© 2011 EAGE www.firstbreak.org
99
special topic first break volume 29, July 2011
Passive Seismic
Identifying faults and fractures in
unconventional reservoirs through
microseismic monitoring
Scott A. Wessels,
*
Alejandro De La Peña, Michael Kratz, Sherilyn Williams-Stroud and Terry
Jbeili of Microseismic describe how in microseismic monitoring of low permeability reservoirs
the use of source mechanism inversion, b values, and energy release rates enables identifica-
tion and differentiation between fracture stimulation and fault activation, critical issues for
effective hydraulic treatment.
M
icroseismic monitoring in low permeability res-
ervoirs is a valuable source of information for
unconventional resource play optimization. The
application of new technologies, such as efficient
horizontal drilling and hydraulic fracturing (‘fracking’)
has resulted in the ability for the industry to produce from
organic rich shales. Prior to such technology developments,
hydrocarbons from such formations were typically not
economically accessible or even recoverable. Monitoring of
microseismicity is essential to understanding how a forma-
tion responds to the injection of frac fluids and proppant
because many of the most active shale plays remain in the
early stages of development and a wide range of geologic
hazards may be present, such as faults, karst collapse fea-
tures, and proximal aquifers. Microseismic monitoring
provides the necessary information to estimate stimulated
reservoir volume (SRV) and identify faults that are unresolv-
able with reflection seismic data that may pose a hazard to
completion operations.
A successful frac will typically increase the permeability
of fine-grained hydrocarbon reservoirs – thus enhancing the
well’s production and delivering a significant rate of return.
This is achieved by stimulating an existing network of
natural fractures (Maxwell et al., 2006; Gale et al., 2007).
Hydraulically stimulated natural fractures are generally near
the wellbore and are a primary receiver of proppant neces-
sary to create a flow pathway to the wellbore. In some cases,
a horizontal wellbore will encounter a pre-existing stressed
tectonic fault. Pumping of fluids and proppant into a fault
can bring about one or more unintended negative effects.
A non-target formation may be stimulated and give rise to
hydraulic connectivity with aquifers that ultimately increase
water production. Another risk is diversion of fluid and
proppant to a fault zone that lies several hundred feet away
from the target fracture stage. The end result is decreased
stimulation of the target formation, a potential increase in
water production, and a significant cost to the operator in
terms of time and materials. Having the ability to differenti-
ate between faults and fractures in a timely manner is critical
to reducing such material waste which could be otherwise
employed in areas that are more favourable for effective
stimulation.
Understanding the source mechanism of a microseismic
event leads to improved event location and provides infor-
mation vital to generating realistic reservoir models. Source
mechanisms indicate how the formation fails under stress;
the polarity of the first P-wave arrival indicates relative
motion along the failure plane. Identification of one or
more source mechanisms within microseismic data recorded
during hydraulic fracturing provides information about the
current stress state of the formation and, where multiple
source mechanisms exist, can also be used to differentiate
between reactivation of a stressed tectonic fault and desirable
natural fracture stimulation.
Statistical analysis using frequency magnitude distribution
histograms (FMD) may be indicative of changes in the stress
magnitude (Schorlemmer et al., 2005; Gulia et al., 2010),
and in the following case study, determines if a population
of events is generated by fault motion or natural fractures.
The FMD relationship was first identified by Gutenberg and
Richter (1954) and demonstrated in the formula:
log N = A - bM
s
where N is the number of events with magnitudes within
a fixed interval around M
s
. A and b are constants. The
constant b for a specific event population represents the
frequency of occurrence for different size events; a higher
slope indicates fewer large events and more small events than
a lower slope b value. Maxwell et al. (2009) and Downie et
al. (2010) observed during a hydraulic treatment that fault
related microseismicity is correlated to b values of ~1 while
desirable induced natural fracture related microseismicity
exhibits a b value of ~2.
* Corresponding author, E-mail: swessels@microseismic.com