Deep directional resistivity measurement for reservoir navigation … · 2020. 4. 14. · resistivity boundary of typical reservoir rocks. For a given frequency and coil spacing,

Deep directional resistivity measurement for reservoir navigation and formation evaluation while drilling: Case Study – North Sea

Tsili Wang, Roland Chemali, Jack Signorelli, Tron Helgesen

Baker Hughes INTEQ Chris Bell, John Hampson, and Andy Poppitt

Chevron Upstream Europe

Abstract Tested in some of the more complex reservoirs of the North Sea, an advanced azimuthally sensitive propagation resistivity tool provided information for clear geosteering advice in cases where standard LWD resistivity tools lead to ambiguous interpretations. Field experiments have demonstrated that combined with conventional LWD resistivity measurement, azimuthal resistivity data have successfully predicted approaching directions of both sand and water zones at a much larger depth of investigation than borehole imaging tools. The results were confirmed by both LWD azimuthal gamma ray (GR) and density image logs. Through azimuthally sectoring of the density and GR measurement, the remote bed boundaries can be described as images. GR images are obtained from the main MWD/LWD sub; the density image is acquired by the density tool. Introduction After about two decades of development, LWD resistivity measurements now are one of the most commonly used tools for geosteering, reservoir navigation, and formation evaluation (Clark et al., 1988; Bittar et al., 1991; Meyer et al., 1994). Such measurements provide information about formation resistivity near the bit and approaching beds. Comparison of real-time measurements with pre-calculated tool responses allows the reservoir navigation engineer to determine the bit position relative to the geological target. Although the conventional LWD resistivity measurement allows discrimination of a conductive approaching bed from a resistive one, the measurement does not distinguish the direction of approach. For instance, a decreasing apparent resistivity value indicates an approaching conductive formation. The conductive formation can occur above (e.g., a shale roof) or below (e.g., an oil-water contact)

the tool. To avoid penetrating the conductive formation, the direction of the conductive formation must be known. Reservoir navigation in a deepwater sand channel may require similar directional information to remain in the channel. LWD measurements with azimuth sensitivity include azimuthal gamma-ray, neutron, density, and resistivity imaging tools (Bonner et al., 1994; Greiss et al., 2003; Ritter et al., 2004). These measurements are extremely useful for reservoir navigation and formation evaluation. However, they all have relatively shallow (less than a foot or so) depths of investigation. Some of them are no more than a couple of inches deep. Only until very recently has a deep directional measurement technique been reported (Li et al., 2005). In this paper, we describe a prototype LWD directional resistivity tool that provides deep-reading azimuthal resolution around a wellbore. When combined with the standard LWD resistivity measurement, the new azimuthal measurement allows for resolution of an approaching bed occurring at any azimuthal angle from the tool. The large depth of investigation of the new tool will significantly improve geosteering and reservoir navigation capabilities. The addition of this measurement will also increase our ability to quantify remote bed resistivity properties. Azimuthal Propagation Resistivity Measurement The building block for the new azimuthal resistivity measurement is a crossed coil pair made up of an axial transmitter and a transverse receiver (Fig. 1). The signal from this elementary array is highly sensitive to the direction of an approaching boudary that separates two zones of contrasting resistivities. Consider the case of a conductive formation located above (Fig. 1) the crossed coil pair. As the two coils are well aligned, with their centers positioned on the tool axis, there is no direct coupling between them. The only contribution to the measured response comes

from the remote bed. The sensitivity of this crossed-coil arrangement to the direction of the remote bed is best illustrated with the “mirror image” principle. According to this principle, the contribution of a remote bed may be approximated, qualitatively, with a mirror image of the transmitter with respect to the bed boundary. If the remote bed is conductive, the “mirror image” transmitter points in the same direction as the real transmitter. Consequently, if one adopts the intuitive sign convention, the cross receiver signal points in the direction of the approaching conductive bed. For instance, in case of a shale bed overlaying the reservoir (Fig. 1) the cross-receiver points up.

Fig. 1. Operating principle of the azimuthal propagation resistivity measurement: Case of conductive shale overlaying a resistive sand reservoir. The signal points to the shale.

Fig. 2. Operating principle of the azimuthal propagation resistivity measurement: Case of high resistivity hydrocarbon interval overlaying a low resistivity water zone. The signal points to the water. Next consider the case of a conductive water bearing interval, located below the hydrocarbon zone (Fig. 2). The cross-receiver signal points down. For other conductive formations approaching the reservoir at different azimuths from the well path, the signal will likewise point to the direction of the actual boundary. The “image principle” cited above is approximate. A more exact theoretical modeling, not presented here confirms that the crossed-coil, or fully-tilted-coil

arrangement provides the necessary information to resolve the direction of a remote bed relative to the tool. Traditional coaxial coil configurations do not provide such information because they inherently lack directional sensitivity. In the crossed-coil arrangement discussed above, the transmitter and receiver are fully orthogonal to each other with one of them oriented axially. Transmitter and receiver may also be partially tilted, at intermediate angles, between 0 degree (axial) and 90 degrees (fully tilted) (Hagiwara et al., 2003; Li et al., 2005). Such configuration mixes direction-sensitive and direction-insensitive information. The data must be processed to extract the azimuthally sensitive information from the gross measurement. Signal Strength Theoretical models show that the azimuthal resistivity signal reaches a maximum when the tool straddles a resistivity boundary of typical reservoir rocks. For a given frequency and coil spacing, and a given distance to the boundary, the signal strength varies as a periodic function of the tool face (Fig. 3). The magnitude of the signal depends primarily on the conductivity difference across the boundary. The chart in Fig. 4 suggests that the maximum signal is is nearly a linear function of the conductivity difference, whereas conventional resistivity responses depend on the resistivity contrast across a boundary.

1 Ohmm100 Ohmm

1 Ohmm

100 Ohmm

Azi

mut

halR

esis

tivity

Sig

nal

Tool Face Angle (degree) Fig. 3. The azimuthal resistivity response as a function of the tool face angle. When a resistivity boundary is right above the tool, the maximum response poinst at the 0 degree azimuth. If the boundary is deviated, the maximum response points towards the actual bed boundary direction.

Upper bed resistivity

Azi

mut

halR

esis

tivity

Sig

nal (

nV)

Fig. 4. Dependence of the azimuthal resistivity response on the conductivity difference across a resistivity boundary. The tool is parallel to the boundary. The response depends almost linearly on the conductivity difference across the boundary. Data Acquisition The new resistivity data is acquired at a series of tool face angles, as the drilling string rotates. The current version of the tool acquires data in 16 sectors evenly spaced over 360 degrees. The 16 sectors result in a minimum azimuthal pixel size of 22.5 degrees. The magnitude of the signal is at a maximum when the transverse antennas point normal to a resistivity boundary. The signal decreases to zero when the transverse antennas point in a direction parallel to a resistivity boundary. Opposing sectors exhibit the same signal strength but with different signs. The amplitude attains the maximum value when the receiver coil faces toward (0 degree) and against (180 degrees) the nearest point of the tank boundary. The amplitudes at other angles vary according to cos(θ) where θ is the tool face angle measured from the nearest point of the tank boundary. The phase measurements have only two distinct values which differ from each other by 180 degrees. Interpretation Concepts Several important concepts have been developed to aid in data interpretation. The most important ones are electrical midpoints and electrical saddle points. An electrical midpoint is where the strength of the azimuthal signal crosses zero as its direction flips by 180 degrees. A saddle point is where the magnitude of the signal dips to a minimum value, surrounded by two local maxima. The tool response at a saddle point may or may not be zero. Below, we give a few scenarios that can produce either electrical midpoints or saddle points.

An electrical midpoint occurs when the tool is surrounded by two shoulder beds of the same or similar resistivity values. Signals from opposing beds cancel each other. Fig. 5 illustrates an electrical midpoint with the tool located within a 10-ohmm formation, sandwiched by two 1-ohmm beds. In this case, the electrical midpoint coincides with the physical midpoint of the 10-ohmm bed. When the shoulder beds have different resistivities the electrical midpoint moves away from the physical midpoint of the bed, toward the more resistive bed. This is clearly illustrated in Fig. 6 where the lower bed resistivity is increased to 2 ohmm. The electrical midpoint moves from 115 ft measured depth to about 117 ft measured depth, closer to the shoulder bed with smaller contrast. A few scenarios can cause an electrical saddle point. As stated above, a saddle point occurs when the tool signal magnitude exhibits a local minimum. In order to have a saddle point, the tool response on both sides of the saddle point must have the same sign; otherwise, an electrical midpoint will result. One common case that causes a saddle point is shown in Fig. 7 where the resistivity boundary is flat but the well trajectory is “V”-shaped. The new tool achieves the minimum response magnitude at the lowest well trajectory point.

1 ohmm

10 ohmm

1 ohmm

Well path

Azm

. Res

istiv

ityS

igna

l (nV

)

Fig. 5. An electrical midpoint in the azimuthal resistivity response occurs when the well traverses two parallel shoulder beds having the same resistivity. In this case, the electrical midpoint coincides with the physical midpoint of the bed. A curved boundary or a “porpoising” well may also cause a saddle point. In Fig. 8(a), the well enters a roof from one side and exits through the roof from the other side. The tool exhibits the maximum responses when it crosses the boundaries and a minimum

response between the crossovers. Because the responses at the crossovers have the same sign, the minimum response in between appears as a saddle point. Several other scenarios for a saddle point are demonstrated in Fig. 8(b)-(d). A special case including unparallel boundaries is a fault zone. A fault with the same or similar formations across the fault zone can produce a saddle point. As shown in Fig. 9, when the tool is close to the tip point of the hanging wall, a saddle point may occur. In this case, a saddle point indicates the transition from being close to the underlying formation to close to the fault zone or foot wall.

1 ohmm

10 ohmm

2 ohmm

Well path

Azm

. Res

istiv

ityS

igna

l (nV

)

Fig. 6 Same as Figure 5 except that the lower shoulder bed resistivity is increased to 2 ohmm. The electrical midpoint is moved closer to the lower shoulder bed boundary, the one with less conductivity difference.

1 ohmm

10 ohmmWell path

Azm

. Res

istiv

ityS

igna

l (nV

)

Fig. 7 An electrical saddle point in the azimuthal resistivity response is caused by a “V”-shaped well path.

Finally, a cascaded resistivity profile can also generate a saddle point. In a cascaded resistivity profile the formation resistivity increases (or decreases) monotonically with depth. One example is shown in Fig. 10 where the bed resistivities increase from 0.7 ohmm to 2 ohmm at the top and to 50 ohmm at the bottom. The relative well deviation is 70 degrees. The standard propagation resistivity shows the typical high-angle horn effect near the resistivity boundaries. In this model, the two peaks of the tool response have similar magnitudes because of the similar conductivity differences across the two boundaries.

Shale

Wellbore

Shale

Wellbore

Shale

(a) (b)

(c) (d)

Wellbore

Shale

Wellbore

Fig.8 Several scenarios may generate electrical saddle point.

Shale

Wellbore

Saddle point

Fig. 9. A fault zone may cause a saddle point.

0.7 ohmm

2 ohmm

50 ohmm

Well path

Azm

. Res

istiv

ityS

igna

l (nV

)

Fig. 10. An electrical saddle point in the azimuthal resistivity response is caused by a cascaded resistivity profile.

Generating an Azimuthal Resistivity Image Because the azimuthal resistivity response in a thick, uniform formation is zero, regardless of the formation resistivity level, it is convenient to transform the response such that it bears information not only about the azimuthal direction of remote beds but also about the resistivity level. To do so, we create a pseudo resistivity image by combining the azimuthal resistivity measurement with the standard propagation resistivity measurement. Figs. 11 and 12 show two examples of the pseudo resistivity image for a conductive bed from above the tool and a conductive approaching from below the tool. The pseudo resistivity image is interpreted in the same way for bed azimuths as for a borehole imaging log except that the former can not be interpreted directly for the bed dip angle through the traditional fitting of a sinusoidal pattern. Indeed the electrical diameter of the azimuthal propagation resistivity is much larger than that of electrical wellbore imaging tools. Examples of resistivity images from actual well logs are discussed in the following field examples. Reservoir Navigation with Azimuthal Propagation Resistivity The objectives of the new azimuthal resistivity measurement for reservoir navigation include (1) staying below a shale roof, (2) staying above an oil-water contact, (3) early identification of the azimuth and distance of approaching resistivity boundaries, and (4) avoidance of shale lenses and calcite stringers. Staying below shale roofs. Production wells often need to be placed as high in a reservoir as possible for best recovery, but without entering the overlying shale. This is an important challenge because in many instances overlying shales are neither flat nor horizontal. The azimuthal signal alerts the reservoir navigation engineer of an approaching shale roof and shows its direction relative to the well path. The magnitude of the signal, together with the resistivity level, provides an indication of proximity to the roof. By comparing the measured signal magnitude to that of the pre-modeled signal, an estimate of the distance to the roof can be assessed. A more accurate algorithm of the distance-to-boundary was recently developed.

Well path10 ohmm

1 ohmmWell path

Fig. 11. Standard propagation resistivity measurement and pseudo resistivity image for a well entering from a 10-ohmm formation into a 1-ohmm formation from below. The letters “T”, “R”, “B”, and “L” stand for top, right, bottom, and left, respectively.

Well path

10 ohmm

1 ohmm

Well path

Fig. 12. Same scenario as Figure 11 except the well enters the 1-ohmm formation from above. Early identification of approaching boundaries: This is a general application of the azimuthal resistivity propagation measurement, particularly for geosteering in non-layered cake reservoirs. The new instrument is specifically designed to provide the direction of an approaching boundary as well as the distance to that boundary. An approaching boundary can be an oil-water interface, a shale stringer, the top of a reservoir or any formation with a resistivity contrast to the bed being drilled. Recall that traditional geosteering with propagation resistivity lacks azimuthal sensitivity. A conventional propagation resistivity tool helps detect an approaching bed and possibly estimate its distance, but such a tool does not distinguish “above” from “below” or “left” from “right”. A conductive roof formation produces the same or similar signal as a

shale stringer approaching from the side, or an oil-water contact approaching from below. The evasive action for one case is contrary to those of the other cases. Expert reservoir navigation engineers are trained to recognize such situations from non-azimuthal propagation, but they must have prior detailed knowledge of the subsurface; and they generally have difficulty handling non-layered cake situations such as in the examples below. Avoidance of shale lenses or calcite stringers. One of the challenges in drilling shaly sand reservoirs is to minimize non-productive intervals like shale lenses or calcite stringers. Calcite stringers present additional risks of damaging drill bits and the bottom hole assembly (BHA). Traditional non-azimuthal propagation resistivity tools will likely warn of an approaching high-resistivity bed with an elevated resistivity reading but will not tell anything about the direction of approach. The azimuthal resistivity measurement will be able to distinguish the approaching direction of a bed as it comes within the depth of investigation of the tool. The magnitude of the response and warning time increases with the relative approaching angles. If the well approaches a shale lens or calcite stringer along a sub-parallel well path, the new measurement is expected to provide enough advance warning for evasive action. Case Histories Well “M”, Chevron, North Sea The field examples described below were collected for a reservoir navigation feasibility case study of the azimuthal resistivity tool. In the first case history reported below, no data was available for real-time use. This was an early field test to assess the tool feasibility. Actual geosteering was conducted using the traditional multi-propagation resistivity together with azimuthal gamma-ray imaging. The former exhibits no azimuthal sensitivity; the latter has strong azimuthal sensitivity but with a depth of investigation limited to only a few inches. The experimental azimuthal resistivity log was recorded in memory with the expectation that it would provide both azimuthal sensitivity and depth of investigation. From the memory data, we were able to gain better understanding of the subsurface, as well as confidence in the validity of the measurement. We also constructed hypothetical real-time logs to get a better sense of the benefits of the real-time information. The objectives of the azimuthal resistivity measurement for this well were (1) staying

immediately below the shale roof, (2) early warning of approaching resistivity boundaries, and (3) avoidance of shale lenses. The reservoir cross section and the pre-job modeling are shown in Fig. 13. The reservoir layer dips in general toward the right. A pilot well and a side track well were drilled. Based on the information acquired from the pilot well, the side track well was planned to avoid the shale roof. The actual drilled well entered the shale roof in the highlighted interval. This entrance may have been avoided with a real-time azimuthal resistivity measurement. The downhole hardware operated properly throughout. All 16 sectors were recorded to memory at short time intervals. When the tool was pulled out to the surface, the raw acquired data was re-processed. The post-job modeling results are shown in Fig. 14. Both the gamma ray image and the standard propagation resistivity show that the well entered the shale roof at around 10,700ft (measured depth) and exited at around 11,100ft. The well re-entered the shale roof at 11,500ft. The azimuthal resistivity signal strength (Track 4) correlates well with approaching bed boundaries. In the reservoir section, the azimuthal resistivity signal is low to very low, as expected.

A short section zooming in on the first entry to the shale roof is shown in Fig. 15. Signals from all 16 sectors are shown in the uppermost track. The quality of the measurement can be readily appreciated by observing the regularity of the distribution of the 16 curves. The signal strength decreases from the peak-signal sector gradually to the minimum-signal sector that is 8 sectors apart from the peak-signal sector. A proprietary algorithm converts the 16-sector curves and the standard propagation resistivity measurements to a resistivity image seen in the second track. This image is similar to those from wireline or LWD borehole imaging instruments, but with a much larger depth of investigation. A darker color indicates a more conductive formation and a lighter color reflects a more resistive formation. Instead of focusing on the borehole wall and its near vicinity, the new azimuthal resistivity tool looks deep into the formation to detect approaching resistivity boundaries well before they intersect the well path, especially when the angle of approach is high.

The azimuthal resistivity image starts sensing the approaching shale roof at about 10,410ft. The azimuthal Gamma Ray image detects the well actual entry into the shale roof at about 10,690ft, i.e. nearly 300 ft later. Prior to that point the GR images do not show any evidence of the approaching shale. Even the standard propagation resistivity does not exhibit an

appreciable change in resistivity measurement in this instance, prior to getting close to the shale. Had the real-time data been available, the azimuthal resistivity measurement would have provided about 280 ft of warning distance.

Two factors contribute to the early detection of the approaching shale before it intersects the well path. First, the azimuthal resistivity has a large depth of investigation. Second, the well path has a high apparent dip angle relative to the approaching shale. The Gamma Ray image indicates an apparent dip angle of approximately 88 degrees. The apparent dip was also calculated independently on a larger scale from the distance to the shale derived from the azimuthal resistivity and the standard propagation resistivity. The triangulation calculation yields an apparent dip of 87.5 degrees, in agreement with the gamma ray.

Observe that for the particular set of bed resistivities seen here, the new tool detects the approaching boundary from 15 ft away when in the reservoir looking for shale. In higher resistivity reservoirs the tool is expected to exhibit a depth of detection as high as 17 ft.

Well “P”, Chevron, North Sea Real-time data from the azimuthal resistivity tool in Well P was successfully transmitted to the surface and applied to geosteering. Most of the high-angle portion of the well stayed in the reservoir. The resistivity image together with the gamma ray image, the standard propagation resistivity, and the azimuthal resistivity signal strength are shown in Fig. 16. The azimuthal resistivity signal in the reservoir section is relatively low, but well above the instrument noise floor. A distance-to-bed calculation indicates that the distance to the shale roof is over 17 ft. However, the gamma ray image suggests a clean lithology and the absence of azimuthal formation variation. One particular event shown in Fig. 16 corresponds to a drop in resistivity, but with no signature on the gamma ray image. The azimuthal resistivity data suggests a conductive bed, probably shale, approaching from above slightly to the left side. The bed is far enough from the wellbore not to show any detectable gamma radiation above background.

The TD portion of well “P” (Fig. 17) encounters several thin streaks and finally the oil-water contact. The well direction near TD is pointing downward. Starting at 9720ft the propagation resistivity shows an overall lower resistivity caused by the thin streaks indicated by the gamma ray. The azimuthal resistivity image suggests the conductive streaks dip upward, as confirmed by the gamma ray image. The image shows the well entering the water zone from above, while the

gamma ray image shows no response to the oil-water contact.

Well “Q”, Chevron, North Sea Real-time data from the azimuthal resistivity tool in well “Q” was also applied to geosteering. The detailed image was generated after drilling from the memory data (Fig. 18). An approaching conductive bed, below the well path, is clearly identified starting at 15,605 ft. This conductive interval is interpreted to be an oil-water contact. The well path is directed upward to stay clear from the conductive zone, and remains for an additional 200 ft in the reservoir.

Conclusions Both laboratory and field data have demonstrated the high quality of the new experimental LWD resistivity measurement. This azimuthal resistivity measurement, when combined with the conventional LWD resistivity data, produces a pseudo image that allows for direct determination of the bed direction. Field examples have proven that the deep azimuthal resistivity measurement provides a geologically meaningful interpretation with valuable information for proactive geosteering. Acknowledgments

We would like to thank those who have contributed to the project including S. Fang, Hu Po, G. Itskovich, A. Kirkwood, H. Meyer, I. Munn, L. Tabarovsky, B. Tchakarov, and L. Yu. Special thanks go to Dan Georgi and Holger Stibbe for their continuous support and enthusiasm. Permission by Chevron to use the field data is greatly appreciated. We also thank Baker Hughes for permission to publish this paper.

References Bittar, M.S., Rodney, P.F., Mack, S.G., and Bartel, R.P., 1991, A true multiple depth of investigation electromagnetic wave resistivity sensor: Theory, experiment and field test results: SPE 22705.

Bonner, S.D., Bagersh, A., Clark, B., Dajee, G., Dennison, M., Fredette, M., Grogan, O., Hall, J.S., Jundt, J., Kwok, E., Lovell, J.R., Rosthal, R.A., and Allen, D., 1994, A new generation of electrode resistivity measurements for formation evaluation while drilling, SPWLA 35th Annual Well Logging Symposium, paper OO.

Clark, B., Luling, M., Jundt, J., Ross, M., and Best,

David, 1988, A dual depth resistivity measurement for FEWD, SPWLA 29th Annual Logging Symposium, paper A.

Greiss, R.-M., Webb, C.J., White, J., McDonald, B., Flanagan, K.P., Rodriguez, J.M., and Scholey, H., 2003, Real-time density and gamma ray images acquired while drilling help to position horizontal wells in a structurally complex North Sea field, SPWLA 44th Annual Logging Symposium, paper Z.

Hagiwara, T., Banning, E., Ostermeier, R.M., and Haugland, M., 2003, Effects of Mandrel, Borehole, and Invasion for Tilt-Coil Antennas: SPE 84245.

Li, Q., D. Omeragic, L. Chou, L. Yang, K. Duong, J. Smits, J. Yang, T. Lau, C. Liu, R. Dworak, V. Dreuillault, and H. Ye, 2005, New directional electromagnetic tool for proactive geosteering and accurate formation evaluation while drilling: SPWLA 46th Annual Well Logging Symposium, paper UU.

Meyer, H.W., Thompson, L.W., Wisler, M.M., and Wu, J.Q., 1994, A new slimhole multiple propagation resistivity tool: SPWLA 35th Annual Logging Symposium, paper NN.

Ritter, R., Chemali, R., Lofts, J., Gorek, M., and Fulda, C., Morris, S., and Krueger, V., 2004, High resolution visualization of near wellbore geology using while-drilling electrical images: SPWLA 45th Annual Logging Symposium, paper PP.

Pre-job Modeling

Track 1 Track 2 Track 3 Track 4 Track 5

Trac

k 5

Trac

k 4

Trac

k 3

Trac

k 2

Trac

k 1

1000096009200880084008000 10400 10800 11200 11600 12000

Fig. 13. Pre-job modeling of a side track well “M”, Chevron, North Sea. No real time azimuthal data was sent to the surface in this early test. The pilot well entered the shale roof multiple times. The side track well was then planned to avoid the shale roof. Pre-job modeling was completed for the standard propagation resistivity, gamma ray imaging, and for azimuthal propagation resistivity.

10 SPE 102637

Track 1 Track 2 Track 3 Track 4 Track 5

Trac

k 5

Trac

k 4

Trac

k 3

Trac

k 2

Trac

k 1

1000096009200880084008000 10400 10800 11200 11600 120001000096009200880084008000 10400 10800 11200 11600 12000

Post-job Modeling

Fig. 14. Post job modeling of gamma ray imaging, standard propagation resistivity, and azimuthal propagation resistivity for well “M”.Good agreement is reached between pre-modeled data and acquired logs. Post job analysis helped update the subsurface map.

SPE 102637 11

280-ft warning

ShaleSand

106001050010400 10700 10800

Trac

k 5

Trac

k 4

Trac

k 3

Trac

k 2

Trac

k 1

GR

Imag

e

Track 1 Track 2 Track 3 Track 4 Track 5 (Resistivity)TVD (ft)

+++++ Target direction2MHz (PD) Short2MHz (PD) Long400K (PD) Short400K (PD) Long2MHz (AT) Long

16-sector Azimuthalresistivity signals

Azimuthal resistivityimage

Azimuthal resistivitysignal strength

5150

5050

0.2

2000

Fig. 15. A short sectionof well“M” zooms in on the first entry into the shale roof from Figure 14. The azimuthal resistivity would have given 290 ft of warning of the shale roof approaching from above.

12 SPE 102637

870086008400 8800 8900

Trac

k 5

Trac

k 4

Trac

k 3

Trac

k 2

Trac

k 1

GR

Imag

e

Track 1 Track 2 Track 3 Track 4 Track 5 (Resistivity)TVD (ft)

+++++ Target direction2MHz (PD) Short2MHz (PD) Long400K (PD) Short400K (PD) Long2MHz (AT) Long

16-sector Azimuthalresistivity signals

Azimuthal resistivityimage

Azimuthal resistivitysignal strength

5300

4900

0.2

2000

8500 9000

Fig. 16. The low magnitude azimuthal propagation resistivity signal indicates that well “P” was clear from the shale roof.The residual signal in this instance may be due to non-symmetrical conductive invasion. In the interval 8760-8940fl, we see a conductive bed, probably shale, approaching from above, and then retreating slightly to the side. The gamma ray image does not detect this remote event.

SPE 102637 13

Trac

k 5

Trac

k 4

Trac

k 3

Trac

k 2

Trac

k 1

GR

Imag

e

Track 1 Track 2 Track 3 Track 4 Track 5 (Resistivity)TVD (ft)

+++++ Target direction2MHz (PD) Short2MHz (PD) Long400K (PD) Short400K (PD) Long2MHz (AT) Long

16-sector Azimuthalresistivity signals

Azimuthal resistivityimage

Azimuthal resistivitysignal strength

5300

4900

0.2

2000

115001130011100 1160011200 1170011400

Exiting the Reservoir at TD

Fig. 17. Near TD, well”P”dips down and encounters several thin streaks and an oil-water contact. The azimuthal propagation resistivity suggests that the streaks dip up, which is confirmed by the gamma ray image. The well finally enters the oil-water contact near 11,650ft.

14 SPE 102637

Trac

k 5

Trac

k 4

Trac

k 3

Trac

k 2

Trac

k 1

GR

Imag

e

Track 1 Track 2 Track 3 Track 4 Track 5 (Resistivity)TVD (ft)

+++++ Target direction2MHz (PD) Short2MHz (PD) Long400K (PD) Short400K (PD) Long2MHz (AT) Long

16-sector Azimuthalresistivity signals

Azimuthal resistivityimage

Azimuthal resistivitysignal strength

1580015700 1590015600

Fig. 18. In well “Q” a conductive event, probably an oil-water contact is first identified at depth 15605 ft, and seen approaching the well path from below. The well is steered upward to avoid entering this low resistivity bed. The maneuver allows to staymuch longer in the payzone.