Micro-Seismic Detection using Distributed Acoustic Sensing

P. Webster*, J. Wall, C. Perkins, M. Molenaar, Shell Upstream Americas

Summary
In particular, the system can be left recording during the
Distributed Acoustic Sensing (DAS) deployed in a well- hydraulic fracture treatments of neighboring wells with the
bore can be used to detect P-waves and S-waves generated goal of detecting micro-seismic activity. This can be
in the subsurface. In a well which has geophones and a extended to detecting micro-seismic activity in the fibered
DAS cable deployed, micro-seismic events have been treatment well during and immediately after the treatment;
detected on both instruments at the same time, establishing an activity often not done with geophones, since it is
the concept of using DAS as a micro-seismic detector. difficult and expensive to deploy geophones in a treatment
While DAS is still less sensitive than geophones, it has the well.
advantage of being non intrusive and a permanent
installation, so both recording in treatment wells and 4D Since Shell’s first field deployment of DAS technology in
recording concepts are realistic options that can be 2009, many fibered wells have been interrogated for DAS
implemented once the fibered cable is installed. data for the purposes of in-well hydraulic fracture
monitoring and VSP acquisition. In this paper the
Introduction application of this technology is discussed for detecting
micro-seismic events, and will discuss a field trial along
Distributed Acoustic Sensing (DAS) is an acoustic with a sensitivity analysis comparing DAS with geophones
detection technology that has recently been applied in as well as possible location methods.
production (Molenaar et al. 2011) and geophysical settings
(Mestayer et. al, 2012). A single mode silica optic fiber in a Field Trial
cable is attached to a well bore and interrogated with laser
pulses. An Interrogator Unit on the surface generates the The field case discussed in this section is a pad of cased &
laser pulses which in turn are back-scattered by impurities cemented horizontal wells in a low permeability, over-
within the silica lattice structure. This Rayleigh back- pressured sand/siltstone reservoir completed with isolated
scattered light is collected back at the surface using the hydraulic fracturing stages using the plug & perf
Interrogator Unit and recombined with the input signal to completion method. The field trial well layout is shown in
determine an amplitude and phase associated with the depth Figure 1.
from which the signal came. Thus, the fiber can effectively
be segregated into many acoustic channels of a chosen
length along the whole length of the fiber, limited by the
speed of the switch generating the laser pulse. The resulting
signal has a bandwidth of 20 kHz on a 4km fiber (although
it can be much higher on shorter fibers) with channel
lengths ranging from 1-10m. The signal is effectively a
representation of the instantaneous strain on the fiber,
which can be generated by sound (pressure waves and
shear waves) and, at low frequencies, changes in
temperature.

Fiber-optic cables can be installed in vertical and horizontal

wells, which can be treatment wells, injector wells or
observation wells. Within the cable there are often both
single mode fibers (for DAS) and multi-mode fibers (for Figure 1. Plan view of pad layout at the field trial.
Distributed Temperature Sensing, or DTS). The extra fibers Wells D and F have single mode fibres deployed. Well F
can offer redundancy and the ability to interrogate with had 16 geophones at the two locations marked during the
different instrumentation simultaneously. Frequently a hydraulic fracture of the D and E wells, and the vertical
cable is deployed at a well for the purpose of monitoring observation well had 32 geophones for the duration of the
the effectiveness of treatments in that well. However while trial.
the well is not being treated, the fibered well is effectively
an observation, or offset, well which can be used to detect Wells F and well D had cables installed with single and
subsurface activity in the vicinity. multi-mode fibers present. The operations consisted of

© 2013 SEG DOI [Link]

SEG Houston 2013 Annual Meeting Page 2459
 Distributed Acoustic Sensing for Micro-seismic

hydraulically fracturing the wells D and E alternately from of the P-wave and S-wave match well between the two
toe to heel, and then the wells C and F alternately from toe instruments (sub millisecond). The S-wave on the
to heel. The vertical observation well had 32 geophones geophones is more diffuse looking due to all three
deployed for the duration of completion operations and component geophones detecting it. The P-wave response is
well F had 16 geophones deployed at two locations diminished on the DAS at the apex of the hyperbola. This is
(marked) during the course of the hydraulic fracture due to the broadside insensitivity of the DAS system, which
operations on the wells D and E. During deployment, the has a cosine squared dependence of amplitude and
geophones could not be pushed any further than the incidence angle (Mateeva et. al, 2011).
locations marked on well F, but this still left 4 stages on the
wells D and E where a direct comparison between DAS and
geophone responses on well F could be performed.

During the toe stages of the hydraulic fracture sequences on

wells D & E, some large micro-seismic events were
recorded on the vertical set of geophones. Although the
geophones in well F were not recording during these stages
(due to the poor incidence angle and large distance from the
treatments), the events were picked up on DAS at the same
time and are shown in Figure 2. These events exhibit some
interesting features. Firstly one can see a clear P-wave and
S-wave response with velocities that match those of the
velocity model used for the geophone micro-seismic
processing. Secondly, the shape and amplitude of the
hyperbolas associated with these wave fronts changes as
the wave approaches the heel and the vertical leg of the
well due to the apparent velocity change caused by the
Figure 3. A micro-seismic event of magnitude -1.07,
changing incident angle of the wave on the well bore.
detected on DAS (grey scale) and on the 16 horizontal
Indeed the event is large enough that we see the wave form
geophones in the same well (red wiggles). It can be seen
reflected from the surface.
that the arrival times agree well (sub millisecond),
demonstrating that the event on DAS is indeed the same
one detected on geophones.

Figure 4 shows an example of well F (with fiber deployed)

being hydraulically fractured. The three lines with hot
colors are the locations that the hydraulic fracture fluid is
entering the perforations and creating noise that is detected
on DAS. This information can used to determine the
effectiveness of the limited entry. Since DAS is also
recording on all the channels and during and after the
treatment, the data can also be used to determine whether
there was any micro-seismic activity. The lower picture
shows a zoom of the upper one, where a micro-seismic
event was detected by DAS, which was confirmed by the
Figure 2. A strong micro-seismic event recorded on DAS. geophones in the vertical well. This particular one was
The vertical axis is channel, or measured depth along the probably close to the well bore as the hyperbola is
horizontal well, and the horizontal axis is time. Colors are somewhat linear in appearance.
the energy of the DAS signal. The event was
simultaneously detected on geophones on the vertical This is the type of situation in which we would normally
observation well. Its moment magnitude was determined to (in the absence of a vertical observation well) have no
be +0.4 (calculated from the geophone data). information about micro-seismic activity due to the
difficulty in deploying geophones in treatment wells.
During the heel stages of wells D and E micro-seismic
activity was recorded on both DAS and geophones in well
F, as shown in Figure 3. It can be seen that the arrival times

© 2013 SEG DOI [Link]

SEG Houston 2013 Annual Meeting Page 2460
 Distributed Acoustic Sensing for Micro-seismic

On a distance-magnitude plot (Figure 6) there is the

expected decrease in amplitudes detected by the geophones
as the events are detected further away from the
observation well. When the DAS events observed and not
observed are plotted on top, it can be seen that DAS is less
sensitive than geophones and has a similar distance-
magnitude relationship, only translated up in magnitude.
There is a line representing the sensitivity of the DAS
micro-seismic events with respect to distance and then a
second one representing when the DAS events have a clear
P-wave as well as an S-wave visible. Assuming this to be
the limit of detect-ability of DAS, then the difference is
~1.75 orders of magnitude in the magnitudes of detectable
Figure 4. Above, treatment can be seen entering 3 events between the two instruments.
perforations and then stopping. The DAS is recording
continuously above and below those perforations and micro-
seismic activity can be seen during and immediately after
the treatment. One event is zoomed; it can be seen to have a
fairly linear looking hyperbola, meaning the event is close to
the well-bore. This is the kind of event that would be hard to
detect without DAS.

Sensitivity Analysis and Location of Events

Initially, in order to locate micro-seismic events on the

DAS data (and to confirm them as genuine micro-seismic
events), the geophone data was used to guide the times at
which to look (see Figure 5). It can be seen that many Figure 6. A distance-magnitude plot based on geophone
geophone micro-seismic events of magnitude greater than micro-seismic data. The dots are micro-seismic events
-1.6 line up nicely with the events recorded on DAS. There calculated from geophones. The red dots are the ones that
are also some events recorded on DAS that were not have been seen on DAS, green are ones that have not been
processed on the geophones, perhaps due to low SNR on seen on DAS, and the blue are ones not checked.
the geophones or co-incident events making the picking
difficult.
The fidelity and signal to noise (SNR) for the geophones
and DAS are compared in Figure 7. Although it can be seen
that DAS is less sensitive than geophones, it is also
apparent that the aperture of the event captured on DAS is
much larger; about 1600m over 300 channels, compared to
450m over the 16 geophones.

DAS can only detect strain changes along the length of the
fiber; i.e. it has no broadside sensitivity. In the case of a
non-deviated well-bore, this could be problematic since we
only know the distance the event occurred from the well-
bore (using P and S-wave separation, or a velocity model),
but not the azimuth since we only have the equivalent of a
Figure 5. The top of the figure shows the DAS energy. The
single component geophone. However if the well-bore is
vertical lines are micro-seismic events, one of which is
deviated or horizontal, the DAS orientation will effectively
zoomed and shown on the bottom of the figure. The plot in
change and as long as the event is recorded over the
the middle shows magnitude of micro-seismic events,
deviated portion of the well-bore (such as in Figure 2), we
calculated from geophone data, with respect to time. It can
should be able to locate the origin of the event. See Figure
be seen that the events located on DAS align well with
8.
those found with geophones.

© 2013 SEG DOI [Link]

SEG Houston 2013 Annual Meeting Page 2461
 Distributed Acoustic Sensing for Micro-seismic

Figure 8. A micro-seismic event recorded on DAS. From

channel 200 located in the vertical section of the well we
only know the distance the event occurred away from the
well-bore; that is, we have no azimuthal information. This
gives a circle of possible origin locations. If the event is
also recorded on the horizontal leg of the well we will have
another circle which will intersect the first circle at the only
2 possible origin locations. Depending on how deviated the
trajectory is, we could narrow this down to one location
given the number of DAS channels.

Recently DAS has been more frequently deployed in wells

Figure 7. The top display is DAS data showing a micro-
for geophysical applications, and the detection of micro-
seismic event and the bottom shows the same event
seismic adds another application of this technology. Events
detected on geophones. The aperture of the DAS event is
can be seen in treatment wells, opening the possibility of
about 200 channels covering about 1600m, whereas the 16
mapping micro-seismic activity in these logistically
geophones cover 450m of the well-bore.
difficult situations as well as perhaps reducing the
dependency of drilling dedicated observation wells in order
Conclusions to obtain micro-seismic data. When combined with other
information such as tracer data, pressure data and pump
DAS has been used to detect micro-seismic events. This curves we can start to obtain a clearer picture of the
has been verified by the detection of the same events with behavior of hydraulic fractures.
geophones in the same well-bore; and the timing agrees to
within a millisecond. A variety of event magnitudes have Acknowledgements
been observed on DAS, with events often being detected
over many channels giving a wide aperture compared to Shell Teams: Upstream Americas ULET team; B.
geophones and some large events can be observed over the Westwood, Areal Monitoring; Jorge Lopez, Jeff Mestayer,
entire well-bore. A similar distance-magnitude relationship Albena Mateeva. Asset team; Fred Mandybura, Barbara
is seen on DAS as that observed on geophones, and while a Cox, Kevin Slauenwhite. Geophysical Operations; Mark
comparison of the SNR of geophones to DAS indicates that Anderson, Carlos Nieto. Technical Services; Yue, Wu, Jeff
DAS is currently less sensitive than geophones, the Tan, Zuolin Chen. IT Support; Richard Tummers, Colin
aperture is much larger and this may help with detection Wiebe. SIEP; Alexander Droujinine. OptaSense; David
and location of micro-seismic events. DAS micro-seismic Hill, Andrew Lewis, Jim Roy. We thank Shell and
events could be located in deviated wells by using the OptaSense for permission to publish this paper.
changing orientation of the DAS channels.

© 2013 SEG DOI [Link]

SEG Houston 2013 Annual Meeting Page 2462
 [Link]

EDITED REFERENCES
Note: This reference list is a copy-edited version of the reference list submitted by the author. Reference lists for the 2013
SEG Technical Program Expanded Abstracts have been copy edited so that references provided with the online metadata for
each paper will achieve a high degree of linking to cited sources that appear on the Web.

REFERENCES
Mateeva, A., J. Mestayer, B. Cox, D. Kiyashchenko, P. Wills, J. Lopez, S. Grandi, K. Hornman, P.
Lumens, A. Franzen, D. Hill, and J. Roy, 2012, Advances in distributed acoustic sensing (DAS) for
VSP: 82nd Annual International Meeting, SEG, Expanded Abstracts, doi:10.1190/segam2012-
0739.1.
Mestayer, J., B. Cox, P. Wills, D. Kiyashchenko, J. Lopez, M. Costello, S. Bourne, G. Ugueto, R. Lupton,
G. Solano, D. Hill, and A. Lewis, 2011, Field trials of distributed acoustic sensing for geophysical
monitoring: 81st Annual International Meeting, SEG, Expanded Abstracts, 4253–4257.
Molenaar, M., D. J. Hill, P. Webster, E. Fidan, and B. Birch, 2011, First downhole application of
distributed acoustic sensing (DAS) for hydraulic fracture monitoring and diagnostics: Hydraulic
Fracturing Technology Conference, Society of Petroleum Engineers, Paper 140561.

© 2013 SEG DOI [Link]

SEG Houston 2013 Annual Meeting Page 2463