American Oil and Gas Reporter - August 2015 - 70

SpecialReport: Hydraulic Fracturing Technology

Method Key For Real-Time Optimization
By Michael Kratz

HOUSTON-One of the biggest advantages of real-time microseismic monitoring of hydraulic fracturing operations
is that it enables oil and gas companies
to evaluate fracture growth as it is occurring and make key treatment decisions
or design alterations on the fly to optimize
results and mitigate risk.
Real-time treatment evaluation improves stage-by-stage stimulation effectiveness, increases productivity, lowers
development costs, and avoids operational
risks such as geohazards. Ultimately, the
technology leads to enhanced economics
and reserves recovery by identifying patterns of fluid movement, fracture development, connectivity, compaction, and
determining whether the induced fracture
and proppants are staying in the target
reservoir zone or propagating out of zone
into unproductive intervals.
These critical insights not only enable
real-time decision making to allow operators to make refinements to the fracturing
plan while a treatment is being pumped,
but also provide valuable data for longterm asset management and optimization,
such as improved horizontal well spacing
and completion design to lower development costs while boosting well productivity. The specific benefits of evaluating
hydraulic fracture performance in horizontal wellbores in real time as treatment
proceeds include:
* Optimizing completions through
clear microseismic correlations of the
reservoir's response to each stimulation
stage (including estimating event magnitudes, hydraulic fracture growth patterns
and stimulated rock volume) while allowing for quick comparisons of different
hydraulic fracture design parameters;
* Determining where to drill the next
well, and the ideal spacing and wellbore
orientation of subsequent wells to maximize production;
* Improving production and recovery
through better placement of wellbore
landing zones, more effective stage placement throughout the lateral, and keeping
the fracture extent within the target zone;
* Reducing overall well costs through
real-time monitoring and more efficient
well spacing and stage optimization; and
* Avoiding faults, water-bearing formations and out-of-zone fracture growth.
70 THE AMERICAN OIL & GAS REPORTER

To achieve these benefits, however,
fast delivery of the processed microseismic
data is fundamental to enabling effective
real-time treatment decisions. While traditional approaches use a manual picking
method to process source mechanisms
and other types of microseismic data,
automatic picking methods are much
faster and equally accurate.

Full Moment Tensor Inversion
Developing an algorithm that can calculate source mechanisms automatically
for the recorded microseismic events and
creating a model of the full moment
tensor can improve the efficiency of microseismic data processing and the interpretation of a more robust catalog of microseismic events. Automatic moment
tensor inversion calculates accurate source
mechanisms on all events, and details
moment tensor inversion in real time.
During microseismic event processing,
analysts must determine the direction and
length of that fracture plane (i.e., the
source mechanism) and the associated
moment tensor solution (i.e., a mathematical representation of the source mechanism) for each microseismic event. This
can be a time-consuming process using
manual picking methods to select firstarrival compressional (P)-wave amplitudes,
and then determine a source mechanism
solution for each individual microseismic
event. A computerized algorithm that
performs automated calculation of moment
tensors requires very little involvement
from the analyst, resulting in more efficient
data processing.
Being able to automatically calculate
estimates of a full moment tensor solution
allows for faster identification of changes
in the population of source mechanisms,
which in turn, dramatically enhances the
speed at which real-time microseismic
monitoring can reflect large systematic
changes in event source mechanisms,
and reduces the need for reprocessing in
real time. The gains in processing speed
can lead to faster identification of obstacles
such as geohazards and other valuable
information such as stress changes in the
reservoir, enabling better and faster treatment decisions.
Automatic moment tensor technology
uses a linear inversion to estimate the
moment tensor through waveform fitting
of the recorded microseismic data and

other event-location parameters. The calculation factors in vertical ground motion
at the receiver, source-receiver geometry,
and wave propagation. The source function
is estimated using the event location and
origin time. The estimated moment tensor
then is determined by finding the leastsquares solution that best fits the data
trace at each receiver to the modeled
data trace.
The moment tensor is analyzed further
by decomposition into double-couple,
compensated linear vector dipole, and
isotropic components, which help characterize whether the movement was purely
tensile or contained volumetric components. The double-couple component can
be analyzed further to determine the orientations of the maximum and minimum
compressive stress directions in addition
to the strike/dip/rake of the fracture plane
associated with the moment tensor. The
orientations of compressive stress directions can be used to understand the orientation and relative magnitudes of regional stress, which allows treatments to
target fractures of interest and/or avoid
potential geohazards.
Theoretically, analysts would handpick individual P-wave arrival amplitudes
manually to calculate the moment tensor
inversion, but this method makes it impractically time-consuming to solve for
each event in a large microseismic catalog.
Therefore, analysts typically would instead
pick a smaller representative population
of events and select a set of pure shear
solutions that generally described the
entire event catalog. The problem with
this method is that it can overly simplify
the fracture geometry.
Automated moment tensor inversion
improves discrete fracture network geometry accuracy and robustness because a
fracture plane is modeled for each microseismic event, making the solution
deterministic.
Eagle Ford Case Study
Using example data from an Eagle
Ford Shale horizontal well, Figure 1
shows a map view of the microseismic
events from the original source mechanism
analysis with manual moment tensor inversion. The events are colored by the
source mechanism class from the original
processing. The blue events clustered
around the wellbore are the dip-slip class,
American Oil and Gas Reporter - August 2015

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