American Oil and Gas Reporter - July 2017 - 60

SpecialReport: Horizontal Wellbore Construction

Tools Optimize STACK Development
By Kyle Haustveit,
Kyle Dahlgren,
Hart Greenwood,
Tom Peryam,
Breanne Kennedy
and Matt Dawson

OKLAHOMA CITY-Understanding
fracture geometry and well spacing early
in the development of a field allows for
optimal development strategies that can
drastically improve net present value, especially in the case of stacked reservoir
intervals such as the Sooner Trend
Anadarko Canadian and Kingfisher play
in the Anadarko Basin.
To better understand fracture geometry
in its STACK leasehold, Devon Energy
Corp. applied multiple diagnostic tools
as part of a five-well "underground laboratory" pilot project. Each tool provided
insights into different hydraulic, propped
and conductive fracture geometry characteristics, and integrating the results has
led to significant completion changes in
this world-class unconventional asset.
The technologies applied included:
* Fiber optic monitoring with distributed acoustic sensing (DAS) and distributed temperature sensing (DTS) to
assess cluster efficiency, fluid and proppant
distribution per cluster, and diverter effectiveness;
* A five-array borehole microseismic
system (vertical and horizontal arrays)
to estimate hydraulic half-lengths, heights

and fracture azimuths;
* Electromagnetic imaging to provide
insight on hydraulic half-lengths;
* Offset well pressure monitoring
with advanced pressure-based fracture
geometry technology to assess hydraulic
and propped half-lengths, heights and
fracture azimuths (the offset well pressure
data, coupled with fiber optic monitoring,
also led to the optimization of diverter
applications);
* Water hammer analysis; and
* Fracture modeling calibrated using
a diagnostic fracture injection test and
vertical well log data.
The validation tools used in the project
included production interference testing,
rate transient analysis, oil-soluble tracers
and fracture fluid identifiers.
A variety of completion variables were
tested, including fluid design, proppant
size, perforation designs and diverter
types, and the results have been integrated
into an improved completion design. Results from the technologies has led to an
improvement in well spacing and suggest
an increase in well density is required to
maximize NPV.
Fiber Optic Monitoring
The center well (Well C) in the fivehorizontal well layout for the pilot project
was equipped with fiber optic monitoring,
offset by wells A and B on one side and
wells D and E on the other side. Diversion
was identified by evaluating DAS and

FIGURE 1
Well C Proppant Distribution/Cluster from DAS

60 THE AMERICAN OIL & GAS REPORTER

DTS data. The DAS results suggest a
strong "heelward" bias present in a majority
of the stages. Stress shadowing within a
given stage, and from adjacent fracture
stages, resulted in a consistent geometric
predominance for fracture growth in the
most heelward perforation clusters.
A variety of completion variables targeting improved fluid distribution was
evaluated throughout the completion of
the lateral. Notable variables included
perforation design, fluid rheology, diverter
and proppant size. Two perforation designs
were trialed on several stages, resulting
in improved fluid distribution compared
with the base perforation design. Results
from DAS influenced a change to the
timing of proppant size changes.
Figure 1 shows the proppant distribution of individual clusters for several
stages of the fiber optic well. An algorithm
applied to the DAS data provided the
proppant volume placed into each cluster
of Well C. The average proppant distribution in each stage again shows a heelward bias. The ability to calculate the
amount of fluid and proppant placed into
each cluster aids in calibrating fracture
models and improving forward modeling
capabilities.
With any alteration of the treatment
schedule, both primary perforation designs
tested on Well C were used in succession
to normalize for the changes. Looking at
the proppant distribution of each stage
as an absolute percent deviation from
perfect efficiency, the second perforation
design averaged 18 percent improved
distribution compared with the base perforation design. Modeling the improved
distributions derived from fiber optic
monitoring indicates that the second perforation design created more equal fracture
heights and half-lengths, as well as increased the overall number of effective
fractures in the wellbore.
Figure 2 shows the heelward bias of
the total treatment on Well C. Nearly 50
percent of the designed treatment was
placed in the two heelmost clusters. On
average, the heel cluster received more
than twice as much treatment as the toe
cluster.
Along with the preplanned variable
testing, multiple optimizations of pump
design were discovered during the treatment of the fiber optic well using realtime DAS and DTS data. The pumping

American Oil and Gas Reporter - July 2017

Table of Contents for the Digital Edition of American Oil and Gas Reporter - July 2017