PROCEEDINGS, Thirtieth Workshop on Geothermal Reservoir Engineering

Stanford University, Stanford, California, January 31-February 2, 2005

SGP-TR-176

COMPARISON OF ACOUSTIC AND ELECTRICAL IMAGE LOGS FROM THE COSO

GEOTHERMAL FIELD, CA

Nicholas C. Davatzes and Steve Hickman

U.S. Geological Survey, Menlo Park, CA, 94025

e-mail: ndavatzes@usgs.gov

ABSTRACT INTRODUCTION
Electrical and acoustic image logs collected from Electrical and acoustic image logging tools provide
well 58A-10 in crystalline rock on the eastern margin an invaluable opportunity to characterize the fracture
of the Coso Geothermal Field, CA, reveal different populations that typically control fluid flow in
populations of planar structures intersecting the geothermal systems. However, these tools detect
borehole. Electrical image logs appear to be sensitive fractures by measuring different properties of the
to variations in mineralogy, porosity, and fluid borehole wall. Fractures interpreted from electrical
content that highlight both natural fractures and rock image logs are identified by contrasts in conductivity
fabrics. These fabric elements account for about 50% between the fracture and the adjacent borehole wall.
of the total population of planar structures seen in the By contrast, fractures in acoustic image logs are
electrical image log, but locally approach 100%. This associated with changes in borehole wall surface
fabric is unlikely to contribute to permeability in the roughness or acoustic reflectivity. In both types of
reservoir. Acoustic image logs reveal a similar logs, fractures with the largest apparent apertures are
natural fracture population, but generally image often—but not always—observed to dominate
slightly fewer fractures, and do not reveal rock fabric. subsurface fluid flow in geothermal fields (Barton et
Both logs also record textural properties of deformed al., 1998; Sheridan and Hickman., 2004). Similarly,
materials within fractures; these textures can be other properties affecting subsurface permeability
related to variations in mineralogy, alteration, or such as rock type variation, foliation, and potential
porosity using the electrical log and can be used to hydrothermal alteration can also be detected through
infer slip history. In addition, locations of high these methods.
fracture density occur adjacent to major faults, but
also occur as zones confined within intervals of Stress concentrations associated with drilling and the
distinct rock type. introduction of the borehole can produce structures
such as petal-centerline fractures (Kulander et al.,
The acoustic image logs collected in Well 58A10 1990; Lorenz et al., 1990), borehole wall tensile
reveal large numbers of drilling-induced borehole fractures and breakouts (Moos and Zoback, 1990).
wall tensile fractures and breakouts. In addition, These structures are directly related to the current
previously un-recognized drilling-induced petal- stress state and can be used to infer the orientations
centerline fractures are evident in images of the of the principal stresses. Small-scale rotations of the
borehole wall from this well. These features are principal stresses revealed by changes in orientation
thought to form just ahead of the drill bit and provide of these structures can also indicate which faults have
additional constraint on the orientation of the recently slipped. Currently slipping fractures, or
minimum horizontal stress. Both breakouts and petal- those well oriented for slip in the current stress state,
centerline fractures are difficult to identify in typically comprise the majority of permeable
electrical image logs because of the low percentage fractures in crystalline rocks (Barton et al., 1995;
of the borehole wall imaged. Analysis of induced Barton et al., 1998; Sheridan and Hickman, 2004).
structures in acoustic image logs indicates that the
minimum horizontal stress is oriented along an In this study, we analyzed populations of natural and
azimuth of 102º±18º. This orientation is consistent induced fractures from well 58A-10 (Figure 1) in the
with the local strike of the Coso Wash normal fault Coso Geothermal Field using both electrical and
that bounds the basin containing well 58A-10. acoustic image logs to: (1) determine how geologic
features detected by each technique differ, (2)
determine each technique’s effectiveness in inferring
the local stress state from drilling induced structures,
(3) relate natural fracture populations to the local injection. We analyzed both acoustic and electrical
state of stress and rock type variation, and (4) image logs from a measured depth of about 6800 to
evaluate the use of these techniques for interpreting 10238 ft through diorite, quartz diorite, and
fracture-dominated fluid flow in geothermal systems. granodiorite. In addition to providing a better
understanding of image logging methods in
GEOLOGIC SETTING geothermal wells, the analysis of fractures crossed by
this well and the local state of stress inferred from
The Coso Geothermal Field is located along the
drilling-induced structures provides a basis for
western edge of the Eastern California Shear Zone,
comparison with the adjacent producing geothermal
subjected to both strike slip and Basin and Range field. Previous studies in other Coso wells, such as
style normal faulting (Figure 1a). Unruh et al. (2002) 38C-9, have been based on the interpretations of
suggested that Coso lies within the right (releasing)
electrical image logs. Thus this comparison between
step between the right lateral Owens Valley fault to
electrical and acoustic image logging methods also
the north and the Airport Lake fault to the south. The
provides a necessary benchmark for the interpretation
field sits above a shallow heat source presumed to be
of these different data sets. In particular, this study
a partially molten magma body as shallow as 4-5 km will help combine insights from both techniques in
depth (Wicks et al., 2001). preparation for the stimulation of well 34-9RD2 as
part of the Enhanced Geothermal Systems (EGS)
Well 58A-10 is located in Coso Wash adjacent to the
experiment in the Coso Geothermal Field.
exploited geothermal field (Figure 1b), but outside
the region impacted by current production or

Figure 1: (a) Tectonic map of southern portion of Eastern California Shear zone modified from Unruh et al. (2002).
(b) Simplified geologic map of basement crystalline rocks in the Coso Geothermal Field. Faults of
different ages are shown, as are measured orientations of the minimum horizontal stress (Shmin) from
induced structures in well 58A-10 and based on observations by Sheridan and Hickman (2003) from wells
83-16, 38B-9, 38C-9, and 38A-9.
 ABI88. Each pulse has an optimal footprint of 5 to 7
IMAGE LOGGING METHODS mm. Two images are available for analysis: (1) two-
way travel time (proportional to borehole radius) and
Electrical image logs were obtained in well 58A-10 (2) amplitude. Image quality is sensitive to borehole
using Schlumberger’s Hot Hole Formation Micro- shape and tool position. Smooth, cylindrical
Scanner tool (FMS) that has been used extensively
boreholes with a well-centered tool return maximum
throughout the Coso Geothermal Field. Electrical
acoustic energy because of the resulting high
images are produced by placing pads with arrays of
incidence angle and low scattering of the acoustic
electrodes maintained at a constant electrical
beam. Departures from this ideal degrade the image
potential against the borehole wall, and measuring quality. BHTV logs were collected in 58A-10 with
the current drop as the electrodes travel along the vertical pixel dimensions of 5 to 7 mm and azimuthal
borehole wall (Ekstrom et al., 1987). Data from
pixel dimensions of 2.5°,about 3.5 mm. BHTV image
multiple electrodes are combined to produce
logs were processed using two different programs:
electrical conductivity images. Because a current is
WellCadTM by ALT and ImagerTM by Geomechanics
being passed into the borehole wall, this technique
International (GMI).
actually measures the properties of a volume of rock
within a few inches of the borehole wall. The
azimuthal coverage of the image and its resolution is
During FMS and BHTV log analysis we
determined by the size and number of electrodes,
distinguished drilling-induced structures, natural
their arrangement, the pad dimensions, and the
fractures, rock fabric, rock type transitions, and
borehole diameter. Resolution approaching 5 mm
indications of alteration wherever possible. The
was attained in well 58A-10 with the FMS tool, and orientation of structures in both types of image logs
the image spanned about 40% of the borehole was measured interactively within ImagerTM. In
circumference within four equally spaced strips.
addition, we ranked the quality of natural fracture,
Overall image quality is strongly related to pad
fabric, and rock type transitions on a scale from A to
contact and thus is sensitive to borehole shape and
C. A indicates the highest quality, least ambiguous
roughness, and to mud cake (Hearst et al., 2000). pick which occurs in locations of good overall image
This tool can operate at temperatures up to 200° C for quality, continuous fracture trace across the borehole
up to 1 hour, with longer operating times at lower and clearly defined orientation. Structures assigned a
temperatures. quality ranking of C required greater interpretation,
usually resulting from connecting discontinuous or
The acoustic log used in this study was produced by a ambiguous line segments to define the structure trace.
new state-of-the-art borehole televiewer (BHTV) Conservative picking in this study probably
built by Applied Logic Technologies (ALT) under underestimates the total number of fractures, but
joint funding from the Department of Energy and leads to a more reliable orientation analysis and
Navy Geothermal Program Office in collaboration relative distribution of structures. For high quality
with Sandia National Lab and the U.S. Geological fracture picks we also measured the apparent
Survey. The new BHTV is specifically designed for aperture, or thickness of the feature, in the image of
use in geothermal wells up to 300°C can operate at the borehole wall. To ensure an unbiased comparison
temperatures of 275°C for approximately 12 hours. of results from each image log, structures were
This study combines results from two BHTV tools initially picked independently in each log and
from ALT, the ABI85 which includes a combined subsequently followed up with detailed comparison.
temperature sensor, and the prototype Interpreted image logs were cross-referenced with the
cuttings log to correlate overall image log character
Acoustic image logs are produced by bouncing an with rock type. Future work includes incorporation of
ultrasonic acoustic pulse from the borehole wall temperature and magnetic field strength recorded by
(Zemaneck et al., 1970). The tool uses a transducer the ABI85 into our structural and lithologic
that both emits and records the acoustic pulse. The interpretations.
travel time of the pulse from the transducer to the
wall and back indicates the dimensions of the
IMAGE LOG ANALYSIS AND COMPARISON
borehole and the relative position of the tool. The
energy of the returning pulse, recorded as amplitude,
Natural fractures and fault rock texture
is a function of the degree of scattering of the pulse
due to the borehole shape and rugosity as well as the Planar fractures appear as sine waves when the 360º
acoustic impedance contrast between the borehole borehole images are unwrapped onto a flat plane.
fluid and wall. Complete azimuthal scans of the Fractures with large apparent aperture and distinct
borehole wall are composed of 72, 144, or 288 from neighboring fractures are easy to pick using
acoustic pulses directed by a rotating mirror. both FMS and BHTV logs. However, in cases where
the fracture is slightly non-planar, or in locations of In well 58A-10 both types of image logs revealed
high fracture density, it becomes difficult to reliably natural fractures with varying thickness or slightly
pick fractures in the FMS because of incomplete non-planar geometry (Figure 2). For example, the
azimuthal imaging. For the same reason crosscutting fracture visible at center-depth in Figures 2a, b, and c
and abutting relations are far harder to distinguish in has a large asperity along its surface. The basal trace
the FMS than in the BHTV. However, where large is the most nearly sinusoidal, and successively less
faults are evident from observations in the FMS log planar segments (less sinusoidal) in the un-wrapped
or cuttings log, or where caliper data suggests image are associated with the “descending” limb of
widening of the borehole, the BHTV typically the sinusoid. The shape of the asperity suggests
provides poorer image quality than the FMS due to successive opening or shearing along multiple
data loss resulting from scattering of the signal along surfaces, possibly representing multiple crack-seal
rough borehole walls. events (Figure 2d). Either geometry indicates
apparent extension consistent with oblique normal
slip across this fracture (Figure 2e). This geometry is
similar to core from well 64-16 (Figure 1) from the
East Flank of the field (Figure 2f), which contains
multiple shear zones of foliated chlorite-smectite
surrounded by brecciated host rock.

The BHTV clearly reveals fine scale geometry of the

structure in Figure 2. The FMS log reveals less of the
internal structure, but indicates variation in the
conductivity of the fracture where more resistive
zones are bright and more conductive zones toward
the margins of the fracture are dark. In the BHTV
image the locations corresponding to higher
conductivity coincide, with signal loss (white areas in
the travel time data) indicating a rough or recessed
surface. Both the variation in the loss of data and
varying conductivity indicate heterogeneous physical
properties within the fracture that might be related to
porosity production accompanying slip, the
Figure 2: Natural fractures in BHTV (a) amplitude, distribution of minerals with different conductivity,
(b) travel time, and (c) FMS image logs or preferential erosion of the fracture and adjacent
from a measured depth of 9857 ft. The wall rock during drilling.
BHTV and FMS logs are oriented with
respect to magnetic North (MN) and true
(TN), respectively, with a magnetic
declination at this site of 14ºE. Each image
progresses from North, then through East,
South, West, and back to North. (d) Sketch
of natural fracture geometry and texture of
fault rock seen in Figures 1a through 1c.
(e) Stereogram of natural fracture colored
red in (d). The gray arrows indicate the
dilation direction inferred from the texture
of the natural fracture. (f) Similar
geometry to (a-d) is visible in core from
990 ft measured depth in well 64-16 Figure 3: BHTV amplitude and FMS image logs of
(Figure 1). In this figure and all intervals containing two transitions in rock
subsequent figures brighter images in the type indicated by changes in image log
BHTV amplitude indicate greater signal character and changing depth of wear
strength and in two-way travel time caused by rubbing of the drill string (tool
indicate shorter times. Brighter images in marks). The upper rock type contains a
FMS indicate lower conductivity. White in foliation, which is only visible in the FMS
travel time data indicates signal loss. log and is similar to fabric visible in core
from well 64-16.
Figure 4: Plots of rock type variation indicated by the cuttings log and changes in image log character (as in Figure
3), natural fracture frequency, fabric frequency and true fabric orientation corrected for borehole
deviation in the logged interval of well 58A-10 from 6820 to 10238 ft. Alteration was logged from the
cuttings during drilling as part of the cuttings log.
Figure 5: Natural fracture true dip and true dip direction for all fractures and for those with greater than 5 mm
apparent aperture from FMS and BHTV logs for the interval from 6800 to 10238 ft. (5 mm is roughly
equivalent to the minimum resolution of the FMS and BHTV.) Major faults are plotted as yellow boxes.

Rock type and rock properties changes abruptly across such transitions, and
Both image logs show changes in image log qualitatively indicates differences in rock hardness or
character, image quality, and pipe-wear along the abrasion resistant. The FMS does not clearly
borehole wall that occur across abrupt near-planar distinguish the change in pipe-wear (although in
transitions. Transitions that separate large (10’s of m) some cases it might be detectable from the associated
depth intervals of distinct image log character caliper log), but does suggest different conductivity
coincide with changes in rock type indicated by the of the borehole wall in this region. We interpret these
cuttings log. Smaller depth intervals distinguished by changes in the physical properties of the borehole
distinct image log character in both BHTV and FMS wall to suggest that either rock type or alteration vary
(Figure 3) are not associated with distinct changes in rapidly over shorter distances than are resolved by
cuttings mineralogy. However, as shown here, the cuttings analysis.
depth of pipe-wear visible in BHTV images often
The depth interval above the first rock type transition significant apparent aperture than the BHTV (Figure
at the top of the images in Figure 3 is also 5). In FMS data this “thickness” is a function of both
distinguished by a pervasive set of sub-parallel the relative conductivity and the actual thickness of
sinusoids only visible in the FMS image. This type of material defining the structure. The measured
thickness could thus reflect the local mineralogy or
nearly sub-parallel (over discrete depth intervals) and fluid chemistry associated with the structure. In
uniformly spaced planar structure consisting of BHTV data this thickness is a function of the
alternating dark and bright bands characterizes broad geometric irregularity of the borehole wall at its
regions of the FMS image log. Caliper logs indicate intersection with the structure and the footprint of the
that these structures have little impact on the acoustic beam. Thus, FMS and BHTV estimates of
borehole wall geometry, consistent with their absence apparent aperture do not directly measure the
in the BHTV log (Figure 3). Crosscutting aperture or thickness of a fracture zone and are not
relationships with natural fractures always indicate directly comparable. Consequently, apparent aperture
that these structures are older. In addition, the only provides insight into the relative importance of
boundaries of these regions generally coincide with fractures and most likely over-estimates true
fault zones and/or changes in image log character thickness or aperture.
(Figure 3) consistent with changing rock types that
are corroborated by the cuttings log (Figure 4). Most Drilling induced structures and stress state
importantly, these characteristics are similar to core In this study we distinguished three types of drilling
and surface outcrops elsewhere in the field that induced structures: failure of the borehole wall as (1)
contain non-brittle fabrics (Figure 3). Therefore, we breakouts and (2) tensile fractures, and failure of rock
interpret these structures as non-brittle fabrics such as ahead of the drill bit as (3) petal-centerline fractures.
foliation that can be distinguished from natural
fractures in image logs because of their consistent (1) Borehole wall breakouts
geometry (attitude and spacing) and absence in the
BHTV log. Introduction of a borehole concentrates the remote
stresses at the unsupported free surface of the
borehole wall. Where the hoop stress exceeds the
Distribution of geologic structures in depth
compressive strength of the rock, compressive failure
Comparison of FMS and BHTV logs from the same of the borehole wall produces breakouts (Zoback et
interval revealed differences in the populations of al., 1985). In near vertical wells, where the
natural fractures and other planar structures imaged overburden represents one of the principal stresses,
by the two techniques (Figure 4). In general, fabric the compressive hoop stress is greatest along the
elements comprise more than 50% of all planar azimuth of the minimum horizontal stress (Shmin).
structures measured in the FMS log, and locally Thus, breakouts indicate the azimuth of Shmin and will
approach 100%. This subset of planar structures tend to occur in pairs 180º apart. The occurrence,
shows consistent orientations over short depth width, and depth of breakouts are primarily a
intervals but systematic changes in dip and dip function of the rock strength, remote stress state
direction with depth (Figure 4). At any given depth, (breakouts being favored in locations with large
the orientation of foliation is clearly distinct from the
orientations of the natural fracture population (Figure
5).

The residual fracture population left by subtracting

these fabric elements from the total FMS data set is
similar to the fracture population interpreted solely
from the BHTV image log in frequency distribution
(Figure 4) and orientation (Figure 5). However, the
FMS data still indicates more fractures than the
BHTV (Figure 4). Both data sets suggest that most
fractures are associated with distinct zones of image
log character, suggesting that there is a correlation
between fractures density and specific rock types or
alteration recorded in the cuttings log.
Figure 6: BHTV amplitude and two-way travel time,
In general, apparent apertures estimated from the and FMS image logs of intervals
FMS logs are larger for small fractures, are smaller containing borehole wall tensile fractures
for large fractures, and show more fractures with and breakouts.
Figure 7: (a) Sketch of petal (green) and centerline (red) fractures visible in (b) borehole televiewer image log at
~7000 ft measured depth from well 58A-10. (c) Petal-centerline fractures are commonly observed in core
(modified from a photo by Al Lacazette, www.naturalfractures.com). (d) Cartoon of stress concentration
around the bottom of a borehole subjected to a mud pressure in excess of the formation pressure based on
studies by Li and Schmitt (1997, 1998).

horizontal stress magnitudes), mud weight relative to oriented parallel to the maximum horizontal stress
formation fluid pressure and well bore orientation. (SHmax). Tensile failure of the borehole wall is also
Observations of breakout width in conjunction with promoted by high mud weights and thermal stress
estimates of rock strength can be used to constrain induced by cooling of the borehole wall during
the range of horizontal stress magnitudes (Moos and circulation of drilling fluids.
Zoback, 1990).
Borehole wall tensile fractures are common in both
In borehole cross section, breakouts typically have a BHTV and FMS image logs of well 58A-10 (Figure
“dog ear” geometry (Zoback et al., 1985) which is 6). We identified this type of borehole wall failure in
only visible in the BHTV log due to the ability of the logged intervals where tensile fractures occurred as
acoustic signal to map the complete borehole shape. paired vertical lineations 180° apart or as en echelon
We limited our picks of breakouts to logged intervals lineations 180° apart.
where breakouts occurred as paired irregularly
shaped patches 180º apart and where they could be (3) Petal-centerline fractures
clearly distinguished from any tool marks. In well
Petal-centerline fractures are interpreted to form
58A-10, breakouts are preferentially distributed
ahead of the drill bit due to the complex
within zones of distinct image log character (Figure
concentration of stress at the floor of the borehole
6), which are typically associated with deeper tool-
(Kulander et al., 1990; Lorenz et al., 1990; Li and
wear than immediately adjacent regions. It is thus Schmitt, 1997, 1998). These induced fractures are
likely that these regions are distinguished by favored in normal to strike slip faulting environments
relatively low compressive strength.
and are promoted by high mud weights and the rough
geometry of the borehole floor (Li and Schmitt,
(2) Borehole wall tensile fractures 1998). They have long been recognized in core, and
Tensile failure of the borehole wall occurs where the both empirical studies of oriented core (Kulander et
hoop stress around the borehole is negative and al., 1990) and theoretical studies (Li and Schmitt,
exceeds the tensile strength of the rock (Moos and 1997, 1998, 1998) demonstrate that they strike
Zoback, 1990). For near-vertical wells, this failure parallel to SHmax. However, they have generally gone
produces vertical tensile fractures 180º apart and un-recognized in both types of image logs. In this
study we take advantage of these fractures and mistake a petal-centerline fracture for a
reaffirm that they provide stress orientations true borehole wall tensile crack (arrows).
consistent with the other induced structures discussed
above. natural fractures in an FMS log. The potential for
such misinterpretation is exacerbated where these
In image logs petal fractures appear as smoothly fractures are nested (Figures 7b and c). Practically,
curving chevrons (pointing up or down). The tips of use of an FMS log alone is thus prone to increased
the chevrons merge with pairs of centerline fractures scatter in measured orientations of Shmin. Using a
oriented parallel to the borehole axis (Figure 7a and BHTV log and applying the same averaging
b) consistent with examples in core (Figure 7c). technique to all tensile fractures, including those
Whereas the symmetrical stress concentration around produced by borehole wall failure, helps limit this
the borehole free surface causes classic borehole wall pitfall.
tensile fractures to form 180º apart, petal-centerline
fractures propagate through intact rock ahead of the Figure 8 illustrates an example where such confusion
drill bit as opening-mode fractures that open in the is likely in the FMS. This example also reveals an
local direction of the least compressive principal abutting relationship visible in BHTV log that
stress (Figure 7d). The resulting fracture plane strikes confirms that classic borehole wall tensile fractures
parallel to SHmax, but does not necessarily lie along form after petal fractures.
the borehole axis. Thus, once the bit has drilled
through the induced fracture, the petal-centerline Orientation of principal stresses
fracture will have variable azimuthal spacing along Analysis of breakouts, pairs of borehole wall tensile
the borehole wall. The orientation of Shmin can then cracks and pairs of petal-centerline fractures from the
only be reliably estimated by calculating the average BHTV logs (Figure 9) indicates that Shmin near well
azimuth of pairs of centerline fractures or paired 58A-10 is oriented along an azimuth of 102º ± 18º. A
measurements on both limbs of petal fractures. large proportion of the tensile fractures evident in
these BHTV image logs are interpreted as petal-
In FMS images, pairs of centerline fractures are centerline fractures. Our analysis of the BHTV data
unlikely to be completely imaged because they need shows that petal-centerline fractures provide
not form 180º apart and FMS pads are spaced at 90º comparable stress orientations to breakouts and
increments (although greater coverage in other borehole wall tensile fractures (summarized in Figure
electrical imaging tools will reduce this limitation). 9). In contrast, the FMS data shows considerably
As a result, centerline fractures spaced less that 180º more scatter and fewer picks because of the poor
could be misinterpreted as classic borehole wall azimuthal coverage of the borehole wall and the
tensile fractures (FMS image in Figure 8). Similarly, inability to distinguish petal-centerline fractures from
petal fractures might mistakenly appear to be parts of borehole wall tensile cracks. The FMS mean Shmin
direction also departs by 13º from the BHTV
analysis. Careful calibration (not presented here)
between the logs demonstrates that they are well
oriented with respect to each other. Thus, this
difference is probably a result of the sampling bias
resulting from FMS pad spacing and the much
smaller population sampled (including lack of
breakouts).

The average azimuth of Shmin is consistent with the

local orientation of the Coso Wash normal fault just
to the west of the well and another similarly striking,
but west dipping normal fault to the east (Figure 1).
Figure 8: BHTV and FMS image logs of classic
Inside the East Flank (boxed area in Figure 1), stress
borehole tensile cracks (blue) abutting
orientations show greater heterogeneity between
against a petal-centerline fracture
wells separated by relatively short distances,
(green). The abutting relationship circled
presumably because of the proximity of active faults.
in red in the sketch confirms that the
petal-centerline fractures formed before
the tensile failure of the borehole wall.
This relationship is not visible in the
FMS image, where one could easily
The depth distribution of induced structures visible in CONCLUSIONS
the BHTV data also shows one prominent stress
FMS logs generally reveal more planar structures
otation at ~9750 ft near the locations of several large
than BHTV logs and are sensitive to characteristics
like foliation, layering, or systematic variations in
resistivity related to mineralogy or fluids. More than
50% of the planar structures imaged by FMS logs in
well 58A-10 appear to be fabric elements not well
imaged by BHTV logs. These fabric elements
probably do not contribute to reservoir productivity.
In contrast, BHTV logs preferentially record open or
un-cemented fractures that are more likely to break
the continuity of the borehole wall; thus it is unlikely
to image fabric. As a result, FMS logs may provide a
more complete record of the entire fracture
population, possibly including healed fractures and
foliation, while BHTV logs more selectively image
“open” natural fractures that are expected to
contribute to reservoir permeability.

FMS and BHTV logs to varying degrees reveal rock

type transitions indicated by changes in image log
character, image quality, and pipe-wear along the
borehole wall and are generally consistent with the
cuttings log. Fault zone textures potentially related to
the slip history and relative ages of fractures are also
revealed in both image logs, although the BHTV is
better at revealing cross-cutting relationships and
non-planar fracture step-overs by virtue of its greater
azimuthal coverage. Combining both techniques
allows a clearer characterization of fault texture that
might reflect the distribution of minerals, fluids, or
porosity.

Azimuthal coverage provided by the BHTV log

reveals a much larger suite of stress state indicators
than the FMS log. Pairs of petal-centerline fractures,
and borehole wall breakouts and tensile fractures
Figure 9: Orientation of Shmin as a function of depth. indicate imaged in the 58A-10 BHTV log an Shmin
The means Shmin orientation shown as azimuth of 102º±18º. This azimuth is generally
dashed red line in the depth section is consistent with image log analyses from other wells
derived only from analysis of the BHTV within the producing geothermal reservoir and is at
logs. Red arrow shows location of a high angle to the strike of the active Coso Wash
significant stress rotation possibly related normal fault.
to slip on one or more active faults
observed at this depth. ACKNOWLEDGMENTS
Steve Bjornstad and Frank Monastero of the Navy
faults visible in the image logs (arrow in Figure 9).
Geothermal Program Office provided data, insight,
Other less prominent stress rotations, such as the
and review that greatly helped with this study. Other
potential rotation at ~7350 ft, are evident, but do not
collaborators include: Coso Operating Company and
stand out as clearly from the scatter. These rotations
the Energy and Geoscience Institute ( U. Utah).
probably indicate active slip on faults nearby or
Funding for N.C. Davatzes was provided by a U.S.
intersecting the borehole.
Geological Survey Mendenhall Fellowship. Judith
Sheridan and Dan Moos provided input on previous
image log interpretations at Coso.
REFERENCES Moos, D. and Zoback, M.D., (1990), “Utiliation of
observations of well bore failure to constrain the
Barton, C.A., Zoback, M.D., and Moos, D. (1995),
orientation and magnitude of crustal stresses:
“Fluid flow along potentially active faults in
Application to continental, deep sea drilling project,
crystalline rock,” Geology, 23(8), 683-686.
and ocean drilling program boreholes,” J. Geophys.
Barton C.A., Hickman, S., Morin, R., Zoback, M.D., Research, 1000(B), 12791-12811.
and Benoit, D., (1998), “Reservoir-scale fracture
Sheridan, J.M. and Hickman, S., (2004), “In situ
permeability in the Dixie Valley, Nevada,
stress, fracture, and fluid flow analysis in well 38C-9:
Geothermal Field,” PROCEEDINGS, Twenty-Third
An Enhanced Geothermal System in the Coso
Workshop on Geothermal Reservoir Engineering,
Geothermal Field,” PROCEEDINGS, Twenty-Ninth
SGP-TR-158, Stanford University, Stanford,
Workshop on Geothermal Reservoir Engineering,
California, January 26-28, 1998. 8 p.
SGP-TR-175, Stanford University, Stanford,
EkstromM.P., Dahan,C.A., Chen, M.Y., Lloyd, P.M., California, January 26-28, 2004. 8 p.
and Rossi, D.J., (1987), “Formation imaging with
Unruh, J.R., Hauksson, E., Monastero, F.C., Twiss,
microelectrical scanning arrays,” Log Analyst, 28,
R.J., and Lewis, J.C., (2002), “Seismotectonics of the
294.
Coso Range—Indian Wells Valley region, California:
Hearst, J.R., Nelson, P.H., and Paillett, F.L., (2000), Transtensional deformation along the southeastern
Well Logging for Physical Properites:A Handbook margin of the Sierran microplate,” in: Glazner, A.F.,
for Geophysicists, Geologists and Engineers: Second Walker, J.D., and Bartley, J.M., (eds), Geological
Edition, John Wiley & Sons, Ltd, New York, NY, evolution of the Mojave Desert and Southwestern
483 p. Basin and Range: Boulder, CO, Geological Society
of America Memoir 195, 227-294.
Kulander, B.R., Dean, S.L., and Ward, B.J.Jr., (1990)
Fractured Core Analysis: Interpretation, Logging, and Wicks, C.W., Thatcher, W., Monastero, F.C., and
Use of Natural and Induced Fractures in Core, , Hasting, M.A., (2001), “Steady-state deformation of
American Association of Petroleum Geologists, the Coso Range, East-Central California, inferred
Tulsa, OK. AAPG Methods in Exploration Series, from satellite radar interferometry,” J. Geophys.
No. 8, 88 p. Research, 106(B7), 13769-13780.
Li, Y. and Schmitt, D.R., (1997), “Well-bore bottom Zoback, M.D., Moos, D., Mastin, L., and Anderson,
stress concentration and induced core fractures,” R.N., (1985), “Well bore breakouts and in situ
AAPG Bulletin, 81(11), 1909-1925. stress,” J. Geophys. Research, 90(B7), 5523-5530.
Li, Y. and Schmitt, D.R., (1998), “Drilling-induced Zemaneck, J., Glen, E.E.Jr., Norton, L.J., and
core fractures and in situ stress,” J. Geophys. Cardwell, R.L., (1970), “Formation evaluation by
Research, 103(B3), 5225-5239. inspection with the borehole televiewer,” Geophysics,
35, 254-269.
Lorenz, J.C., Finley, S.J., and Warpinski, N.R.,
(1990), “Significance of coring-induced fractures in
Mesaverde core, Northwestern, Colorado,” AAPG
Bulletin, 74(7), 1017-1029.