High Quality Electrical Borehole Images While-Drilling Provides Faster Geological-Petrophysical Interpretation, With Increased Confidence

Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

SPWLA 46th Annual Logging Symposium, June 26-29, 2005

BB

HIGH QUALITY ELECTRICAL BOREHOLE IMAGES WHILEDRILLING PROVIDES FASTER GEOLOGICAL-PETROPHYSICAL


INTERPRETATION, WITH INCREASED CONFIDENCE
Jeremy (Jez) Lofts, Stephen Morris, Ren N. Ritter, Roland Chemali, Christian Fulda;
Baker Hughes INTEQ
Drilling called StarTrak1 is part of the ongoing
migration of wireline technology to LWD. This
migration is driven for the most part by expectations of
rig time saving and justified in high cost drilling
environment. One salient feature of the new device is
the continuous full-bore coverage giving a higher level
of confidence in the interpretation of the images.

Copyright 2005, held jointly by the Society of Petrophysicists and Well


Log Analysts (SPWLA) and the submitting authors.
This paper was prepared for presentation at the SPWLA 46th Annual
Logging Symposium held in New Orleans, USA, June 2629, 2005.

ABSTRACT
The drilling environment surprisingly, offers an ideal
platform for electrical borehole imaging. At the time of
drilling, the borehole wall rugosity is often minimal and
electrical images generated by sensors that rotate with
the drill string provide a full coverage of the borehole
(when compared to the pad coverage observed on
conventional wireline borehole images). Field tests and
characterization in a laboratory of a new high resolution
electrical imaging tool deployed while-drilling confirm
its field worthiness as well as accuracy and repeatability
of the images.

The resolution of the electrical images enables detailed


geological evaluation.
Further image interpretation discussed in this paper
highlights the predictive ability of StarTrak images in
terms of image fabric characterization and
interpretation of sedimentary facies and structures. It
also highlights the possibility for Sedimentary Steering
in which the bit may be steered to follow particular
target sedimentary facies, possible because of the high
resolution nature of the sensor.

We show further design improvements and conclude


that with a properly dimensioned imaging sensor, and
use of advanced focusing techniques, a nominal image
resolution comparable to that of wireline borehole
resistivity imagers may be achieved. With a pixel
resolution comparable to that of wireline, we advocate a
more confident understanding of the geological
features, achieved because of the complete borehole
wall coverage.

An important aspect of imaging while drilling is the


real-time information about the subsurface that helps
the field geologist and the drilling engineer recognize
the terrain surrounding the wellbore. This article
address examples of memory recorded logs.
Geosteering applications will be published at a later
date, but resolution here for the first time allows us to
consider the concept of sedimentary steering.

Electrical images from various wells recorded whiledrilling show a broad range of high resolution
sedimentary features including laminated and disturbed
injected/dewatered sands/mud rock, and cross-bedding
within laminated and bioturbated sandstones, as well as
composite fractures, fracture clusters and faulting. With
full borehole coverage textural and facies
discrimination is clear and can be extended to the realtime environment and to consider the concept of
Sedimentary Steering.

IMAGE ACQUISITION
A single sensor placed on the side of the collar scans
the surface of the wellbore and delivers a continuous
stream of data covering the entire circumference of the
borehole. A reasonable rate of penetration provides
continuous scanning of the borehole wall in the axial
direction (Ritter et al. 2004). The maximum pixel
resolution has been calculated at 0.25 (6.4 mm).
Image Processing and Interpretation

INTRODUCTION

The data presented here were processed into images


using Baker Atlass RECALL software with data QC
performed in following methodology of Lofts and
Bourke 1999. After processing, the images were
displayed with interpolation which further enhanced
geological features.

High resolution electrical imaging is now possible


while drilling. A recently developed LWD tool, capable
of producing such image logs in conductive water based
mud was described by Ritter et al. 2004. The design
and deployment of the Electrical Imaging Tool While

StarTrakTM Trade Mark of Baker Hughes INTEQ

SPWLA 46th Annual Logging Symposium, June 26-29, 2005

IMAGE FABRIC CHARACTERISTICS FROM


STARTRAK

hot shale: high gamma response mudrock


indicating a possible maximum flooding surface.

Interpretation comprised three stages:

marine mudrock: a slightly upward decreasing


gamma profile indicating slowly increasing influx
of sandy material. Abundant, low angle lamination
seen on the StarTrak images is indicative of low
energy conditions and accumulation of hemipelagic
(settling) and dilute, muddy underflows of a lower
delta front environment.

upward cleaning sandstones:


mudrock to
sandstone transition with increasingly clean
sandstones upward. Dip magnitudes tend to
increase upwards indicating increased energy and
bedform development. This is shown by Package
5A which is characterized by intervals of flat-lying
dips, interbedded with steeper dips.
This
represents a delta-front deposit with localized
slumping that caused the steeper dips and with
possible distributary channels (infilled with thin
sandstones).

channel sandstone: thick, clean sandstone with


interspersed structureless and structured intervals.
In Sub-package 4A, cross bedding is diagnostic of
strong, unidirectional currents with deposition of
sand in migrating bedforms. The mud drapes
associated with these (along set surfaces and in
discrete layers) may also indicate tidal influence
with relaxation in flow power leading to mud
deposition. This is likely to represent high energy
conditions of mouth bars and distributary river
channels.

coal/limestone:
elsewhere in the interpreted
interval there are coal and limestone beds at the
tops of cycles. These facies indicate emergence
with plant growth and accumulation/open, shallow
marine conditions respectively and indicate
maximum outbuilding of the delta front before
subsidence and resumption of deep marine
conditions in the following cycle.

dip picking of the images using a dip scheme


classified according to the lithology, bed boundary
styles and dip magnitudes (DOI = 0.4 in),

automated thresholding of the gamma log to obtain


log facies and

separate designation of fabrics from the images


(in which lithology was a minor factor).

Test Well Data


The tool was field tested in the Mounds area of
Oklahoma State, USA. Mudrock dips picked from the
images allowed the accurate calculation of structural
dip which can be used to rotate sedimentological dips
back to their original orientation. This was not done in
the test well due to its low magnitude. The structural
dip was comparable to that measured from the wireline
run.
Lithological Interpretation
The area is interpreted as a stack of deltaic mudrocks
and sandstones with classic, repeated-upwards decline
in gamma log response that reflects increasing
sandstone content. The high gamma (hot shale)
intervals are likely to represent maximum flooding
surfaces (Figure 1 at 2424 ft), and are the deposits
associated with the deepest marine water when relative
sea-level was highest. The bases of large sandbodies
may be sequence boundaries that formed when the
relative sea level dropped rapidly and channel/deltaic
sandstone then encroached basinward over previous
deposits (Figure 3 at c. 2400 ft).
StarTrak images have enabled a confident
sedimentological interpretation of the test well using a
combination of the distribution of StarTrak image
fabrics, associated dip patterns and gamma log facies
(Figure 1). These were subsequently used to divide the
interpreted section into genetic sedimentary packages.
In the case of shallow marine (deltaic) sedimentary
systems, this packaging is a strong indication of
sequence-stratigraphic cyclicity (cf. Emery and Myers
1996) and is caused by relative sea-level changes. In
addition, several packages have been subdivided as they
are likely to correspond to single, smaller deltaic
parasequences.

Image Fabrics
The interpretation of the test well showed a wide range
of image fabrics that can be resolved and identified
with the tool, and these have been designated largely
independent of the host lithology (Figure 2):
STRK-01 Structureless: uniform or slightly variable
resistivity response with no discernable fabrics. In rare
examples, a blotchy texture may be developed although
no sinusoids can be fitted to the features. This lack of
fabric may indicate a truly structureless lithology or
possibly a previously structured lithology that has been
modified (e.g. by bioturbation, where the new fabric is
below the resolution of the tool).

Discrete components of deltaic sequences are


identifiable from the gamma log profile and the
distribution of sedimentary structures derived from
images (facies interpretations: Figure 1):

SPWLA 46th Annual Logging Symposium, June 26-29, 2005

BB
contact. The equivalent core example shows a water
escape structure caused by rapid sand deposition which
may also be associated with larger-scale remobilization
and slumping. In addition, bioturbation and the
development of cementation fronts within compacted
sediments may also exaggerate the visibility of
contacts.

STRK-02 Weakly laminated: sandstones commonly


display intervals of weak lamination in which there is
poor resistivity contrast between individual layers, or
the lamination is fuzzy due possibly to modification.
The example shows a well-defined sedimentary surface
at c. 2105.4 ft dividing a bright (resistive) response
facies from weakly laminated facies below. The fabric
is mostly blotchy with several poorly resolved
sinusoids. Weak (relict) structures such as these may
have developed in structured sandstones that do not
comprise substantially differentiated material and
which may therefore have poor resistivity contrast.
From core, some weakly laminated sandstones
contained mm-scale mudclasts along sedimentary
surfaces which may account for the blotchy nature in
images.

STRK-07 Patchy: this fabric is characterized by smallscale patches of alternating resistivity with no overall
organization and containing, at most, very weaklydefined (relict?) structures.
In core, the fabric
corresponds to intervals in sandstones that contain
patchy and disorganized fabrics, distributed mudclasts
and bioturbation, all of which may be exaggerated by
patchy cement development.
STRK-08 Vuggy:
a distinct fabric that can be
differentiated clearly from the Patchy type, with wellresolved, dark (conductive), circular or slightly oblate
patches. The lack of resistive halos indicates that these
are vugs (hollows) in the host lithology and generally
coincide with limestone. The vugs are commonly
concentrated along weakly-defined lamination and are
likely to indicate differential dissolution of various
components of the limestone.

STRK-03 Strongly laminated: both sandstones and


mudrocks commonly display strongly-developed
lamination or other sedimentary structures where thin,
high resistivity contrast alternations are seen on images.
Individual lamina can be defined down to a single pixel
which equates to a vertical resolution of c. 7.6 mm.
Such fine lamination, as seen in the example, is likely
to comprise interlayered mudrock and silty bands or
sandstones containing clay-rich surfaces, both of which
give good resistivity contrasts.

STRK-09 Uninterpretable: rare intervals where image


quality is poor and the fabric cannot be seen, although
these are usually overlapped by subsequent drilling or
relog passes over which good images were acquired.
Typical artifacts include poor resolution due to low tool
rotational velocity during run-up and rare stick-slip
effects (zig-zag distortion). Borehole spiraling is rare
in the test well but its presence does not affect image
interpretation to a great extent (Lofts and Bourke 1999).

STRK-04 Banded: a common fabric that comprises


parallel zones of similar resistivity in bands that are
thicker than average lamination. The example shows
two, c. 0.4 ft thick, bright (resistive) bands, the lower of
which contains sparse lamination. The bright bands are
interlayered with more conductive material which is
variable in character (laminated, structureless to blotchy
and vuggy). The bright and dark banding may be due
to different lithologies or, as shown in the core
example, by cement zones within the host sandstone.

APPLICATIONS OF FABRIC
INTERPRETATION: TEST WELL

STRK-05 Laminated with resistive halos: resistive


halos around conductive spots are common in the test
well and commonly occur along laminae. These
comprise a dark (conductive) spot, commonly
elongated along the lamina and surrounded by a bright
(resistive) halo. The halo is an artifact of resistivity
imaging devices caused by current gather into the
conductive spot, in this case the pyrite nodules seen in
the core. Commonly, the halo diminishes slightly along
the host lamina due to its slightly more conductive
nature (e.g. 1185.2 ft).

The high resolution of the StarTrak tool allows the


interpretation of detailed sedimentary and structural
history using lithological and image fabric
identification. This is illustrated using examples from
the test site and the North Sea (Figures 3 and 4).
A detailed interpretation of a 17 ft representative
section of sandstones from the test well is shown in
Figure 3 and compares the StarTrak images to the
wireline STAR Imager2, image fabric interpretations,
facies associations and core from the offset well. Four
main sedimentary fabrics have been focused on here,
although the images display additional variations.
These fabrics form a larger-scale genetic facies
association which constitutes the channel sandstone
seen in Figure 1:

STRK-06 Irregular blotchy: intervals characterized by


large (borehole-width) irregular or convoluted fabrics
and irregular, non-planar, highly disrupted contacts
between lithologies. The example shows a resistive
lithology containing high dip magnitude, non-planar
structures overlying a less structured conductive
lithology with a highly convoluted and distorted

Mark of Baker Atlas

SPWLA 46th Annual Logging Symposium, June 26-29, 2005

1659.2-1660.6 ft: a unit of irregular (and slightly


blotchy) fabric within relatively conductive and
mud-rich sandstones with well-defined boundaries.
The dip magnitude is greater than in surrounding
fabrics although not entirely chaotic. This is likely
to be a slumped unit in which unconsolidated sand
may have been remobilized and flowed as a
coherent unit. Truncation is seen at the upper
surface, indicating partial erosion by the flow that
deposited the overlying, laminated sand.

wall and therefore measurement of the more conductive


drilling mud. The geometry of these features is
consistent with the development of borehole breakout
over the weaker mudrock lithology and its position in
the image shows that it is lateral (at the sides of the
borehole). The detection of borehole breakout in such
detail can provide valuable geomechanical information
that can be used to predict borehole stability and, in
real-time, could be used to prompt a counteractive
increase in mud weight.

1660.6-1662.5 ft: an interval of laminated and


banded fabrics that are parallel and display distinct
alternations in resistivity. Hints of subtle variation
in fabric may also been seen within some of the
bands and may indicate smaller-scale structures
such as ripples or mudclast layers.

Drilling-Induced Fractures

1662.5-1664.5 ft: a spectacular example of cross


bedding within laminated sandstones displaying
parallel set-bounding surfaces between which
steeper-dipping foresets (the remnants of accreting
bedforms) are preserved at clear truncation
surfaces. The cross bedding is displayed by subtle
variations in resistivity response that may be a
result of preservation of mud drapes, small
mudclasts or different grain sizes along the
bedform foresets.

1669.0-1671.0ft: a zone of laminated sandstone


with variable dip magnitudes and localized blotchy
fabrics. This indicates the original deposition of
flat, parallel laminated sand followed by partial
disruption that may be due to pore water expulsion
and partial remobilization. In addition, the growth
of small, in-situ, non-displacive cement zones may
also disrupt the sedimentary fabric.

Other mudrock intervals contains a range of natural and


induced fractures that are not present in the sandstone
intervals (Figure 4b). Several steep, full boreholewidth to partial, and variably intersecting open
fractures can be seen clearly as dark, conductive
sinusoids (although the width of these features is likely
exaggerated by the tool due to current gather). On the
bottom side of the borehole a non-linear fracture can be
identified which is broadly parallel to the borehole axis
but also contains an offset along a natural fracture and
an abrupt termination against another. The
characteristics and position of this feature (relative to
the borehole breakout) indicates that this is a drillinginduced fracture and also has implications for
geomechanical modeling.
3D
Visualization
and
Sedimentary Steering

Real-time

Images:

The high-resolution images can be used to construct


tube-type 3D projections. These can be rotated and
manipulated to aid in geometric interpretations of
sedimentary structures and reservoir architecture
(Figure 5a).
The acquisition of image data in real-time also allows
the possibility of sedimentary steering by guiding the
drill bit along a preferred sedimentary fabric (Figure
5b) or within a well-constrained zone within a
sandbody (Figure 5c). Sedimentary steering could be
used to accurately position the well to maximize
coverage within a section of a sandbody that contains
the best reservoir quality. In addition, the orientation of
entry and exit bounding surfaces between sandbodies
and mudrocks can be used for the calculation of
sandbody geometries (Figure 5d; cf. McGarva et al.
1999).

DRILLING-RELATED FEATURES: NORTH SEA


WELL
The successful acquisition of good-quality and useful
image data in the North Sea (Ritter et al. 2004) shows
the tools capability in the adverse conditions of a
deviated (horizontal) borehole in a highly resistive
formation (Rm:Rt 2000; Figure 4). The predictive
ability from this well includes the following drilling
related features, which if seen in real-time can serve to
mitigate drilling hazards.
Borehole Breakout

CONCLUSIONS

A c. 10 ft example of the borehole intersecting a sliver


of mudrock is shown in Figure 4a. The two sandstonemudrock contacts of the mudrock interval can be seen
in the images, they are non-parallel and the up-hole
contact is also irregular. Importantly, the StarTrak tool
has imaged dark, elongate, sub-parallel strips within the
mudrock which indicate slight spalling of the borehole

The high resolution of StarTrak has allowed the


identification of detailed image fabric types (8 in
the test well). From these image fabrics we can
make confident interpretations of key sedimentary
structures similar to that derived from wireline
image logs (especially with field or core
calibration).

SPWLA 46th Annual Logging Symposium, June 26-29, 2005

BB

ABOUT THE AUTHORS

When combined with lithological variation, image


fabrics form the elementary building blocks with
which to characterize the sedimentological
variation in a reservoir.

Jeremy Lofts received his PhD from Leicester, UK in


1993 and he is a Chartered Geologist. He has worked as
an image interpretation specialist and he lectures
externally/internally on the subjects of Borehole Image
Interpretation. Jez is author of +20 papers on geological
and petrophysical oilfield applications. He is currently
the Marketing Director, LWD Product Line Baker
Hughes INTEQ, USA Headquarters.

The tool is capable of acquiring drilling-decision


information even within hostile environments (such
as deviated and harsh resistivity-contrast wells).
DIF, borehole breakout and natural fractures are
clearly resolved by the tool. In a similar fashion to
wireline tools, fracture aperture is exaggerated by
current gather effects.

With 3D visualization it is possible to interpret the


internal architecture of sedimentary deposits and
the larger-scale sandbody geometry that is critical
for reservoir development.

The interpretation of image fabrics and image


facies in real-time gives the potential for
sedimentological steering in which the drill bit is
guided through a particular sedimentary
characteristic that may have favorable properties
for well positioning and reservoir development.

Stephen A. Morris received his PhD from Cardiff, UK


in 1998 and has authored papers on sedimentary
processes. He is a specialist in deep-water sedimentary
processes (turbidite systems). He has previously
worked on geological interpretation of wireline and
LWD borehole images and now works as an imaging
specialist at Baker Hughes INTEQ, Houston.
Ren N. Ritter received his Dipl.-Ing. in Engineering
from the University of Lbeck, Germany in 1997. From
1997 to 2001 he held various Technical and engineering
roles in INTEQ. He is now the Project Manager within
the Strategic Technology Development department of
Baker Hughes INTEQ, Celle.

ACKNOWLEDGMENTS

Roland Chemali received his engineering degree from


the Ecole Polytechnique of Paris. He has authored over
20 papers and patents in electrical and acoustic logging.
He is currently Product Line Manager for Emerging
Technologies at Baker Hughes INTEQ, Houston.

The authors wish to thank ChevronTexaco Upstream


Europe and partners for permission to use the North Sea
data.
REFERENCES

Christian Fulda received his M.Sc. in Physics and his


Ph.D. at the University of Heidelberg, Germany in
1998. During his post doctoral work at the LeibnizInstitute for Applied Geosciences, Hanover, Germany,
he was associated lead of the Research Topic
Groundwater. He is currently the Project Manager for
the Advanced Resistivity Measurement R&D team
within the Strategic Technology Development
department at Baker Hughes INTEQ, Celle.

EMERY, D & MYERS, K.L., 1996. Sequence


stratigraphy. Blackwell Science. 297p.
LOFTS J.C., & L. B. BOURKE, 1999. The recognition
of artefact images from acoustic & resistivity
devices, In: Lovell, M. A., Williamson, G. &
Harvey, P. K. (Eds).
Borehole Imaging:
applications and case histories. Geological
Society, London, Special Edition, 159, 59-76.
MCGARVA, R.M., BELL, C. & BEDFORD, J. 1999.
Use of Resistivity At Bit (RAB) images within an
Eocene submarine channel complex, Alba Field,
UKCS. In: Lovell, M. A., Williamson, G. &
Harvey, P. K. (Eds).
Borehole Imaging:
applications and case histories. Geological
Society, London, Special Edition, 159, 177-189.
RITTER, R.N., CHEMALI, R., LOFTS, J., GOREK,
M., FULDA, C., MORRIS, S. & KRUEGER, V.
2004. High resolution visualization of near
wellbore geology using while-drilling electrical
images.
SPWLA
45th
Annual
Logging
Symposium. Later published in Petrophysics, 46,
85-95.

SPWLA 46th Annual Logging Symposium, June 26-29, 2005

Figure 1. Composite StarTrak image interpretation from the test well showing static-normalized images, dips and
dip summaries, image fabrics, facies interpretations, larger-scale sedimentological packaging and sequence
stratigraphic interpretation.

SPWLA 46th Annual Logging Symposium, June 26-29, 2005

BB

Figure 2. Examples of image fabrics seen in the test well. Images are shown with static- and dynamic
normalization alongside representative core. The table includes fabric description, interpretation of sedimentary
structure type and interpretation of sedimentary processes.

SPWLA 46th Annual Logging Symposium, June 26-29, 2005

Figure 3. Large-scale interpretation of image fabrics and sedimentary structures in a c. 17 ft example from the test
well, compared to the wireline image. Four key image fabrics are exemplified in the images with example core
sections, although other fabrics are also present.

SPWLA 46th Annual Logging Symposium, June 26-29, 2005

BB

Figure 4. Interpretations of two key sections of image data from the North Sea well. 4a shows where the borehole
has penetrated through (or close to) a c. 10 ft sliver of mudrock that penetrates downward into the reservoir
sandstones. Lateral borehole breakout is seen in the mudrock but not in the more competent sandstone. 4b shows
where the borehole penetrated a c. 20 ft long mudrock interval containing an injected sandstone bed, open fractures
and an axial drilling-induced fracture terminating on a natural fracture.

SPWLA 46th Annual Logging Symposium, June 26-29, 2005

Figure 5. Example of rendering of StarTrak image data in a tube-like projection (5a). This adds greater 3dimensional visualization of sedimentary structure geometries. This example shows discordant surfaces due to
likely cross bedding and a high-angle, resistive fracture that cuts through sedimentary features. Such visualization
can be used to guide a well trajectory through specific sedimentary facies: Sedimentary Steering (5b). Additionally,
the bit could be steered to stay within a sandbody (5c). On a larger-scale, borehole entry and exit orientations to
sandbodies can be used to calculate sandbody geometries (5d).

10

You might also like