American Oil and Gas Reporter - March 2019 - 54

SpecialReport: Unconventional Resource Science
across the area. Seismic data were not
available, but based on regional geologic
mapping and the XMAC log data, major
fault systems are not thought to exist in
the immediate area surrounding the pad
(if seismic and microseismic data had
been available, a more thorough investigation could have been conducted to review the intersection of major/minor fault
systems or natural fractures with individual
stages, allowing specific natural fracture
system communications to be identified).
Statistical Analysis
After identifying the frac hits, the collected data were further investigated to
identify completion parameters that could
have influenced whether a frac stage connected with the parent well. The parameters
available and examined during the study
ranged from the order of completion operations to frac fluid chemical components
and volumes, proppant size and type,
proppant loading, stage lengths, treatment
pressures, perforation diameter, cluster
spacing, and pump rates.
Figure 4 clearly illustrates that distance

from the parent well was a primary indicator of fracture interactions. C4-M, located only 600 feet from P3-M, had the
highest percentage of stages interacting
with the parent well. Interestingly, C1-M
and C5-M are located the same distance
from P3-M, yet the number of interactions
from C5-M were much higher than from
C1-M (78 versus 53 percent, respectively).
Part of this disparity may be a matter
of timing; the C-5M zipper frac began
prior to the C-1M's zipper frac, so stress
shadowing could have caused C1-M fractures to orient away from the parent well.
In addition, many of the interactions from
the C5-M were not direct. It is suspected
that fracture-to-fracture interaction occurred between the C4-M and C5-M,
which then transferred some of the energy
to the P3-M.
As expected, given the complexity of
fracturing, there were no single, clear
frac hit indicators using simple statistical
analysis of the data. Advanced statistical
testing was necessary, primarily utilizing
nominal logistic regression techniques.
As opposed to multiple linear regression,

nominal logistic regression predicts the
likelihood of a "yes/no" result as opposed
to a numerical result. The quantification
of the magnitude of each individual frac
hit adds depth to the investigation and
allows the use of multiple linear regression
testing.
The goal of logistic regression is to
find the best fitting model to describe
the relationship between the dependent
variable (response variable) and independent (predictor or explanatory) variables. For this study, nominal logistic
regression quantified and predicted the
likelihood of a yes/no result by testing
each independent variable to determine
its level of significance (p-value) in relation to frac hits.
To better quantify and scale the pvalue for graphing, a "logworth" value
was assigned. Any logworth value above
2.0 was considered to be statistically significant. Table 2 shows the independent
variables from the study with p-values
less than 0.01, meaning that these variables
were very likely significant. A larger logworth value does not necessarily imply

Simpler, high-confidence
prospect decisions
...from clear, comprehensive,
state-of-the-art geophysical
interpretations

Our team of experts comes from your
world. They understand the objectives,
pressures and constraints you face in an
E&P division.
When you work with IGC you can expect
* Candid assessment of tools and analyses.
* Comprehensive analyses you can put to use
immediately.
* The best available data and state-of-the-art tools.
* Concise, well-packaged work product and expert
presentation support.

Get the most comprehensive integrated
seismic/gravity/magnetics tool set by using IGC
as your outsourced analytic team.

Integrated Geophysics Corporation
50 Briar Hollow Lane , Suite 400W
50

www.igcworld.com
54 THE AMERICAN OIL & GAS REPORTER

Houston, Texas 77027

Tel (713) 680-9996

http://www.igcworld.com

American Oil and Gas Reporter - March 2019

Table of Contents for the Digital Edition of American Oil and Gas Reporter - March 2019