integrating well and seismic data for characterisation of ... · pdf fileintegrating well and...
Post on 07-Feb-2018
233 Views
Preview:
TRANSCRIPT
23rd International Geophysical Conference and Exhibition, 11-14 August 2013 - Melbourne, Australia 1
Integrating Well and Seismic Data for Characterisation of Shale Plays
Pieter Gabriels Ted Holden Jason Jason
69 Outram Street 6671 Southwest Freeway, Suite 600
West Perth, WA 6005, Australia Houston, TX 77074, USA
Pieter.Gabriels@cgg.com Ted.Holden@cgg.com
INTRODUCTION
The success of shale plays in North America has created a lot
of excitement in Australia recently. Australia’s shale potential
is very large. The U.S. Energy Information Agency (EIA)
ranks Australia’s technically recoverable shale resources sixth
in the world. When it comes to producing from shale
reservoirs though, some tough questions will need to be
answered:
• Where are most hydrocarbons (free and absorbed)?
Hydrocarbon bearing shale formations are both source and
reservoir rocks; the biogenic or thermogenic gas may be
trapped within micropores of the shale, or in local fracture
porosity, or absorbed onto mineral or organic matter
components of the shale.
• Where are the natural fractures? What are their density,
orientation and content? The presence of open natural
fractures allows the gas to flow, facilitating the recovery
process. However fracture communication between a water-
bearing zone and the shale may render the prospective shale
non-commercial.
• Where are the rocks which can be fractured? Shale
formations have low permeability and the horizontal
production wells generally require hydraulic fracturing
(fracking) of the rocks to stimulate production. Fracking
costs an estimated 1/3rd of total drilling budget and carries
large risk: the production-to-frack ratios vary widely. The
latter is illustrated by the following examples. A well
received a total frac of 85,000 Bbl and produced in the first
5 months an average of 43 million scf gas per month. This is
a production-to-frack ratio of 0.51 MCF/Bbl. Another well
in the same field received a total frac of 4,700 Bbl and
produced in the first 5 months an average of 60 million scf
gas per month. This is a production-to-frack ratio of 12.8
MCF/Bbl, which is more than an order of magnitude higher.
In this paper we will discuss an integrated workflow that
addresses these key questions. Some data examples from
studies performed in North America will be used to illustrate
the approach.
PETROPHYSICS
The first part of the methodology is to analyse the core data
and wireline logs from all wells and describe the shales in
terms of their source rock potential: total organic content
(TOC), thermal maturity and kerogen. Shales may also be
described in terms of their producability, using measures such
as their quartz content, the presence of fractures and the
pressure gradient of the rock layer.
Shales contain a complex mixture of minerals, such as clays,
heavy minerals, quartz, carbonates and kerogen. This
complexity calls for an advanced petrophysical analysis. A
stochastic approach was adopted to estimate the mineral
volumes of shale reservoirs using conventional logs (Jensen
and Rael, 2012). The model results for each mineral were
calibrated to X-ray diffraction (XRD) core cutting analysis
(Figure 1). Once the stochastic model is derived and
calibrated, it can be applied to other wells in the area.
SUMMARY
Shale plays have revolutionised the oil and gas industry
in North America and exploitation of these kinds of plays
is steadily gathering pace in other parts of the world.
Because hydrocarbon bearing shales usually have
insufficient permeability to allow significant flow to a
well, production from these unconventional reservoirs
comes with unique challenges. Optimizing recoverable
reserves from shales requires strategic placement of
horizontal wells: placing the well in the best areas,
drilling the lateral in the proper direction and keeping the
lateral portion of the wellbore in the optimum layer. It
further requires production stimulation by hydraulic
fracturing (fracking) of the rocks to connect the natural
fractures with induced near-well fractures.
In this paper, we present a methodology to identify these
optimum areas and layers in the shales using a seismic
characterisation workflow where well and seismic data
are rigorously integrated. The first part of the approach is
well data analysis to extract petrophysical, rock physics
and mechanical information. Shale formations have a
complex mineralogy requiring a sophisticated
petrophysical analysis. Then a seismic inversion is
performed to predict rock properties, which characterise
the shale reservoirs and importantly allow us to predict
how the rocks will respond to fracking. The final part of
the methodology is an interpretation of multiple rock
property models in terms of defined shale facies. A
Bayesian approach was adopted to generate shale facies
models that describe the thickness and complex
architecture of shale reservoirs. These facies models can
be used to significantly reduce the risk of poorly
performing wells and improve asset performance.
Key words: unconventional, shales, inversion, reservoir
characterisation, hydraulic fracturing.
Dow
nloa
ded
12/2
2/15
to 1
04.1
29.1
92.5
5. R
edis
trib
utio
n su
bjec
t to
SEG
lice
nse
or c
opyr
ight
; see
Ter
ms
of U
se a
t http
://lib
rary
.seg
.org
/
Integrating Well and Seismic Data for Characterisation of Shale Plays Gabriels P. and Holden T.
23rd International Geophysical Conference and Exhibition, 11-14 August 2013 - Melbourne, Australia 2
Figure 1. Volumetric mineral results compared to XRD
analysis of cuttings. The 3rd
track shows that the
stochastic model matches Vkerogen from XRD reasonably
(brown), better than the modified Passey method (blue).
The mineralogy, porosity and water saturation results from the
stochastic are then used in a rock physics model to determine
the ability of a rock to fail under stress (Poisson’s ratio) and
maintain a fracture (Young’s modulus) (Figure 2). These
elastic rock properties can be combined to derive a shale
brittleness index (Rickman et al., 2008). The brittleness has
proven to be useful to determine the suitability of the
formation for fracking. Brittle shale is more likely to be
naturally fractured and respond favourably to fracking. On the
other hand, ductile shale is not a good reservoir, because the
formation wants to heal any natural or hydraulic fractures.
Figure 2. The lithology and fluid analysis results from the
stochastic model along with the Poisson’s ratio and
Young’s modulus
As part of the interpretation of wireline logs, the various
petrophysical rock properties (e.g. Vquartz, Vclay, Porosity,
Vkerogen), mechanical rock properties (e.g. brittleness, stress)
and elastic rock properties (e.g. P-impedance, S-impedance,
Density) were carefully investigated using cross-plots to
establish meaningful relationships. For example, a cross-plot
of P-impedance and Vp/Vs overlain by brittleness showed a
clear trend from shales with high brittleness to shales with low
brittleness. All empirical and rock physics information was
then used in the inversion of 3D seismic data.
SIMULTANEOUS INVERSION
Our understanding gained from the well data analysis needs to
be extrapolated away from well locations and simultaneous
inversion of 3D seismic data offers a way to do this. The goal
of rigorously integrating the well and seismic data through
inversion is to accurately predict reservoir and mechanical
rock properties of the shales. These rock properties can then
be used to identify favourable characteristics for optimal well
placement, hazard avoidance and geosteering of long
horizontal wellbores.
The direct outputs from a simultaneous inversion are elastic
rock properties: P-impedance, S-impedance and Density. The
latter usually requires seismic reflection angles exceeding 50
degrees, unless other independent information is available (R.
Roberts ea., 2004). The inversion results were used to
compute models of Poisson’s ratio and Young’s modulus
(Varga et al., 2008). A shale brittleness index was
subsequently calculated from the Poisson’s ratio and Young’s
modulus (Figure 3).
Figure 3. This section shows the shale brittleness index
derived from Poisson’s ratio and Young’s modulus, which
were computed from inversion results. The brittleness logs
from seven wells are overlain.
The relative presence of quartz in shales is important, as the
more of this mineral is present, the easier it is to stimulate the
shale reservoir. Therefore models of Vquartz were computed
through geostatistical co-simulation, using the direct outputs
from inversion and Vquartz logs from wells. The final
Vquartz model was then calculated as the mean from 10
realisations.
INTERPRETATION
Natural fractures can provide pathways for the oil or gas to
move to the wellbore and need to be characterised, if they
exist. This was done by computing discontinuity attributes
(coherence, curvature) from the inversion results.
The multiple rock property models were interpreted with the
aim to delineate shale facies with important producability
characteristics. When deciding how to classify the shale
facies, it is important that the facies not only represent classes
of rocks with significant petrophysical rock properties, but
also that they have reasonably distinct elastic or mechanical
rock properties, so that their distribution can be constrained by
the seismic inversion results. Five shale facies were defined
based on Vkerogen, brittleness and Vquartz: 1) high kerogen,
low brittleness, 2) medium kerogen, 3) high quartz, medium to
high brittleness, 4) low quartz, low kerogen 5) high quartz,
very high brittleness (Figure 4).
low
high
Tim
e (ms)
3.5 km
Dow
nloa
ded
12/2
2/15
to 1
04.1
29.1
92.5
5. R
edis
trib
utio
n su
bjec
t to
SEG
lice
nse
or c
opyr
ight
; see
Ter
ms
of U
se a
t http
://lib
rary
.seg
.org
/
Integrating Well and Seismic Data for Characterisation of Shale Plays Gabriels P. and Holden T.
23rd International Geophysical Conference and Exhibition, 11-14 August 2013 - Melbourne, Australia 3
Figure 4. Probability Density Functions (PDFs) for five
shale facies in a cross-plot of Vquartz and brittleness.
The most productive shale is facies type 3. The very high
brittle shales (facies type 5) do not represent good reservoir as
the amount of kerogen in this facies is low.
A Bayesian approach was adopted to interpret the brittleness
and final Vquartz models in terms of these five shale facies.
This way a probabilistic interpretation of brittleness and
Vquartz models was performed and a probability volume for
each of the five shale facies was calculated. A most likely
facies model is then easily computed (Figure 5).
Figure 5. This section shows the most likely facies model
derived through a probabilistic interpretation of
brittleness and Vquartz models. The Vquartz logs from
seven wells are overlain.
Finally, the facies models were analysed to determine if they
could explain the existing production data. The good
production-to-frack ratios came from wells with their
horizontal section well within a layer of facies type 3, and
usually underlain by a ductile layer of facies type 1. The latter
layer acts as a barrier stopping the induced near-well fractures
from reaching water-bearing rocks underlying the shale
(Figure 6). The poorly-performing wells had missed the
optimum shale layer partly or completely.
Figure 6. This section shows a 1.3 km lateral at 2 km
depth which successfully targeted shale facies type 3 (red);
this agrees with a high production-to-frack ratio of 12.8.
CONCLUSIONS
We have described a methodology to characterise shale
reservoirs. In the approach geology, petrophysical and
geophysical data are rigorously integrated to build realistic
and accurate subsurface models describing the thickness and
complex architecture of shale formations. Stochastic methods
are optimally suited for the petrophysical interpretation of
shale mineralogy and fluids. Simultaneous inversion of
seismic data provides the interpreter with a set of elastic rock
properties, which can be used to predict reservoir and
mechanical rock properties. A Bayesian approach offers a way
to interpret a number of defined shale facies based on
important producability characteristics. The facies models
calculated in this manner can be used to strategically locate
horizontal wells, significantly reducing the risk of poorly
performing wells. Additionally the models can help to
enhance the well completion and stimulation, thereby
increasing the value derived from rigorous data integration.
ACKNOWLEDGMENTS
Some material in this paper was presented at the SEG Annual
Meeting in Las Vegas, 2012. The authors thank our colleagues
at Fugro-Jason for reviewing this paper.
REFERENCES
Jensen, F., Rael, H., 2012, Stochastic modeling &
petrophysical analysis of unconventional shales: Spraberry-
Wolfcamp example.
Rickman, R., Mullen, M., Petre, E., Grieser, B., Kundert, D.,
2008, A practical use of shale petrophysics for stimulation
design optimization: All shale plays are not clones of the
Barnett shale, SPE 115258.
Roberts, R., Bedingfield, J., Guilloux, Y., Carr, M., Mesdag,
P., 2004, A case study of inversion of long offset P-wave
seismic data into elastic parameters, EAGE 66th conference,
D008.
Varga, R., Lotti, R., Pachos, A., Holden, T., Marini, I.,
Spadafora, E., Pendrel, J., 2012, Seismic Inversion in the
Barnett Shale successfully pinpoints sweet spots to optimize
wellbore placement and reduce drilling risks, SEG Technical
Program Expanded Abstracts: pp. 1-5.
Vquartz
Brittleness Index
Probability
Vquartz
Probability
Brittleness Index
Tim
e (ms)
3.5 km
Depth (ft)
Dow
nloa
ded
12/2
2/15
to 1
04.1
29.1
92.5
5. R
edis
trib
utio
n su
bjec
t to
SEG
lice
nse
or c
opyr
ight
; see
Ter
ms
of U
se a
t http
://lib
rary
.seg
.org
/
top related