Table
of Contents
Introduction
Data Assessment
Borehole Image Data: Fracture Analysis |
Figure
2 |
Borehole
Image Data: Wellbore Failure Analysis
Preliminary Stress Model |
| Figure
3 |
Pressure-Temperature-Spinner
Analysis
Identifying Critically Stressed Fracture Orientations
Using Preliminary Stress Models
References
Appendices
|
| Introduction
(back)
Barton et al.
(1995, 1998) showed that optimally oriented, critically stressed fractures
control permeability in areas of active tectonics. This suggests that
critically stressed fracture sets are likely to be responsible for
the majority of the geothermal production in the Coso Geothermal Field.
Knowledge of the well-constrained local stress tensor is needed to
determine the proximity of natural fractures to frictional failure
and therefore, to determine their role in reservoir permeability.
A detailed analysis is required in order to develop a geomechanical
model of the reservoir and to determine which fractures are optimally
oriented and critically stressed for shear failure. The geomechanical
model includes the pore pressure (Pp), the uniaxial compressive rock
strength (C0), and the magnitudes and orientations of the most compressive
(S1), intermediate (S2), and least compressive (S3) principal stresses.
These are derived from in situ pore pressure measurements, laboratory
rock strength tests, wireline log data, minifrac test results, and
observations of wellbore failure. Only through fracture and wellbore
failure analyses of image data, correlated petrographic analyses,
and identifying critically stressed fault orientations and fault orientations
in fluid flow intervals can we then understand the effects of subsequent
stimulation experiments upon increases in fracture permeability.
We
adopted a multi-step approach used in previous studies at Coso and elsewhere
(Barton et al., 1997a, 1997b, 1998) beginning with the identification
of image logs with sufficient data quality to perform the fracture and
failure analyses. We measured the orientation and distribution of fractures
throughout the logged intervals of the study wells from available electrical
Formation Microscanner (FMS) image data. In one well we were able to
use high-precision pressure, temperature, and spinner flowmeter (PTS)
data and compare the areas of fluid flow with the measured fracture
distribution. We developed preliminary constraints on the in situ state
of stress based on observations of wellbore failure and available drilling
and production data. All fractures were then analyzed for their proximity
to frictional failure using two stress state end members. These results
will be used to understand and utilize the permeability fabric in ongoing
EGS studies in the Coso Geothermal Field.
Data
Assessment (back)
Electrical Formation MicroScanner (FMS) borehole image data for 19 wells
drilled in the east flank of the Coso Geothermal Field were inspected
to determine where image data quality is adequate for the purposes of
this study. We identified four wells (Figure 2) with image data whose
quality ranges between excellent and fair. Image data in wells 38A-9,
38B-9, 83-16, and 86-17 are adequate for natural fracture analysis and
for constraining the orientation of stresses within the east flank of
the Coso Geothermal Field.
In order to correlate fluid flow entries with the permeable set of fractures
that are optimally oriented and stressed for shear failure, it is necessary
to possess high-resolution pressure-temperature-spinner (PTS) data.
PTS surveys were originally conducted in the four study wells; however,
only a single survey run in well 83-16 is a high-resolution survey (with
a sample rate approximately 1.0 sample/foot). PTS data currently available
for wells 38A-9, 38B-9, and 86-17 are not of sufficient resolution for
this kind of analysis. Coso Operating Company (COC) is in the process
of obtaining those data, as part of their cost-share participation in
this program. A PTS instrument is currently being developed for the
high flow rates expected in 38B-9, but wells 38A-9 and 86-17 are no
longer available for additional PTS surveys.
Robust stress magnitude information is not yet available as we have
not fully analyzed a recently completed hydraulic fracturing test (to
determine the minimum principal stress, S3 or Shmin), completed pore
pressure monitoring (to determine the pore pressure, Pp), performed
the geophysical logging (to acquire integrated density logs used to
constrain the overburden stress, Sv), or taken these results and performed
the necessary modeling to constrain the maximum principal horizontal
stress (SHmax). These studies are planned for the future phases of this
EGS project.
Borehole
Image Data: Fracture Analysis (back)
Electrical Formation Microscanner (FMS) data were analyzed for Coso
wells 83-16, 38A-9, 38B-9, and 86-17. GMI•Imager™, designed
specifically for the analysis of digital wellbore image data, was used
to interpret natural and drilling-induced features in the FMS image
data for the Coso wells. Planar structural features that intersect the
borehole appear as sinusoids on unwrapped 360° views of the image
data (e.g., Figure A1). Several different plots are used to display
fracture orientations including tadpole plots, plots showing dip and
dip azimuth versus measured depth (feet), and stereonets for different
depth ranges. The results are summarized here and discussed in detail
in Appendix A.
Natural fracture orientations were analyzed using image logs for wells
83-16 (3,674 feet), 38A-9 (3,448 feet), 38B-9 (1,604 feet), and 86-17
(233 feet). An extremely dense fracture network was observed and analyzed
in all four study wells. We also identified a subset of fractures with
significant apparent aperture and analyzed their orientations. The apparent
aperture observed in electrical image logs is based on a high electrical
conductivity contrast that can represent either the presence of a highly
conductive fluid (e.g., drilling mud), or a highly conductive vein-filling
material resulting from hydrothermal alteration. Fractures with significant
apparent aperture due to mud infiltration may be acting as fluid flow
pathways.
|
|
Figure
2. Locations and trajectories of wells within the east flank
EGS study area of the Coso Geothermal Field. (back) |
In well 83-16
fracture dips at all depths range from intermediate to steep. The
upper interval of well 83-16 (2,397–4,053 feet MD) shows a preferred
dip azimuth orientation towards the southwest. In the middle interval
(4,700–7,100 feet MD), there appears to be a higher concentration
of dip azimuths in the southeast direction. In the bottom portion
of the well (7,425–8,800 feet MD), the dip azimuths tend towards
a bimodal distribution, with one set of fractures dipping towards
the south and the other set dipping towards the west. Fracture orientations
in well 38A-9 display a dominant northeast to east dip azimuth and
a subordinate west to northwest dip azimuth with moderate to steep
dips. Fracture orientations in well 38B-9 display dominant northeast
to east dip azimuths with shallow to steep dips. The subordinate northwest-trending
dip azimuths observed in 38A-9 are not observed as frequently in 38B-9,
particularly below 7,650 feet MD. Fracture orientations in well 86-17
display a dominant northwest-trending dip azimuth with a minor northeast-trending
dip azimuth. Fracture dips in 86-17 are generally steep to vertical.
Well 86-17 is a deviated well, with a deviation ranging between 22°
and 26° in the logged interval. As a result, image logs in 86-17
capture vertical fractures that probably exist throughout the rest
of the field but that cannot be intersected as easily in the other
three study wells that have near vertical wellbores.
In all four study wells, the orientations of fractures with significant
apparent aperture have steeper dips with much less scatter in dip
azimuth than the orientation of all fractures measured.
Borehole Image Data: Wellbore Failure
Analysis (back)
Four study wells (83-16, 38A-9, 38B-9, and 86-17) were analyzed for
the distribution and orientation of drilling-induced tensile wall
failures. No stress-induced borehole breakouts were observed in the
image data analyzed. Drilling-induced tensile fractures were observed
and analyzed in wells 83-16, 38A-9, and 38B-9. The geographic azimuths
of these tensile fractures correspond to the geographic orientation
of the SHmax direction in wellbore intervals deviated less than approximately
12°.
The abundance of high-angle natural fractures makes discrimination
of these drilling induced features extremely difficult. In many cases,
the axial propagation of these features was disturbed by their proximity
to natural fractures. In other intervals of the study wells, wellbore
wall features that appear to be tensile failure are observed on all
four FMS pads. Strict criteria were followed to ensure only non-ambiguous
tensile wellbore failure were measured in each well. These measurement
criteria require that tensile wall fractures 1) occur in pairs at
the same depth, 2) occur in pairs that are approximately 180+
apart, and 3) comprise the single set of tensile failure at a given
depth.
A statistical analysis of the azimuths of tensile wall fractures yields
an orientation of SHmax of 50° +18° in well 83-16,
an SHmax orientation of 12° +15° in well 38A-9, and
an SHmax orientation of 65° +6° in well 38B-9 (Figure
3).
Preliminary Stress Model (back)
Appendix B describes in detail our preliminary analysis of the in
situ state of stress in the east flank of the Coso Reservoir. Tensile
wellbore failures were detected in electrical image data from three
of the four wells in this study. Spotty pore pressure information,
waterfrac test results, and estimates of the granitic overburden (Sv)
provide some constraints on the local stress state for the field.
While the least horizontal stress is clearly less than the overburden
we are unable to distinguish between a normal versus strike-slip faulting
regime (SHmax ≈ Sv > Shmin). The local earthquake catalog includes
both normal (Sv > SHmax > Shmin) and strike-slip (SHmax ≈ >Sv
> Shmin) earthquakes. Since neither the absolute nor relative magnitude
of the maximum horizontal stress, SHmax, is known, we bracket the
normal/strike-slip scenarios by including both these two stress state
end members in our preliminary Coulomb failure analyses.
|
|
Figure
3. The orientation and relative magnitudes of the normal faulting
and strike-slip faulting stress tensors are displayed for the wells
analyzed in this study. The radius of the green circle represents the
Sv magnitude for both stress end members, the dark blue bowtie represents
the azimuth range and relative magnitude of SHmax for the normal faulting
stress end member, the light blue bowtie represents the azimuth range
and relative magnitude of SHmax for the strike-slip faulting stress
end member, and the red bowtie represents the azimuth range and relative
magnitude of Shmin for both stress end members. (back) |
Pressure-Temperature-Spinner
Analysis (back)
High-resolution PTS
data are available only for Coso well 83-16. Details of the analysis
are described in Appendix C and the results are summarized here.
Four separate temperature gradient profiles have been analyzed to identify
the depths of flow anomalies in study well 83-16. 70 of 214 temperature
gradient anomalies were observed on more than one PTS logging run indicating
that these flow horizons are likely to represent steady state flow rather
than transient events. There is a partial correspondence between zones
of high fracture frequency and the measured flow horizons supporting
other studies at Coso that production is fracture controlled. The extremely
large number of fractures measured throughout the logged interval and
the lack of distinct fracture sets precludes the automatic correlation
of natural fractures and flow anomalies in this well. A visual inspection
of the image data at the depth of each temperature gradient anomaly
was performed to isolate the fracture likely to control fluid flow.
We compared the orientation of fractures in permeable and non-permeable
intervals and found that fractures in the vicinity of flow anomalies
define a distinct subset of the natural fracture population in the 83-16
well. While fracture orientations in non-flow intervals range from shallow
to steep dips with northeast to southeast strikes, the fracture orientations
associated with flow intervals have distinctly steeper dips, with strike
directions that trend either east-northeast, southwest, or due north.
Additionally, in well 83-16 we find that the orientations of fractures
associated with temperature gradient anomalies resemble the orientations
of fractures with significant apparent aperture, and they both show
much less scatter than the fracture orientations of all observed fractures.
This correlation suggests that apparent fracture apertures may be useful
in identifying fluid flow horizons in wells where high-resolution PTS
data do not exist.
Identifying Critically Stressed Fracture
Orientations Using Preliminary Stress Models (back)
GMI•MohrFracs™ predicts critically stressed fracture orientations
using Mohr-Coulomb faulting theory to calculate the shear stress and
effective normal stress acting on each fracture plane given the orientations
and magnitudes of the three principal stresses and the formation fluid
pressure. Barton et al. (1995, 1998) have shown that optimally oriented,
critically stressed fractures control permeability in areas of active
tectonics. This suggests that critically stressed fracture sets are
likely to be responsible for the majority of the geothermal production
in the Coso Geothermal Field. Two end-member stress states bracket the
likely state of stress at the Coso site. Using a normal faulting (Shmin
= SHmax < Sv) and a strike-slip faulting stress state (Sv = SHmax
> Shmin), we have determined the proximity of all fractures and faults
measured in the electrical image data to Coulomb (i.e., frictional)
failure (see Appendix D).
In well 83-16 the orientation of drilling-induced tensile fractures
(and the SHmax azimuth) is N50°E. If this normal stress end member
describes the active tectonic environment in this area, then faults
striking northeast and dipping approximately 60° towards the southeast,
and faults striking southwest and dipping approximately 60° towards
the northwest are potentially active and, therefore, may be responsible
for geothermal production in the area. Critically stressed fractures
in a strike-slip faulting environment either dip steeply to the southeast
with strikes that range from east-northeast to north-northeast, or they
dip steeply to the northwest with strikes that range from west-southwest
to south-southwest. Of these two fracture sets measured in the 83-16
wellbore image data, more fractures are observed with a steep southeast
dip than with a steep northwest dip.
In well 38A-9 the orientation of drilling-induced tensile fractures
is N12°E. As a result, critically stressed faults defined by the
normal stress end member strike NNE–SSW and dip approximately
60° either towards the WNW or towards the ESE. Critically stressed
fractures in a strike-slip faulting environment either dip steeply to
the east with strikes that range from northeast to northwest, or they
dip steeply to the west with strikes that range from southwest to southeast.
In well 38A-9 more fractures dip steeply to the east than steeply to
the west.
In well 38B-9 the orientation of drilling-induced tensile fractures
is N65°E. Critically stressed faults defined by the normal stress
end member strike NE–SW and dip approximately 60° either towards
the northwest or towards the southeast. Critically stressed fractures
in a strike-slip faulting environment either dip steeply to the south-southeast
with strikes that range from north-northeast to east-southeast, or they
dip steeply to the north-northwest with strikes that range from west-northwest
to south-southwest. In well 38B-9 most fractures dip steeply to the
east.
Neither drilling-induced tensile fractures nor breakouts were observed
in the image data for 86-17; therefore, the SHmax azimuth from the closest
well (N50?E in well 83-16) was used in the analysis of critically stressed
fractures in this well. Critically stressed faults defined by the normal
stress end member strike NE–SW and dip approximately 60° either
towards the northwest or towards the southeast. Critically stressed
fractures in a strike-slip faulting environment either dip steeply to
the southeast with strikes that range from north to east, or they dip
steeply to the northwest with strikes that range from west to south.
In well 86-17 most fractures are orthogonal to these potentially critically
stressed orientations.
This Coulomb analysis using two preliminary end member stress states
shows that orientations of steeply dipping critically stressed fractures
are quite sensitive to the assumed SHmax value. This result emphasizes
the importance of accurately determining the full stress tensor for
the East Flank wells of the Coso Geothermal Field.
Figures (panels) displaying drilling, fracture, well-log, petrographic
and petrologic data for the four study wells are shown in Appendix E.
|
|
Barton,
C. A., S. Hickman, R. Morin, M. D. Zoback, and D. Benoit, 1998. Reservoir-scale
fracture permeability in the Dixie Valley, Nevada, geothermal field,
In: Proceedings, Twenty-Third Workshop on Geothermal Reservoir Engineering,
SGP-TR-158, Stanford University, Stanford, California, January 26–28.
Barton,
C. A., M. D. Zoback, and D. Moos, 1995. Fluid flow along potentially
active faults in crystalline rock, Geology, 23 (8), pp. 683–686.
Barton,
C.A., S. Hickman, R.H. Morin, M.D. Zoback, T. Finkbeiner, J. Sass, and
R. Benoit (1997), Fracture permeability and its relationship to in-situ
stress in the Dixie Valley, Nevada, geothermal reservoir, Proceedings
22nd Workshop on Geothermal Reservoir Engineering, Stanford Univ., Stanford,
CA, 147-152.
|
Appendices
(back)
Appendix A: Fracture Analysis
Appendix B: In Situ Stress
Analysis
Appendix C: Fluid Flow Analysis
Appendix D: Coulomb Failure
Analysis
|
Appendix
E: Geophysical Logs
Well Name |
Depth from (ft.) |
Depth to (ft.) |
| |
4,040 |
5,760 |
| |
5,760 |
7,500 |
| |
7,380 |
9,020 |
| |
2,300 |
4,060 |
| |
4,060 |
5,600 |
| |
5,600 |
7,160 |
| |
7,160 |
8,800 |
| |
9,660 |
9,940 |
|
