treating pressures follow
identical trends for each stage. Because Stage 1 is asymmetric, it would suggest that Stage 2 is also asymmetric. Starting in this
stage, the influence from the previously treated and produced well becomes noticeable. The total fracture-network width on the
northeast side of the lateral or Stage 2 is 320 m, while the network width on the southwest side of the lateral is only 270 m. This
narrowed network width could be attributed to the depleted reservoir near that side of the well. Because the secondary tool array
was also on the southeast side of the lateral, the narrower fracture-network width could also be a function of increased location
accuracy caused by the small observational distance. This can be explored further with the other stages. Judging from the DTS
data, the width of fracture initiation at the near wellbore is significantly larger than that seen during Stage 1, and the related
fracture-network width procured from the microseismic mapping does not appear to drastically change. From this simple
comparison, it can be concluded that an increased width of fracture initiation at the near wellbore does not necessarily effect the
created geometry or the calculated SRV value. It also shows that this substantial increase in near-wellbore initiation width, most
likely a result of poor cement, does not impact the ability to treat the job, given both Stages 1 and 2 treated almost identically.
CSUG/SPE 136565 9

Stage 2 perf interval

Fig. 8—Stage 2 combined real-time results.

The next step in the diagnostic process is to analyze each stage’s production results to see if this increased SRV directly
correlates to improved production. Although no production data was available, some conclusions can be drawn. The created SRV
is not necessarily a function of the points of fracture initiation near the wellbore but is more directly tied to treatment responses
and the completion strategy. This is important information when discussing proper perforation spacing and strategy for future
wells in this area.
Because both stages treated nearly identical, it can be deduced that the Stage 1 geometry is most likely symmetric across the
lateral. Perhaps events on the northeast side of the lateral were not recorded because of the location of the geophone observation
well.

Stage 3
Stage 3 (Fig. 9) reveals a significant conclusion when using both the diagnostic results from the microseismic mapping and DTS
data. The first observations will be made about the microseismic results individually. In Stage 3, we again see reasonable fracture
symmetry across the lateral. The southwest side of the lateral shows two very distinct geometries being created, which is most
likely a function of the depletion from the vertical producer in that area. On the northeast side of the lateral, the stimulated width
appears more evenly treated, based on the microseismic-event spacing. The network width on the northeast side of the lateral is
similar to Stage 2 at 320 m. There is, however, a significant amount of event overlap on the northeast side of the lateral between
Stages 2 and 3. The SRV percent overlap between stages is a valuable quantity that can be assessed in the post-fracture analysis.
The reason for this overlap is revealed when looking at the DTS data at the near wellbore. There is poor isolation between Stages 2
and 3, which allows the fluid from Treatment 3 to enter at the same point in the wellbore as Treatment 2. This creates a very large
near-wellbore area of fracture initiation into the reservoir, which is likely the reason for the significant microseismic-event overlap.
It is challenging to quantify that impact in the treatment chart because the average wellhead pressure does increase. However, that
is directly related to an incremental increase in wellhead rate. The calculated SRV between Stages 2 and 3 remains fairly similar.
This SRV overlap is of some concern because it is quite high on the northeast side of the lateral. Without the DTS data, the reason
for this SRV overlap would have likely been given as reservoir related or seen as a function of the perforation-interval spacing.
Because it can be assessed that the overlap is most likely caused by isolation ineffectiveness, the overall recommendations and
results change significantly.
10 CSUG/SPE 136565

Stage 3 perf interval

Fig. 9—Stage 3 combined real-time results.

Without detailed production information, it is challenging to make final recommendations for this stage. Perhaps this SRV
overlap actually shows an improvement in production. If that is the case, then a further recommendation could be made to decrease
the perforation spacing on future projects to promote increased SRV overlap. Understanding the effects of the near-wellbore
environment on the created fracture geometry provides great insight to the overall completion effectiveness. Combining this
information with fracture data is extremely valuable in analyzing the post-treatment diagnostics. In this case, the fracture data does
not provide any direct evidence of the poor isolation. So, the presence of the DTS cable bridges that knowledge gap when
analyzing the results of the microseismic-event locations. The major tie-in, however, is grouping all this information with the
individual interval production results. This is the key when making sound recommendations on future completion strategies within
the same reservoir.
Another way to gain information through the DTS is by the use of the warm-back data. After fluid pumping during a stage has
been completed, the formation will begin to return to geothermal temperature. The speed at which the near-wellbore area returns to
the geothermal temperature can be assessed through the DTS cable. Typically, the area of the formation that accepts the most fluid
will return to geothermal temperature at a much slower rate. When reviewing the warm-back for Stage 3, it can be seen that the
area near the Stage 2 perforation interval was the slowest to warm back after Stage 3 was completed. This leads to the conclusion
that most of the fluid pumped during Stage 3 did indeed enter into the formation across the Stage 2 perforated area.
Fig. 10 highlights an instance in time after Stage 3 has been completed and before Stage 4 is started. Clearly, the area across
the Stage 2 perforated interval is at a lower temperature than the near-wellbore area across the Stage 3 perforated interval. Most of
the fluid for Stage 3 entered the area that is warming back at a slower rate; in this case, the Stage 2 interval.
CSUG/SPE 136565 11

Fig. 10—Warm-back data from Stage 3.

Observing the warm back after a stage is the most accurate way to observe which area during a job took the most fluid if a
temperature tracer is not run during the stimulation treatment. Other information that can be assessed from shut-in analysis would
be if any form of crossflow exists between intervals.

Stage 4
The results from Stage 4 (Fig. 11) are similar to Stage 3. The first impression of the microseismic data is that the event locations
from Stage 4 directly overlap the event locations from Stage 3. The calculated percentage of overlap from the SRV calculations is
high between these two stages. The other key observation made with regards to the microseismic data is again the more planar
geometry present on the southwest side of the lateral, which is again associated with the depletion of the reservoir from an adjacent
vertical producing well. With respect to the total fracture-network width that is created on the northeast side of the lateral, total
growth is 250 m. This is noticeably less than what is observed during Stages 2 and 3.
The treatment information from Stage 4 is almost identical to the treatment information from Stage 3. These two stages treated
at almost identical wellhead rates, surface-treating pressures, and sand and fluid volumes. So, the small network width in Stage 4
does not appear to be a function of the treatment effectiveness or design.
12 CSUG/SPE 136565

When coupling all this information with the DTS data, it becomes clear that an issue with effectively isolating Stage 3 from
Stage 4 has occurred. The points of fracture initiation directly outside of the lateral for Stage 4 directly overlap the same area for
fluid entry seen during Stage 3. Instead of focusing on perforations strategy or perforation spacing as the reason for the fracture
geometry overlap, the conclusion can now be drawn that the overlap is a function of poor isolation between stages. Another factor
of note when reviewing the DTS data is the narrower near-wellbore width for fluid entry for Stage 4 vs. the near-wellbore width
from Stage 3. In fact, the near-wellbore fracture width seen during Stage 4 is roughly half of that seen during Stage 3. This might
also directly attribute to the reduced network-fracture width created. It is, however, challenging to draw direct correlations between
points of fracture initiation and far-field geometric width in this case because all the fluid pumped during Stage 4 enters across an
identical area at the near wellbore where the Stage 3 fluid entered the formation. Most likely, fracture-network width of Stage 4
will be dependent on the previously created fracture from Stage 3.

Stage 4 perf interval

Fig. 11—Stage 4 combined real-time results.

As with previous stages, the next obvious step is to analyze the production of this particular zone. It would be extremely
valuable to analyze the area of fracture initiation shared between Stages 3 and 4 to determine if the added slurry volume to that
area revealed any increase in production. If substantially increased production is garnered from this particular interval, then that
might be case to increase job volume size for wells in this area moving forward.

Stage 5
The results obtained during Stage 5 (Fig. 12) are significantly different from those seen during the other four stages analyzed so
far. Simply looking at the microseismic results tells a completely different story than the other four stages. The percent of SRV
overlap for this particular stage is extremely low compared to other stages. The total fracture-network width measured on the
northeast side of the lateral is roughly 350 m. The fracture growth to the southwest side of the lateral again appears to be
dependent on the presence of the vertical producing well in that area. With this microseismic information, the near-wellbore effects
seen with the fiber-optic DTS can be assessed. Clearly, the isolation for Stage 5 is significantly improved over Stages 2, 3, and 4,
and the width of fracture initiation exiting the wellbore is well contained near the perforation interval. These findings assisted in
creating stimulation growth into virgin reservoir. The conclusion that can be drawn based on this new information is that the poor
isolation, as seen through the previous stages, is affecting proper fluid placement within the reservoir.
Obtaining individual production results for this particular stage would certainly improve the understanding for the value of
SRV overlap and the role it plays in generating any improvements on production. This stage could help to clarify some of the
CSUG/SPE 136565 13

issues regarding proper perforation spacing because the stimulation growth of this particular stage was not directly influenced by
the previous treatments.

Stage 5 perf interval

Fig. 12—Stage 5 combined real-time results.

Another result that can be drawn from this particular stage is similar to what was seen during Stage 1. The generation of
fracture-network width in the far-field is not a function of the initiation width seen at the near wellbore. Both Stages 1 and 5 had
significantly shorter widths of fracture initiation at the near wellbore, which had no effect on the overall SRV generated far field. If
this SRV proves to provide ample production results, then increasing perforation clusters or the number or perforations per stage
would not be recommended. The information gained from this particular stage would be an excellent place to start when designing
future completion designs and strategies for this particular reservoir because the geometry is not influenced by the intrusion of
other stages at the point of fracture initiation at the near wellbore.

Total Well Results

Analyzing the diagnostic tools with respect to the real-time results leads to some firm conclusions that can be made about this
dataset. Clearly, an improved method of isolation needs to be attempted. In this particular case, it appears a poor cement bond was
the cause of the isolation issues. This poor isolation did not have an effect on the ability to complete the stage as designed. So, for
future projects, an increase in fluid and proppant volume could easily be attempted. The poor isolation also played a key role with
the amount of SRV overlap seen between the stages. It remains to be seen if this increased overlap actually improves production;
therefore, that would need to be assessed on a stage-by-stage basis to determine if perforation spacing would need to be adjusted
on future wells in this area. With good isolation, as seen during Stage 5, the amount of SRV overlap is significantly decreased.
Using production data, it can be confirmed if the presence of significant SRV overlap is desirable or not.
The presence of the vertical well on production to the southwest side of the lateral played a key role in the fluid travel in that
direction. Total SRV for Stages 3 and 4 was affected on that side of the lateral caused by the depletion from the vertical well. More
information could be obtained with the assistance of production data from each individual stage, which would be analyzed through
the fiber-optic cable. Having this information would allow for some better answers with respect to proper perforation spacing and
improved understanding for adjustments in fluid volumes moving forward.
14 CSUG/SPE 136565

Conclusion
Having the ability to accurately assess created far-field geometry of a hydraulic-fracture treatment and points of fracture initiation
at the near wellbore provides valuable information. If a created geometry is seen to be a result of the near-wellbore environment,
then remedial solutions can be run while the job is being pumped because both these technologies can be observed in real time.
The conclusions gained from post-fracture analysis are significantly improved using both these technologies. In the case of planar
geometries, typically created in a sandstone play, limited-entry effectiveness can be directly linked to perforation strategy and
spacing. The effect good isolation has on the overall completion is also easily assessed. In the case of complex network
geometries, as highlighted in our case, understanding fracture initiation at the near wellbore provides evidence as to why certain
geometries are created for each individual stage. As seen in our example, individual stages can now be assessed as case studies for
a particular field, as opposed to only having the ability to procure case-study information on a well-by-well basis.
The key information that links all this data together is having accurate production data available for each individual stage. This
production information can be acquired through the fiber-optic DTS cable as well and can be analyzed at any time during the life
of the well with little effort required to procure the information. Having the ability to relate all the information to accurate
production information provides a firm foundation for making sound conclusions and recommendations with respect to completion
of future wells in that field.
There are clear benefits to running microseismic mapping and fiber-optic DTS projects as individual diagnostics tools.
However, the benefits gained from running these tools simultaneously truly outweigh the benefits of each diagnostic tool as a
separate entity. The far-field geometric results gained from microseismic mapping has been a proven technology for the past
decade, and the understanding of what conclusions can be drawn from that information is continually evolving. While fiber-optic-
based DTS is a newer technology of recent interest within the oilfield, the technology has shown significant improvements over
the past few years. Coupling the microseimic information with accurate near-wellbore fluid-travel information from DTS and tying
that diagnostic information together with accurate production results provides a significant understanding for completion
effectiveness and allows for more resolute information about how to complete specific pay intervals for future projects.

Acknowledgements
The authors wish to thank the management of Pinnacle and Halliburton for their support and permission to publish this paper.

References
Clarkson, C.R., and Beierle, J.J. 2010. Integration of Microseismic and Other Post-Fracture Surveillance With Production
Analysis: A Tight Gas Study. Paper SPE 131786 presented at the SPE Unconventional Technical Gas Conference,
Pittsburgh, Pennsylvania, 23-25 February.
Huckabee, P. 2009. Optic Fiber Distribution Temperature for Fracture Stimulation Diagnostics and Well Performance Evaluation.
Paper SPE 118831 presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, Texas, 19-21
January.
Jaripatke, O., Grieser, B., and Chong, KK. 2010. A Completions Road Map to Shale Play Development – A review of Successful
Approach Towards Shale Play Stimulation in the Last Two Decades. Paper SPE 130369 presented at the SPE Deep Gas
Conference and Exhibition, Manama, Bahrain, 24-26 January.
Mayerhofer, M.J., Bolander, J.L., Williams, L.I., Pavy, A., Wolhart, S.L. 2005. Integration of Microseismic Fracture Mapping,
Fracture and Production Analysis with Well Interference Data to Optimize Fracture Treatments in the Overton Field, East
Texas. Paper SPE 99508 presented at the SPE Annual Technical Conference and Exhibition, Dallas, Texas, 9-12 October.
Mayerhofer, M.J., Lolon, E.P., Warpinski, N.R., Cipolla, C.L., Walser, D., Rightmire, C.M. 2010. What is Stimulated Reservoir
Volume?. Paper SPE 119890 presented at the SPE Shale Gas Production Conference, Fort Worth, Texas, 16-18
November.
Sierra, J., Kaura, J., Gualtieri, D., Glasbergen, G., Sarkar, D., Johnson, D. 2008. DTS Monitoring Data of Hydraulic Fracturing:
Experiences and Lessons Learned. Paper SPE 116182 presented at the SPE Annual Technical Conference and Exhibition,
Denver, Colorado, 21-24 September.
Warpinski, N.R. 2009. Integrating Microseismic Monitoring With Well Completions, Reservoir Behavoir, and Rock Mechanics.
Paper SPE 125239 presented at the SPE Tight Gas Completions Conference, San Antonio, Texas, 15-17 June.
Warpinski, N.R., Mayerhofer, M.J., Vincent, M.C., Cipolla, C.L., Lolon, E.P. 2008. Stimulating Unconventional Reservoirs:
Maximizing Network Growth while Optimizing Fracture Conductivity. Paper SPE 114173 presented at the SPE
Uncoventional Reservoirs Conference, Keystone, Colorado, 10-12 February.
Warpinski, N.R., Wolhart, S.L., Wright, C.A. 2001. Analysis and Prediction of Microseismicity Induced by Hydraulic Fracturing.
Paper SPE 71649 presented at the SPE Annual Technical Conference and Exhibition, New Orleans, Louisiana, 30
September-3 October.