Copyright 2005, Society of Petroleum Engineers This paper was prepared for presentation at the 2005 SPE Annual Technical Conference and

This paper was selected for presentation by an SPE Program Committee following review of information contained in a proposal submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to a proposal of not more than 300 words; illustrations may not be copied. The proposal must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

Abstract A new logging-while-drilling (LWD) tool that incorporates

directional antennae and long measurement spacings has been

developed and field tested. The directional electromagnetic

(EM) tool measurements are more sensitive to approaching

resistivity boundaries than existing propagation resistivity

tools. Combining measurements from symmetrically arranged

pairs of antennae further amplify this boundary effect while

minimizing undesirable sensitivity to dip and anisotropy.

Novel data processing and structure visualization software was

developed to aid the decision-making and planning process.

Field test results from Oman and the North Sea illustrate how

the directional EM measurements fulfill the requirements for

geosteering in thin, dipping, and curving targets with lateral

resistivity variations. In addition, the directional EM tool also

enables improved characterization of resistivity and resistivity

anisotropy in high-angle and horizontal wells.

Introduction In the past decade, we witnessed increased activity and

progress in the development of EM-sensor technology for well

logging applications. In wireline area, the major development

was the 3D induction, designed primarily for detecting

anisotropy in vertical wells, 1,2 by utilizing a set of tri-axial

antennas or tilted coil antennae. 3,4 Having a full set of triaxial

measurements opens new possibilities in formation

evaluation. Use of cross-dipole couplings, in particular, brings

new quality and allows building more accurate structure

models. 5,6 This becomes very important as the number of

wells drilled at high deviation and horizontally increase.

However, these measurements have not been fully utilized

due to limitations in sensor design, and big environmental

effects, such as borehole eccentering and invasion. 2

LWD technology has matured and reached the quality of

wireline measurements. 7-9 It offers a clear advantage of

providing real-time answers while drilling, and being less

affected by the environment. That is especially important for

well placement and high-angle and horizontal well

applications. 10,11 The objective there is to stay in the sweet

spot of the reservoir, or at a defined distance with respect to

geological boundaries, where the decision to steer the well up,

down, left, or right has be made in real time.

Conventional geosteering relies on logs from an offset well

or a pilot well, 12 and the use of imaging technology.

13-15

Typically, it is assumed that layered structure extends to the

targeted reservoir without much variation. This assumption is

often not valid, particularly in wells whose horizontal length

may be on the order of kilometers. The geosteering also uses

resistivity logs with horns as indicators of boundary proximity. 16 This indicator is not quantitative (does not indicate

distance to the boundary), the tool must be very close to the

boundary, and its presence depends on other factors. On the

other hand, correcting the shoulder-bed effect is a major

problem in high-angle and horizontal well applications. 17,18

Anisotropy in electric conductivity and permittivity of

surrounding medium make the quantitative use of horns even

more difficult. 12,18

Traditional LWD resistivity measurements have proven

to be insufficient for steering wells due to the limited depth

of investigation and lack of directionality, i.e., the

measurements are insensitive whether the boundary is

approached from above or from below.

A first-generation directional EM measurement-while-

drilling tool has been developed and successfully field-

tested.19-21 The tool allows well placement to be optimized in

real time by mapping distances to geological boundaries.

The new LWD measurement technology is based on

novel tilted and transverse antennae and symmetric

transmitter-receiver configurations. In addition, new data

processing and structure visualization software was developed

to aid the real-time decision-making and planning process.

In favorable conditions, such as in thick resistive beds,

measurements are able to detect conductive boundaries at

distances greater than 15 ft. In typical geosteering scenarios,

when sensing boundaries are tens or hundreds of drilling feet

away, proactive steering decisions can be made to avoid

unwanted structures.

New interpretation software has been developed to

translate measurements into real-time structural maps. The

SPE 97045

Deep Directional Electromagnetic Measurements for Optimal Well Placement

D. Omeragic, SPE, Schlumberger-Doll Research, and Q. Li, L. Chou, L. Yang, K. Duong, J. Smits, SPE, T. Lau, C.B. Liu,

Exhibition held in Dallas, Texas, U.S.A., 9 – 12 October 2005.

SPE, R. Dworak, V. Dreuillault, J. Yang, and H. Ye, Schlumberger Sugar Land Technology Center

2

data processing is based on a model-based (parametric)

inversion algorithm using a simple three-layer model. At each

measurement point, data are inverted to obtain distance to

nearby boundaries, horizontal and vertical resistivity of the

bed, as well as the resistivities of the beds above and below

the measurement point. This interpretation facilitates proactive

geosteering decisions and provides significant geological

insight into the reservoir structure, which benefits future well

planning.

Directional EM Tool The basic receiver-transmitter layout of the new deep

directional EM tool is shown in Fig. 1. The measurement

system includes the set of conventional propagation resistivity

measurements, with the antennas aligned to the tool axis; i.e.,

transmitters T1−T5 and receivers R1, R2. 8 At both ends of the

tool are two tilted receiver antennae R3, R4, inclined 45° with respect to tool axis, and the transverse transmitter T6.

R3 T5 T3 T1 R1 R2 T6 T2 T4 R4

Fig. 1— Directional EM tool transmitter and receiver layout showing the axial and transverse transmitter antennas and tilted receiver antennae

In addition to the 2−MHz and 400−kHz standard operating frequencies of conventional resistivity EM tools, , a 100− kHz frequency has been added in the new directional EM tool. The

tool design is optimized to maximize depth of investigation,

defined by the 96-in. longest T−R spacing (T5−R4 and T4−R3); the other available spacings are 84, 34, and 22 in. The

transverse transmitter operates only at 100 kHz and 400 kHz,

and provides anisotropy, as well as directional

(nonsymmetrical) measurements, with spacings 74 and 44 in.

The azimuthal orientation of the tool is provided by a

magnetometer. Other sensors include a directional gamma ray

detector and an annular pressure sensor.

Directional EM Tool Antennae

Tilted and transverse antennas are the key enabling

technologies in the directional EM tool development. The

sensors use titled or saddle coils, covered by special antenna-

protection shields, with sloped slots or with transverse slots. 22

The shields and slots are optimized by using 3D finite-element

(FEM) analysis software to produce minimal attenuation and

distortion of the EM field. These models are shown in Fig. 2.

The modeling and experiments confirmed that the point

magnetic dipole is a good representation of these antennae. 22

a b

Fig. 2—−−−−Directional EM antenna: (a) 45°−tilted antenna shield with the sloped slots over tilted coils; (b) Transverse antenna with saddle coils covered by shield with transverse slots

Directional EM Measurements Standard Born response theory used to analyze conventional

induction measurements23 sensitivities can be used to evaluate

triaxial induction couplings. 5 It is well known that the

crossdipole couplings are zero in a homogenous medium and

in certain symmetric structures.

Base XZ coupling, where X and Z are orientations of the

coils, and Z is tool axis orientation, have cos φ sensitivity, where φ is the tool azimuth measured with respect to X-axis. It means that the contribution of domains above and below the

tool has different sign, therefore, giving basic directionality

(up vs. down) information. That is the sensitivity needed to

place the well in simple layered formations.

Crossdipole transverse couplings XY have cos 2φ (or quadrant) sensitivity. This class of measurement is sometimes

referred to as second-harmonic measurements, while

couplings of axial and transverse antennas are first-harmonic

measurements.

If propagation-style measurements, typically used in LWD

tools, are composed using crossdipole couplings, the

directionality will be lost. The alternative is the use of a tilted

antenna and to take advantage of tool rotation.4

The rotation allows for making measurements with the

antennae at virtually any tool orientation and bedding. 4,24

Directional measurements are composed of a ratio of voltages

at two different tool azimuths (up and down).

The equivalence with XZ induction coupling comes from:

180ln 8.68

2 2ln ln 1

UP

DOWN

ZZ ZX ZX ZX

ZZ ZX ZZ ZX ZZ

VAttn i PhaseShift

V

V V V V

V V V V V

π

= −

+= = + ≅ − −

(1)

where VUP and V

DOWN are voltage measured in “up” and

“down” tool position in the bedding plane, respectively.

Fig. 3 shows the sensitivity of directional measurements

using an axial transmitter and 45°-tilted receiver, assuming zero background conductivity. In that case the sensitivity of

axial-tilted directional measurements is identical to induction

XZ sensitivity.

SPE 97045

3

-0.0001

-0.001

-0.01

-0.1

1

0.1

0.01

0.001

0.0001

Sensitivity

Fig. 3—−−−−Sensitivity of the up and down directional axial-tilted pair of measurements

One should note that contributions from different sides of the

tool have different signs, but also, most of the signal comes

from the area close to the axial antenna.

Axial-tilted directional propagation measurement

responses in a 20-Ωm formation, 20 ft thick, between 2-Ωm and 5-Ωm shoulder beds are shown in Fig. 4. The responses are calculated for the longest spacing of 96 in., and at all three

operating frequencies. The signal amplitude changes sign

depending upon whether the conductive boundary is

approached from above or below. The responses are functions

of conductivity difference and frequency. If the ratio L/δ, where L is tool spacing and δ is the skin depth, for more conductive layers is too high, the response may be more

complicated and more difficult to interpret. Note that in the

case of directional measurements, it is the skin depth of the

more conductive layer that defines the signal magnitude,as is

the case for the 2-MHz response at the 2-Ωm boundary where responses are not monotonic. The depth of investigation of the

measurements is also a function of spacing and frequency, or

L/δ, as with conventional resistivity tools.

2 Ωm 20 Ωm 5 Ωm

2 Ωm 20 Ωm 5 Ωm

Fig. 4—−−−−Directional propagation responses for 96-in. T−R spacing when crossing a 20-ft bed. The tool is parallel to the boundaries.

Symmetrization of Directional Measurements

The simplicity of directional measurement response holds if

the formation is isotropic. However, these measurements

exhibit extreme sensitivity to anisotropy if the tool is not in the

transverse or the vertical plane. Fig. 7 shows the directional

response in an anisotropic formation, for different well

inclinations. It is obvious that there is high risk of

misinterpreting the anisotropy and the boundary effect.

Symmetrization of directional measurements exploits the

remarkable relationship of XZ−to−-ZX coupling. Voltage VXZ-VZX is insensitive to anisotropy and dip at any angle if all

of the coils are in the same medium.25 Its propagation

counterpart uses the same concept dealing with axial-tilted

couplings to significantly reduce the sensitivity to dip and

anisotropy.26

Fig. 5 illustrates the combination of two symmetric

measurements made with tilted antennas. If the two

measurements are subtracted, the tool reads zero far from

boundaries in anisotropic formations at any angle. Fig. 6

shows the sensitivity of such symmetrical measurements.

Compared to Fig. 3, the sensitivity is symmetric with respect

to the transverse plane at the tool’s midpoint, and the half-

space contribution from one side of the tool is all positive or

negative.

θθθθ

θθθθ

Fig. 5—−−−−Symmetrical directional measurements with tilted and axial antennas. Solid and dashed arrows indicate the relative up and down tool orientation with respect to layering.

-0.0001

-0.001

-0.01

-0.1

1

0.1

0.01

0.001

0.0001

Sensitivity

Fig. 6—−−−−Sensitivity of symmetrical directional measurements with axial-tilted pair of antennas

The directional propagation response of 96-in. spacing in a

three-layer anisotropic formation is shown in Fig. 7. The

symmetrization practically removes the anisotropy and dip

effect.

SPE 97045

4

2 Ωm Rh=4 Ωm Rv=20 Ωm

1 Ωm

2 Ωm Rh=4 Ωm Rv=20 Ωm

1 Ωm

Fig. 7—−−−−Directional attenuation responses for 96-in. T−R spacing at 400 kHz in an anisotropic formation for various tool inclinations: (a) single T-R pair; (b) the symmetrical measurements

Another nice feature of symmetrization is the simplicity of

responses when the tool crosses a boundary. Directional

responses are almost independent of tool position and

practically linear for dip angles below 35° as illustrated in Fig. 8. This simple dependence on structural dip when the tool

is crossing a boundary is extremely useful for slightly

deviated and near-vertical wells where images are not reliable

for estimating local dip.

1 Ωm 10 Ωm

1 Ωm 10 Ωm

Fig. 8—96-in., 100-kHz symmetrical directional measurement responses to a boundary when the coils are crossing the boundary. Directional responses are scaled with the dip.

Fig. 9 illustrates the features of symmetrical directional

measurements, which are insensitive to dip (α) and resistivity anisotropy when all antennae are in the same bed. When

crossing a boundary, the measurements become linearly

dependent on dip.

T2

T1

R1

R2

Meas = f ( α ,R u ,R t )

R u

R t

Dip detection Boundary detection

T1R1

R2 T2

Meas=f(h, Ru, Rh)

Independent of α, Rv

Rh, Rv

Ru

h

Fig. 9—−−−−Symmetrical directional measurements for boundary detection and dip estimation

Anti-symmetrization of Directional Measurements

When generating symmetrical directional measurements, the

two base directional measurements from Fig. 5 were added to

remove the sensitivity to resistivity anisotropy and dip. If two

base directional measurements are subtracted, the resulting

measurements have increased sensitivity to anisotropy and

dip, and reduced sensitivity to boundaries.

Fig. 10 shows a response comparison of symmetrical and

antisymmetrical axial-tilted directional measurements for 84-

in. T−R spacing, 100-kHz transmitting frequency, in a 20 ft-bed, with Rh=5 Ωm, Rv=10 Ωm, and shoulder beds of 2 Ωm and 1 Ωm, respectively. The responses are normalized with respect to dip, andboth measurements scale linearly with the

dip. The symmetrical signal is proportional to dip when the

antennas are on opposite sides of the boundary, and

antisymmetric measurements have linear dependence on dip

when the tool is crossing the boundary. In a transversely

isotropic (TI) layered medium, symmetrical measurements can

be used to obtain the structural dip, and antisymmetric

measurements can be used to acquire the anisotropic

resistivities.

2 Ωm 1 Ωm

2 Ωm Rh= 5 Ωm

Rv=10 Ωm 1 Ωm

antisymmetric

symmetric directional

Fig. 10—Symmetrical and antisymmetric response in a 20ft thick anisotropic bed with isotropic shoulderbeds.

SPE 97045

5

Measurement Definition and Downhole Processing Raw measurements are continuously acquired while the tool is

rotating. The tilted receiver voltages vary azimuthally, and

depending on the sensitivity of particular coupling, they may

contain a first harmonic (for axial transmitters T1−T5) or first and second harmonic (for transverse antenna T6). The voltages

are fitted as measurements are acquired. The fitting algorithm

output Fourier coefficients for each frequency, f, transmitter, t,

and receiver r:

( ) ( ) 2

0

1

( , , ) cos sink k

k

V f t r a a k b kφ φ=

= + +∑ , (1)

where φ is the tool face angle and ai, bi are complex coefficients; i.e.,

i REi IM ia a ia= + and i RE i IM ib b ib= + .

For each measurement (f, t, r) channel, the orientation of

layering with respect to tool reference for the first and second

harmonics can be obtained by weighted averaging:

2 2 2 2

1 1 1 11 11 11

1 1 1 1 1 1

( , , ) tan tanRE RE IM IMRE IM

RE IM

a b a bb bf t r

a b a a b aφ − −+ +

= ++ +

(2)

and 2 2 2 2

2 2 2 21 12 22

2 2 2 2 2 2

1( , , ) tan tan

2

RE RE IM IMRE IM

RE IM

a b a bb bf t r

a b a a b aφ − −

+ + = +

+ +

.

(3)

where φ1(f, t, r) and φ2(f, t, r) are boundary orientation for the first and second harmonic directional measurements,

respectively.

From these angles, one can compute the base voltages in

the bedding plane, leading to propagation style directional

measurements for first and second harmonics:

0 1 1 1 1

1 10

0 1 1 1 1

0 1 1 1 1

1

0 1 1 1 1

cos sin( , , ) 20log

cos sin

cos sin( , , )

cos sin

RE

RE

RE

RE

a a bAtt f t r

a a b

a a bPS f t r Arg

a a b

φ φ

φ φ

φ φ

φ φ

+ +=

− −

+ += − −

, (4)

and

0 2 2 2 2

2 10

0 2 2 2 2

0 2 2 2 2

2

0 2 2 2 2

cos sin( , , ) 20log

cos sin

cos sin( , , )

cos sin

RE

RE

RE

RE

a a bAtt f t r

a a b

a a bPS f t r Arg

a a b

φ φ

φ φ

φ φ

φ φ

+ +=

− −

+ += − −

. (5)

where Att1(f, t, r), PS1(f, t, r) and Att2(f, t, r), PS2(f, t, r) are the

directional attenuation and directional phase shifts for the first

and second harmonic directional measurements, respectively.

Because the measurements are ratio-based, the drift in

electronics is significantly reduced, as it is for borehole

compensated resistivity measurements.

Measurement Interpretation When processing conventional resistivity data, raw

measurement signals are mapped to apparent resistivities using

a resistivity transform. In other words, an unbounded

homogenous medium model is used to interpret each

measurement independently.

There is no equivalence to apparent resistivity when

interpreting directional EM measurements because the

measurements are zero if there is no boundary. The simplest

model for interpreting the measurement is a single-boundary

model, which requires at least three parameters to be

described: distance, bed resistivity, and shoulder-bed

resistivity. Therefore, if resistivity of the bed or the shoulder is

known, one additional measurement is required. These

models can be parameterized to read a resistivity and a

distance to the boundary, using any pair of available

measurements.

Crossplot Chart-Based Interpretation

Two types of crossplot charts are commonly used in

interpretation of new directional EM tool measurements.

They combine either two directional measurements or a

directional and conventional resistivity measurement.

If shoulder-bed resistivity is known, the use of one

resistivity measurement is recommended. Fig. 11 shows a

corresponding crossplot for shoulder-bed resistivity, where

Rh=0.8 Ωm, Rv=3.2 Ωm, using 84-in. T−R spacing, 400-kHz directional attenuation, and apparent phase shift resistivity

(RPS) from the 28-in. spacing, 2-MHz T−R. The chart also includes measurements and value of distance and resistivity

read from the chart. It should be noted that use of this chart

produces the shoulder-bed free-bed resistivity if the resistivity

measurement has not peaked out. A sample point on the chart

corresponds to an apparent resistivity value of about 80 Ωm, but the crossplot value is 20.48 Ωm.

Rres=4-40 Ωm

Rh=0.8 Ωm, Rh=3.2 Ωm

h=0 - 12 ft

Fig. 11—−−−−Crossplot chart from 400-kHz, 96-in. T−R spacing directional attenuation and conventional 2-MHz PS resistivity to obtain distance to boundary and bed resistivity, assuming shoulder-bed

resistivity of Rh=0.8 Ωm, Rv= 3.2 Ωm

SPE 97045

6

If bed resistivity is known, it is recommended that a pair of

directional measurements be used in a crossplot chart to

obtain shoulder-bed resistivity and distance to boundary. Fig.

12 illustrates the use of 84-in., 100-kHz directional phase shift

and attenuation for a 1-Ωm shouler-bed resistivity. If the shoulder bed is less resistive, it is not necessary to know the

bed resistivity precisely because the responses are nearly

proportional to conductivity difference.

Rres=15 Ωm

Rshale=0.3-3 Ωm

h=0 - 12 ft

Fig. 12—−−−−Crossplot chart for 100-kHz, 84-in. T−R spacing directional measurements to obtain distance to boundary and shoulder-bed

resistivity, assuming a bed resistivity of 15 Ωm

Fig. 13—−−−−Crossplot chart using 400-kHz, 84-in. T−R spaced directional attenuation and 2-MHz PS resistivity in a complex structure with resistivity transition. Distance to oil-water contact and resistivity can be read from this chart.

The basic crossplot interpretation method can be extended

to more complex structures. For example, transition in

resistivity can also be included, as well as information about

the bedding. Fig. 13 illustrates model building for a more

complex structure, including resistivity transition to an oil-

water contact, and a corresponding chart, allowing for the

estimation of bed resistivity and distance to the oil-water

contact.

Another alternative to simple 2D charts are 3D crossplots,

created from two directional measurements and one resistivity

or anisotropy measurement, allowing for simultaneous

interpretation of both bed resistivities and distance to

boundary.

Building the Structure Model using Real-time Inversion

New interpretation software has been developed to translate

measurements into structural maps in real-time. The

interpretation is inversion-based, fully automated, and driven

by a graphical user interface (GUI). This interpretation

facilitates proactive geosteering decisions and provides

significant geological insight into the reservoir structure,

which also benefits future well planning.

A model-based parametric inversion technique has been

developed to estimate the distance to nearby bed boundaries,

horizontal and vertical resistivity of the bed as well as the

resistivities of the beds above and below.27 The fast-forward

model for TI anisotropic horizontally layered medium is run in

the inversion loop, allowing various EM measurement

responses to be integrated and combined in the inversion

process.

In the automatic interpretation mode, all available real-

time measurements are inverted for a simple three-layer

model. An algorithm runs multiple hierarchical models, from

simple to the most complex with up to six parameters,

including distance to two shoulder beds, anisotropic bed

resistivity, and two shoulder bed resistivities. To avoid

“overinterpretation,” Akaike information criterion (AIC)28 and

physics-based constraints are used to additionally penalize the

model complexity and to select the simplest model that fits the

data. The process of generating an initial formation image is

fully automated and requires no user interaction.

To refine the interpretation, data are reprocessed segment-

by-segment. The model may be built point-by-point, or all

data may be inverted for a common layered structure to

determine bed resistivities and thickness of all layers. The

simultaneous inversion reduces the uncertainty due to random

and systematic measurement errors.

Experimental Verification The directional tool responses were extensively tested and

characterized in a 14-ft diameter, 20-ft high resistivity tank.

The tool responses were measured in a hole 14in. from the

tank wall. Measurements were acquired while the tool was

rotating at 50 rpm and the data was processed using the

downhole software.

The experimental results were verified against the 3D

FEM solver with the model, including the tool details as well

as tank geometry and the ground. The FEM discrete model is

shown in Fig. 14. The diagram shows two boreholes used for

borehole eccentering tests.20 The tool was placed in the offset

hole, while the central borehole was filled with a fluid of a

different resistivity, used for other testing.

SPE 97045

7

Fig. 14—−−−−FEM discreteness for a tool in the offset hole of the resistivity tank. The central hole was filled with a fluid of different resistivity.

To verify the the ground resistivity measurements were

made at 0° and 180° azimuth with a static-positioned tool 22-in. above the ground. Voltages were processed to generate

directional measurements that were inverted for ground

resistivity using a crossplot chart. Diagrams in Fig. 15 show

100-kHz and 400-kHz deep directional measurements and

inverted distance and resistivity. A consistent resistivity of 5.9

Ωm was obtained at both frequencies, while the estimated distances are different, indicating that the ground is not

homogenous.

The experimental and 3D modeling results in an offset

hole 14-in. from the boundary in a 1-Ωm tank are summarized in Tables 1-2. Symmetrical directional measurement responses

are shown in Table1. Directional measurements from 1st and

2nd harmonic from transverse-tilted couplings are presented in

Table 2. Given the variation of position, where setup did not

allow fixing the distance to the wall while the tool was

rotating, the agreement is considered very good. It should be

noted that the directional measurements are extremely

sensitive to position. Response increases exponentially with

the distance, and 15% response in variation corresponds to

about 2 in. uncertainty in estimated distance to boundary.

PS=-11.2° Att=-0.405 dB

PS=-30.22° Att=-2.82 dB

100 kHz 400 kHz

Rg=5.9 Ωm h=23”

Rg=5.9 Ωm h=21”

Fig. 15—−−−−Static tool measurements at two azimuths. The crossplots are used to estimate the ground resistivity.

Table 1—Symmetrical Directional Measurements

22in. 34in. 84in. 96in.

Tool -7 -20.92 41.04 30.86 2

MHz FEM -6.39 -17.94 37.3 28

Tool -2.33 -12.75 23.54 -13.04 400

kHz FEM -3.97 -11.99 25.38 -16.06

Tool -1.26 -4.13 18.42 -18.88

PS

(°)

100

kHz FEM -1.21 -3.94 19.97 -21.96

Tool -1.49 -4.88 -3.97 -1.1 2

MHz FEM -1.21 -3.78 -3.49 -1.06

Tool -0.16 -0.78 -6.92 -7.7 400

kHz FEM -0.16 -0.97 -7.84 -8.46

Tool -0.1 -0.08 -0.59 -0.94

Attn

(dB)

100

kHz FEM 0. -0.07 -1.12 -1.52

Table 2—1st and 2

nd Harmonic Directional Measurements

From Transverse Antenna

1st harmonic 2nd harmonic

44in. 74in. 44in. 74in.

Tool 3.32 -6.08 -8.34 -20.93 400

kHz FEM 3.94 -7.35 -9.86 -24.17

Tool 3.58 3.64 -5.54 -14.42

PS

(°) 100

kHz FEM 4.32 4.22 -6.67 -15.63

Tool 1.3 2.17 -1.57 -3.82 400

kHz FEM 1.54 2.39 -1.81 -3.42

Tool 0.05 0.13 -0.39 -1.29

Attn

(dB)

100

kHz FEM -0.06 -0.06 -0.43 -0.85

The azimuthal system was verified by checking the estimated

boundary orientation with respect to magnetic north. The tool

measured a 73° angle, which is within 2° from the true angle. The variation in measured angle from different transmitter-

receiver couplings, measured at different frequencies and

space harmonics, is within 5°. The total number of angles is 38; 5Tx2Rx3F for axial-tilted and 1Tx2Rx2Fx2H for

transverse tilted measurements.

Measurements were taken at different tank resistivities and

similar agreements were obtained. Other tests, such as

resistivity transform check and sensitivity to borehole

eccentering, was also performed.20

Well Placement in Burhaan Field The typical production from a well in the Shuaiba formation

is driven by the quality of the well placement; i.e., avoidance

of Nahr Umr shale exits and minimization of attic-oil left

behind between the wellbore and the overlaying shale.

An experimental prototype tool was used in central Oman

to successfully place seven Petroleum Development Oman

SPE 97045

8

(PDO) wells in Shuaiba thin carbonate oil rims, close to the

top unconformity.19

Wells were positioned, on average, 3−4 ft from the Nahr Umr shale giving access to significant quantities of oil that

would otherwise have been economically unrecoverable “attic

oil” between the well and the shale cap.

Wells have been drilled in a range of resistivity contrasts.

Real-time distances to the boundaries above and below the

well were used to make well placement decisions while

drilling the well. In all cases, the service has delivered

impressively consistent results that can be correlated to, and

used to explain the responses of other measurements,

including wireline imaging logs.

Identified value to the client includes:

- Higher production per unit length due to extended contact

with the high-permeability, weathered interval

(100%−200% typical increase in production) - Reduced attic oil left behind between the well and the cap

shale when production results in the oil-water contact

moving up to the level of the well.

- Longer duration of production due to the delay in the

onset of water production.

- Sustained high production due to the mechanical isolation

of the unstable overlaying shale preventing collapse of the

shale into the wellbore.

Well Placement in Low-Resistivity Contrast

Fig. 16 shows the consistent structure interpretation in the first

800-ft segment from BRN-27 well drilled in the Burhaan field.

Within the first 100-ft, variation from the planned trajectory

was required. The top track, which shows the resistivities

available in real time, does not show any indication of the

structural changes. The lower track shows the interpreted

structure. Inversion-based data processing indicates that bed

thickness is about 7 ft, showing consistent upper and lower

boundary positions. This measurement interpretation was used

to steer the well drilling and stay in the sweet spot. It is

obvious that by simply looking at apparent resistivities, it

would be very hard and practically impossible to achieve that

goal.

A40H P40H P28H

Fig. 16—Geosteering screen shows real-time directional EM measurements interpretation in BRN-27. The top track shows the conventional resistivity measurements, indicating lateral variation of resistivity, insensitive to boundaries.

In the later portion of the same well, besides the dip

change of overlaying shale boundary, lateral variation of

reservoir properties were experienced. The water saturation

increased, reducing the resistivity contrast to as low as 1.5.

The well was drilled about 1 ft below the 0.8 Ωm boundary. Interpretation is shown in Fig. 17. There was no separation in

resistivity responses caused by the nearby boundary, while

directional EM responses enabled consistent boundary

interpretation even in these extreme conditions.

A40H P40H P28H

Fig. 17—Geosteering screen showing real-time directional EM measurements interpretation in low-resistivity contrast scenario

Well Placement in a Faulted Reservoir

Fig. 18 shows a well placement in the Burhaan field where a

fault was encountered. The directional EM measurement

interpretation enabled the throw of the fault to be determined

without tripping out of the hole and the well was successfully

undercut to reenter the reservoir after traversing the fault. All

faults identified with the new service were confirmed by a

borehole imaging run after the well reached total depth.

fault

Fig. 18—A well traversing a major fault and two minor faults.

Detecting Sub-Seismic Faults

In the same reservoir, the formation exhibits lateral variation.

Using conventional technology, it is not possible to determine

whether the change in resistivity reading is caused by lateral

SPE 97045

9

change in water saturation, presence of layering, near-by

boundary, or subseismic fault.

Fig. 19 shows a segment of data in the Burhaan field

where a subseismic fault was identified. The upper track

shows apparent resistivity readings. At the marked position of

the subseismic fault, only phase resistivities experienced

change of about 30%, while attenuation reading was

unaffected. At the same time, the directional response

increased dramatically, indicating the upper boundary

suddenly became much closer. About 200 ft later, when the

directional signal increased rapidly due to the major fault

proximity, the bed boundary and fault effect combined. Fig.

20 shows the available density and resistivity images in the

marked segment, along with the possible locations of the

subseismic fault. Because there is no significant change in

resistivity, these images are not reliable indicators of the

presence of subseismic faults.

sub-seismic fault

major fault

Fig. 19—Resistivity and directional data from a well drilled in the Burhaan field. The position of subseismic fault and major faults are indicated.

Fig. 20—Images in the zone of the subseismic fault. Due to very small resistivity contrasts, boundary indications are not reliable.

Measurement Consistency

To validate the measurement consistency further, each

directional measurement was analyzed individually on a

segment of trajectory using crossplot charts. Estimated

distance to the Nahr Umr shale from different spacings and

frequencies, assuming the same shale resistivity Rh=0.8 Ωm,

Rv=3.2 Ωm, is shown in Fig. 21. All crossplots use conventional PS 28-in. 2 MHz apparent resistivity. The

estimated distance to the boundary from six different

directional measurements is practically the same. The only

exception is that the deepest measurement SAD1 (100-kHz,

96-in. attenuation) shows a longer distance. That particular

measurement is more sensitive to the lower boundary and

conductive layer below, which is reducing the directional

signal. Therefore, if the single-boundary model is used to

interpret the data, it corresponds to the boundary farther away.

One should also note the scatter corresponding to SPD4 (400-

kHz, 96-in. phase shift) close to the boundary. At that signal

level, the distance solution is not unique due to low shoulder-

bed resistivity.

Fig. 21—Consistency of directional measurements

Well Placement in the Malaan field

Discovering Better Reservoir

While drilling horizontally in the reservoir layer of PDO’s

Malaan-2 well, the unique capability of the deep directional

EM tool to measure the resistivity of shoulder beds several

meters away, led to the identification of a previously

undiscovered high-quality reservoir sweet spot above the well.

The geosteering team immediately recognized the value of

these data and presented the client with a new plan to drill

shallower to explore the potentially better pay zone.

Subsequent analysis confirmed that a large portion of the well

had indeed been placed in a higher-quality reservoir, turning

an exploration well into a producer. Fig. 22 shows the actual

real-time structural interpretation based on the directional EM

data during the well placement. The well was moved up 15

feet from the drill plan, and from a 5 Ωm zone (dark brown) to a zone with four times the expected resistivity (light brown).

SPE 97045

10

Fig. 22—Real-time structure interpretation for the well in the Malaan field. The lighter color identifies the unsuspected reservoir sweet spot with higher resistivity values.

Confirmation of the Boundary Position

In order to gain confidence in the accuracy of the real-time

interpretation and new measurements, a number of wells were

intentionally steered upward at the toe to confirm the position

of the boundary. An example is shown in Fig. 23. The upper

and lower boundaries indicate that the reservoir was about to

pinch out. Not only was the position of the upper overlaying

shale boundary consistently interpreted, but also, the

resistivity of the shoulder bed was consistent on both sides of

the boundary.

Fig. 23—Geo-steering software screen capture confirms interpreted position of the boundary from real-time interpretation.

Well Placement in the North Sea Turbidite Formations Several wells were drilled in a turbidite sand reservoir in the

North Sea. In the past, these kinds of wells were drilled

geometrically based on seismic maps and known position of

oil-water and gas-oil contacts. However, very often, the total

lengths in sand was reduced due to interbedded shale and silt.

Fig. 24 shows the planned trajectory and assumptions

about the reservoir structure based on the seismic map. The

operator planned to drill horizontally about 4,000 ft. Real-time

estimated distances to boundaries above and below the well,

as well as boundary orientations, were used to make well

placement decisions while drilling the well. The real-time

interpretation is shown in the same figure, with aligned start

and end of the planned and drilled well and reservoir model

built by inverting directional EM measurements. Clearly, the

assumptions about the structure are different from the reality.

Fig. 24—Real-time structure map generated from the directional EM tool measurements during well placement in a reservoir with intrabedding shales, compared to original assumptions made about the structure.

Fig. 25 shows the overall interpretation from the

placement of a nearby well in the same field, with planned

trajectory, actual trajectory, and structure built in real-time.

The original plan to drill horizontally was changed, and the

trajectory was adjusted about 30 ft in depth, resulting in

significantly increased footage in the sand.

Fig. 25—Real-time structure map generated from the directional EM tool during well placement in a reservoir with intrabedding shales. The top and bottom lines indicate the tool sensitivity cutoff.

Fig. 26 is a segment of the same well, interpretation

showing the distance to boundary, and the azimuth view,

which shows the boundary orientation. The measured azimuth

angle of 21° agrees very well with the density images that were available in real time. The new directional EM

technology, with its ability to build the un-biased structure

interpretation in real time, allowed the operator to be proactive

and make geo-steering decisions based on distance to

boundaries as well as boundary orientation.

SPE 97045

11

Fig. 26—Using the azimuth view showing the boundary orientation and distance to boundary interpretation to adjust the well path.

Well Placement in a Mature North Sea Field The tool performance was tested in a mature field in the North

Sea. During the well landing, the new directional EM tool was

able to consistently detect the boundary as far as 17 ft away

from the wellbore, shown in Fig. 27.

Directional measurements are shown in the first track and

resistivity in the second track. Clearly, the resistivity

measurement has a smooth change from shale to sand

resistivity, without the typical horn corresponding to a sharp

boundary. At the same time, the directional measurements

clearly indicate approaching the boundary, and allowed

consistent distance-to-boundary estimation up to 17 ft away

from the wellbore. More details about the well placement in

this field will be presented in a companion paper.21

40 Ωm

27 Ωm

5.5 Ωm

DPD1 DPD4 DPS4

P16H

A40H P40H P28H

Fig. 27—Consistent distance-to-boundary interpretation in a North Sea mature field

Conclusions The first deep-directional EM LWD tool and real-time

interpretation answer products for well placement have been

developed and successfully field tested in different parts of the

world.

The new service is based on four innovative technologies

implemented for the first time in LWD EM tools: tilted and

transverse coils with new shields; symmetrical antenna

arrangement with reduced dip and anisotropy effect for

geosteering or amplified anisotropy effect for formation

evaluation; novel measurement acquisition scheme with

azimuthal fitting of responses; and real-time inversion-based

interpretation to build an unbiased model and enable

proactive geo-steering decision making, and reservoir model

updates.

Field tests have demonstrated how the information from

the new tool can be used to optimize well placement through

proactive geosteering; e.g., by placing wells in a thin oil rim

without exiting into a cap shale, or by increasing the

productive length of a wellbore in a reservoir consisting of

complex structures. The field data examples demonstrated that

the new measurements are able to detect and accurately

estimate distances and orientation of nearby boundaries up to

17 ft away from the wellbore in favorable conditions. The

measurements are able to estimate the resistivity of shoulder

beds, allowing operators to detect and steer into neighboring

higher-quality parts of a reservoir.

Novel data processing and structure visualization software

was developed to facilitate timely decision-making and

planning processes. The technology developed includes a real-

time visual display of the interpreted structure, including the

boundary locations and their 3D orientation, as well as the

formation and shoulder resistivities, requiring no prior

knowledge of the geological structure. Real-time inversion

results demonstrate consistent structure interpretation over

long sections. The field data reveal that lateral heterogeneities

of the reservoirs are common, and relying on 3D seismic is not

sufficient for efficient well placement.

Acknowledgments The authors would like to thank the oil companies for

permission to use their data in this paper.

The authors would like to acknowledge their Schlumberger

colleagues: Steve Bonner, Brian Clark, Dean Homan, Alain

Dumont, Cengiz Esmersoy, Mark Frey, Gerry Minerbo,

Richard Rosthal, Jean Seydoux, and Jacques Tabanou, from

the Sugar Land Technology Center for their help, advice, and

useful discussions in the early stage of project development.

The authors are also indebted to Tarek Habashy from

Schlumberger-Doll Research for his contribution on

development of original distance to boundary inversion. We

would also like to thank the Schlumberger field organization

and geosteering community for collaboration during the field-

testing.

Nomenclature

ft x 3.048 E - 01 = m

in. x 2.54 E + 00 = cm

Att Attenuation measurements

A40H 2MHz 40-in. attenuation resistivity curve

DTB Distance to boundary

MD Measured depth

SPE 97045

12

P28H 2MHz 28-in. phase-shift resistivity curve

P40H 2MHz 40-in. phase-shift resistivity curve

P28L 400kHz 28-in. phase-shift resistivity curve

P40L 400kHz 40-in. phase-shift resistivity curve

PS Phase-shift measurements

Rd Resistivity of the lower shoulder

Rh Formation horizontal resistivity

Rm Borehole mud resistivity

Rt True formation resistivity

Ru Resistivity of the upper shoulder

Rv Formation vertical resistivity

SAD1 Symmetrical 100kHz 96-in. directional

attenuation measurement

SAD4 Symmetrical 400kHz 96-in. directional

attenuation measurement

SAS1 Symmetrical 100kHz 34-in. directional

attenuation measurement

SAS4 Symmetrical 400kHz 34-in. directional

attenuation measurement

SPD1 Symmetrical 100kHz 96-in.

directional phase-shift measurements

SPD4 Symmetrical 400kHz 96-in. directional

phase-shift measurements

SPS1 Symmetrical 100kHz 34-in. directional

phase-shift measurements

SPS4 Symmetrical 400kHz 34-in. directional

phase-shift measurements

TVD True vertical depth

References 1. Kreigshauser, B. et al.: “A New Multicomponent Induction Tool

to Resolve Anisotropic Formation,” paper D presented at the

2000 41st Annual SPWLA Symposium, Salt Lake City, Utah, 30

May−3 June. 2. Rosthal, R. et. al.: “Field Test Results of an Experimental Fully-

Triaxial Induction Tool,” paper QQ presented at the 2003 44th

Annual SPWLA Symposium, Houston, TX, 22−25 June. 3. Mechetin, V.F. et. al.: “TEMP – A new dual electromagnetic

and laterolog apparatus- technological complex,” paper K

presented at the 1990 13th European Formation Evaluation

Symposium, Budapest, Hungary; Trans., 1-16.

4. Sato M. et. al.: “Apparatus and Method for Determining

Parameters of Formations Surrounding a Borehole in Pre-

selected Direction,” U.S. Pat. No. 5508616 (1996).

5. Spies, B.R. and Habashy, T.M.: “Sensitivity Analysis of Cross-

Well Electromagnetics,” Geophysics (1995), 60, No. 3, 834.

6. Nabighian M. N.: “Electromagnetic Methods in Applied

Geophysics,” . 1-2, Investigations in Geophysics (1987), 3, 285.

7. Rodney, P.F. and Wisler, M.: “Electromagnetic Wave

Resistivity MWD Tool,” SPEDE (1986), 337. 8. Clark B. et. al.: “A Dual Depth Resistivity Measurement for

FEWD,” paper A presented at the 1988 29th Annual SPWLA

Symposium, San Antonio, TX, 5-8 June.

9. Bittar, M. et. al.: “An MWD Multiple Depth of Investigation

Electromagnetic Wave Sensor,” paper D presented at the 1991

32nd Annual SPWLA Symposium, Midland, TX, 16−19 June. 10. Meyer, W.H.: “Interpretation of propagation resistivity logs in

high angle wells,” paper D presented at the 1998 39th SPWLA

Annual Symposium, Keystone, CO, 26−29 May. 11. Seydoux, J. et. al.: “A Deep-Resistivity Logging-While-Drilling

Device for Proactive Geosteering,” The Leading Edge (2004),

23, No.6, 581.

12. Edwards, J. et. al.: “Geosteering Examples Using Modeled 2-

MHz LWD Response in the Presence of Anisotropy,” paper N

presented at the 2000 41th Annual SPWLA Symposium, Dallas,

TX, 4−7 June. 13. Rohler, H. et. al.:“The Use of Real-Time and Time-Lapse LWD

Images for Geosteering and Formation Evaluation in the

Breitbrunn Field, Bavaria, Germany,” paper SPE 71331

presented at the 2001 SPE Annual Technical Conference and

Exhibition, New Orleans, LA, 30 Sept – 3 Oct.

14. Rasmus, J., Esmersoy, C., Seydoux, J., Hawthorn, A.: "LWD for

Imaging, Wellbore Placement and Formation Evaluation," paper

17646 presented at the 2005 Offshore Technology Conference,

Houston, TX, 2−5 May. 15. Li, Q. et. al.: “Real-Time LWD Image: Techniques and

Applications,” paper WW presented at the 2001 42nd SPWLA

Annual Symposium, Houston, TX, 17-20 June.

16. Luling, M.: “Method for controlling directional drilling in

response to horns detected by electromagnetic energy

propagation resistivity measurements,” U.S. Pat. No. 05241273

(1993).

17. Li, Q. et. al.:“Automated Interpretation for LWD Propagation

Resistivity Tools Through Integrated Model Selection,”

Petrophysics (2004), 54, No. 1, 1.

18. Yang, J. et. al.: “Bed-Boundary Effect Removal to Aid

Formation Resistivity Interpretation from LWD Propagation

Measurements at All Dip Angles,” paper presented at the 2005

46th SPWLA Annual Symposium, New Orleans, LA, 26-29

June.

19. Daveridge, S. et al.: “An Innovative Business Model to

Leverage Innovative Well-Placement Technology,” paper 17591

presented at the 2005 Offshore Technology Conference,

Houston, TX. 2−5 May. 20. Li, Q. et. al.: “New directional electromagnetic tool for

proactive geosteering and accurate formation evaluation while

drilling,” paper presented at the 2005 46th SPWLA Annual

Symposium, New Orleans, LA, 26−29 June. 21. Laastad, H. et. al.: “Geosteering Using New Directional

Electromagnetic Measurements and 3D RSS in a North Sea

Well,”paper 95725 presented at the 2005 SPE Annual Technical

Conference and Exhibition, Dallas, TX, 9−12 October. 22. Omeragic, D. et. al.: “Characterization of LWD antennas shield

effects,” paper presented at the 2002 Joint IEEE AP/URSI

Symposium, San Antonio, TX, 16−19 June. 23. Doll, H. G., “Introduction to induction logging and application

to logging of wells drilled with oil based mud,” JPT (1949), 1,

No.6, 148.

24. Hagiwara, T. et. al: “Effects of Mandrel, Borehole and Invasion

for Tilt-Coil Antennas,”, paper 84245 presented at the 2004 SPE

Annual Technical Conference and Exhibition, Houston, TX,

26−29 September. 25. Minerbo, G. and Omeragic, D.: “A directional electromagnetic

measurement for bed boundary detection in geosteering,”

Schlumberger Internal Technical Report, 2001.

26. Omeragic, D. et. al.: “A directional propagation style

electromagnetic measurement for bed boundary detection

insensitive to anisotropy at any dip,” Schlumberger Internal

Technical Report, 2001.

27. Omeragic, D. et. al.: “Method for calculating a distance between

a well logging instrument and a formation boundary by

inversion processing measurements from the logging

instrument”, US Pat. No. 6594584 (2000).

28. Akaike, H.: “A new look at the statistical model identification”,

IEEE Trans. Aut. Control, (1973), 19, 716.

SPE 97045