SPE 160784

Addressing Wellbore Position Challenges in Ultra-Ex tended-Reach Drilling in Russia's Far East Benny Poedjono, Sheldon Andre Rawlins, Chandrasekhar Kirthi Singam, Alexander Van Den Tweel, Alexey Dubinsky and Rustam Rakhmangulov, SPE, Schlumberger; Stefan Maus, SPE, Magnetic Variation Services

Copyright 2012, Society of Petroleum Engineers This paper was prepared for presentation at the SPE Russian Oil & Gas Exploration & Production Technical Conference and Exhibition held in Moscow, Russia, 16–18 October 2012. This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract Drilling in Russia's Far East has always been associated with industry-defining ultra-extended-reach drilling. With the emergence of more powerful drilling rigs and advances in measurement- and logging-while-drilling (MWD and LWD) tools, these wellbores can be designed to reach farther. Therefore, accurately penetrating and exploiting distant reservoirs have resulted in critical dependence on high-accuracy surveying techniques.

Successful target penetration and meeting anticollision requirements without the need for shutting production in nearby wells are key proponents for a geomagnetic referencing service (GRS). Geomagnetic referencing is the technique to minimize the lateral position uncertainties when using MWD. This is particularly important for wellbores that extend the boundary of the drilling envelope with stepouts greater than 13 km. The wellbore azimuth accuracy is highly dependent on the quality of the magnetic data used to produce the geomagnetic reference model. This model characterizes the absolute magnitude and vector direction of the natural magnetic field for every point along the wellbore. Representation of the local crustal magnetic contribution is key to the process since it constitutes a significant error in the lateral wellbore position.

Since 2011, a new, highly accurate geomagnetic referencing methodology has been used in Russia’s Far East. Global contributions are accounted for by a high-definition geomagnetic model (HDGM). In addition, the local crustal magnetic anomaly is represented by 3D ellipsoidal harmonic functions tracking the shape and depth of the Earth, thereby providing seamless integration with HDGM and avoiding distortions faced by conventional plane-Earth approximations. A comparison with the previous industry standard shows improvements of 0.5° in azimuth determination. This high-degree geomagnetic technique will serve well for a number of upcoming developments in Russia’s Far East, continuing to push the drilling envelope and providing essential, accurate wellbore positioning, while offering significant time and cost savings. Introduction On the far Eastern Coast of the Russian Federation, there are a number of clients that have started and will continue to drill extended reach wells to exploit outlying hydrocarbon reservoirs. As a result of drilling these challenging wells, there would be a high dependence on providing highly accurate measurements that would result in reduced true vertical depth (TVD) and lateral uncertainty errors. These wells will be longer than any others drilled in the world, and thus any opportunity to minimize the uncertainty in the geomagnetic reference model used in MWD surveying must be taken advantage of. However, from previous drilling in the area, it has been observed from aeromagnetic study and field measurements that there exists a crustal anomaly effect that can impact the accuracy of survey measurements and thus introduce greater error. This, coupled with the fact of existing wellbore anticollision concerns in the earlier stages of the well and the challenging geological targets that had to be penetrated, required the creation of a more accurate geomagnetic referencing model. An improved geomagnetic reference model will bring significant value to the Project as the drilling is already complicated by the challenges of drilling at higher latitudes at high inclinations. Also, the well designs in this region are evolving into having longer stepouts, generating the need for an

2 SPE 160784

improved geomagnetic referencing model. In the ERD industry, a key performance indicator used in demonstrating the extent of complexity and challenges of an ERD well is an ERD Ratio of Horizontal reach to TVD. The current industry limit and records are the wells that have this ratio excess of 5:1 and 6:1. An envelope of this complexity can be graphically represented to define the limit that the industry is pushing and also provide a quantitative basis for comparison of various ERD Projects worldwide. This is shown in Fig. 1. In the Project site, the majority of wells that are planned are within the green circular which shows that these wells will continue to push the limit and boundaries of ERD.

Fig. 1— Envelope of Extended Reach Drilling Wells Key Perf ormance Indicator of ERD Ratio of Horizontal reach to True Vertical Depth Earth’s Magnetic Field Many Earth magnetic system processes generate magnetic fields, either primary magnetic fields or as electromagnetic induction in response to other magnetic fields. At any location on or near the Earth’s surface, the magnetic field B may be expressed as the vector sum of the contributions from three sources: the main field Bm due to the dynamo in the Earth’s liquid-iron outer core; the crustal field Bc from the magnetic mineral content of local rocks; and a disturbance field Bd from electrical currents flowing in the upper atmosphere and magnetosphere, which also induce electrical currents in the sea and ground. The relationship among these components can be summarized as:

B = Bm + Bc + Bd .

The main field Bm contributes about 95% of the total magnetic field and can be modeled by a spherical harmonic multipole expansion of the magnetic potential. The magnetic field intensity (magnitude) and inclination both increase towards the magnetic poles. The main field changes slowly over time, and this change is referred to as secular variation.

2:1 ERD Ratio

3:1 ERD Ratio

4:1 ERD Ratio

6:1 ERD Ratio

SPE 160784 3

The International Geomagnetic Reference Field (IGRF) model, updated every five years, is a mathematical description of the main field used in the scientific community. The change is assumed to be linear over five-year intervals. In oilfield directional drilling projects, a model called the British Geological Survey (BGS) Global Geomagnetic Model (BGGM) updated annually has been used instead of the IGRF. The recent development of the High Definition Geomagnetic Model (HDGM) brings greater accuracy in challenging wellbore positioning by taking into account the long-wavelength crustal magnetic anomalies.

The crustal field Bc is associated with induced and remnant magnetization within the crust. It is determined by conducting land, marine, or airborne magnetic surveys. For the Project location field, the crustal field was verified using a proprietary crustal modeling process developed by Magnetic Variation Services LLC (see the validation study by Poedjono, Montenegro, et al. 2012). When this methodology is combined with the global geomagnetic model, the local distortion of the geomagnetic field can be accounted for.

The (external) disturbance field Bd includes several components. Time variations are much more rapid than for Bc. Correcting for these variations can be accomplished by setting up a magnetic station at the drillsite or by forward modeling of data from existing remote geomagnetic observatories. In practice, Bc is extracted by removing Bm and Bd from the measured value of B at the local magnetic base station or observatory.

Correction of magnetic sensor readings for local variations in the Earth’s magnetic field is an essential component of effective geomagnetic referencing. An accurate map of the local magnetic field and accurate measurement of time-dependent local declination variations are two key elements of accurate geomagnetic referencing. Challenges Aeromagnetic surveys and field measurements show the presence of significant local magnetic anomalies; these impact the accuracy of survey measurements and introduce errors in the wellbore position. This has a direct impact on two critical aspects in both the planning and execution phases of the well design process.

Firstly, there would be existing wellbore anticollision concerns in earlier stages of the subject well that must be navigated against offset wells that may have some degree of geometrical misplacement with respect to their locations as indicated by the definitive survey for the wells. The collision concerns must be considered as significant. Anticollision standards and policies of the client and drilling services contractors must be adhered to and met, typically without shutting production in nearby wells.

Secondly, the geometric misplacement of wells within a productive hydrocarbon-bearing reservoir as a result of an inaccurate referencing model for a geomagnetic referencing service (GRS) can compromise the targets of detailed reservoir penetration and production plans. This would result in the wells not being in the “sweet spots” of the reservoir and thus having lower than optimal production rates of oil and gas. In severe cases, this can lead to wells placed completely outside of the reservoir or passing outside of the field lease lines.

Why More Accurate Geomagnetic Referencing is Needed A critical aspect of the provision of a GRS is the utilization of a suitable geomagnetic reference model. In previous drilling at the drillsite, the geomagnetic reference model used had been provided based on earlier study in 2003. The model had been built from public domain data of the U.S. National Geophysical Data Center (NGDC). However, since that time, there have been advances in data availability and processing methods, enabling us to attain higher resolution geomagnetic models with accurate specification of the strong crustal magnetic anomalies, which are prevalent in the area of new drilling. Furthermore, the large spatial extent of the anomalies in this region requires special techniques to account for long‐wavelength contributions that may not have been fully captured in the previous model.

In preparation for a new campaign in the area, the reference values had been rechecked. The comparison revealed significant discrepancies in the magnetic field strength and declination between the original model and the new HDGM.

Because of the strong crustal magnetic anomalies in this region (reflected in the differences between the reference values in Table 1, accurate specification of crustal magnetic anomalies is of particular importance for the Project location. Furthermore, the large spatial extent of the anomalies in this region requires special techniques to account for long-wavelength contributions. Thus, for the planned ERD wells at the drillsite, this difference and uncertainty in the magnetic declination could result in a significant geometric misplacement of the well.

4 SPE 160784

TABLE 1—SAMPLE MAGNETIC REFERENCE VALUES FOR LATITU DE=N 52 28 36.541, LONGITUDE=E 143 17 9.132, AT SURFACE

Magnetic Field Strength

Magnetic Dip

Magnetic Declination

BGGM 54358.8 66.384 -11.724

BGGM + Crustal 54400.6 66.447 -11.673

HDGM 54279.0 66.430 -11.530

In the meantime, eight years after the earlier study, two additional data sets have become available for the area:

1. The Russian Aeromagnetic Study (RAMS), sponsored by a set of major oil companies, resulted in an improved 0.5-km resolution grid for the whole of Russia.

2. Recent measurements of the German CHAMP satellite provided essential long-wavelength specification. The higher-resolution aeromagnetic coverage and accurate long-wavelength specification enable the production of a

significantly improved magnetic model for the far East Coast region of Russia. The scope of this work was therefore to provide an updated and more accurate, higher-resolution geomagnetic reference field for Project site field, covering the volume from the surface down to the maximum drilling depth and taking advantage of the refinements in geomagnetic referencing and the development of a robust geomagnetic referencing processing to provide precise, real-time positioning (Figs. 2-4).

Fig. 2—The difference between the measurement-while-drilli ng ellipsoid of uncertainty (EOU, in red) and geoma gnetic referencing (in blue), looking from the east and above.

SPE 160784 5

Fig. 3—The difference between the measurement-while-drilli ng EOU (in red) and geomagnetic referencing (in blu e), looking directly from the east.

Fig. 4—The difference between the measurement-while-drilli ng EOU (in red) and geomagnetic referencing (in blu e), looking from above. Quality Assurance/Quality Control of Available Magn etic Data Sets The global geomagnetic reference model used is the HDGM2011, which was produced by the US NGDC for the period of 1-Jan-2000 to 31-Dec-2012. In the meantime, the follow-on revision HDGM2012 is available, extending to the end of 2013 (Maus 2010; Maus et al. 2012). The HDGM represents the main field, the crustal field with a resolution of 28 km, and the stable portion of the disturbance field. When specifying the crustal field, there is a significant difference between magnetic anomalies referenced to HDGM2011 and anomalies referenced to BGGM2011, as can be seen in Fig. 5.

6 SPE 160784

Figure 5—Geomagnetic declination at 2000 m depth be low sea level, as given by (a) the British Global G eomagnetic Model, BGGM, (b) HDGM, and (c) HDGM combined with the local MagVAR m odel

To accurately estimate the geomagnetic reference field, it is important to consider the entire wavelength spectrum of the

geomagnetic field. Satellite observations cover wavelengths down to 250 km. The shorter wavelengths are best provided by local marine or aeromagnetic surveys. This study is well covered by large-scale aeromagnetic surveys. The coverage provided by the RAMS data was extended eastward using the NGDC grid, to improve the representation of intermediate wavelengths of 100 km to 250 km.

The long wavelengths of the crustal magnetic field are accurately provided by the German CHAMP satellite which orbited the Earth from 2000 to 2010. It was launched in July 2000 into a polar orbit at an altitude of 450 km, which slowly decayed over the lifetime of the mission. The last measurements were collected at about 250-km altitude, before the satellite burned in the atmosphere in September 2010.

Three data sets were available for the study area. Of these, only the commercial RAMS product of GETECH has the required quality to produce an accurate geomagnetic model.

The RAMS recovered the airborne measurements from flight-line profile maps at the original scale of 1:200,000 of data acquired in the 1960s, 1970s, and 1980s. GETECH digitized the data along the flight lines indicated in the maps. This is significantly more reliable than digitizing the contour lines, as was done for the NGDC grid described below. The flight line spacing was generally 2 km, although it drops to 4 km in some offshore areas, notably east of Sakhalin Island. The survey elevation was about 300 m above terrain. The data supplied were total magnetic intensity. These were digitized along profiles and reprocessed using GETECH microleveling methods, resulting in a grid at 0.5-km resolution. Fig. 6 shows the navigation of the RAMS data available for the far East Coast of Russia.

In 1974, the Ministry of Geology of the USSR published a series of 18 total intensity anomaly maps at 1:2,500,000 scale covering the area of the former Soviet Union. The contours on these maps were digitized in 1982 by the US Naval Oceanographic Office. The digitized contours were then corrected to remove spikes. A 2.5-km grid of total intensity anomalies from the corrected digital contour dataset, on a transverse Mercator projection, was then produced. The US NGDC then “unprojected” the data, on a per map basis, into the latitude and longitude coordinate system, regridded them on a 3-ft grid, and made the data available on a CD. This inferior-quality data set was used in the earlier study in 2003. It was also used here to fill in a void to the east of the RAMS data coverage, at a large distance from Project field, to improve the far-field.

SPE 160784 7

Fig. 6—Flight lines of the magnetic surveys from GE TECH’s (RAMS). The line spacing is 2 km on land and 4 km offshore. The dashed recta ngle shows the desired extent of magnetic coverage. The missing area to the east wa s filled in from the NGDC grid of the former Soviet Union.

In 2007, a team of Russian scientists contributed a 5-km resolution grid to the World Digital Magnetic Anomaly Map

(WDMAM). The source of this data was not disclosed. For the sake of completeness, this data was evaluated against the RAMS and NGDC grids. It was found to be of poor quality.

The RAMS data were compared with the public domain NGDC and Russian data. The results are displayed in Fig. 7 and Fig. 8 as maps, as well as in Fig. 9 and Fig. 10 as east/west and north/south profiles, respectively.

8 SPE 160784

Fig. 7—Overlay of RAMS data onto NGDC grid. The NGD C data set is seen to be shifted and has inferior r esolution.

Fig. 8—Overlay of RAMS data onto 2007 Russian contr ibution to WDMAM. This data set is also shifted and has even poorer resolution than the NGDC grid.

SPE 160784 9

Fig. 9—East/west profiles through the Project locat ion comparing the RAMS data with the public domain data. The x-axis gives longitude in degrees, while the y- axis displays magnetic total intensity anomaly in n anoTesla.

Fig. 10—North/south profiles through the Project lo cation comparing the RAMS data with the public doma in data. The x-axis gives latitude in degrees, while the y-a xis displays magnetic total intensity anomaly in na noTesla. Note the strong magnetic anomaly south of the drill site.

From the onset, the RAMS data appear the most trustworthy since the study used the best digitization and processing method by digitizing the paper maps along flight lines instead of along contours. Furthermore, the data were reliably georeferenced, while the NGDC grid appears to have been transformed multiple times between different coordinate systems.

The comparison in map and profile form shows that the RAMS data have significantly better resolution than the public data. Poor coordinate transformations have shifted anomalies northward in the public data. Shifts in the absolute level of about 200 nT are less of a concern, since the MagVAR processing ties the anomalies into the long-wavelength absolute reference of the satellite measurements.

While the RAMS data provide good coverage of the drillsite, optimal representation of long wavelengths requires a data set extending 130 km in all directions. As seen from the coverage map in Fig. 2, the RAMS data do not extend that far eastward. A section was therefore cut out of the NGDC grid to extend the coverage to the east. This section was adjusted to the level of the RAMS data at the seam, and was then merged without altering the RAMS portion of the grid.

A seamless transition is needed between the short-wavelength information provided by the airborne data and the long-wavelength information provided by the CHAMP satellite. The CHAMP-derived crustal field model was evaluated on a grid at the altitude of 300 m of the airborne data sets. The airborne grid was then draped onto the satellite grid, and the two were merged into a common grid of the total magnetic intensity (TMI) at 300-m altitude, which was used as an input to the geomagnetic modeling.

Challenges Unique to the Project Location Although the new HDGM+MagVAR geomagnetic reference model proposed for drilling in this field would provide improved accuracy of the survey measurements, a suitable work process had to be developed to handle the legacy survey data transfer. In addition to the fact that for this particular situation there were legacy data that were based upon an earlier and less accurate geomagnetic reference model, there is also the existence of the general situation that different drilling contractors may have recorded the well’s survey data inaccurately and the quality control of the data may be suspect. These imported surveys would need to have the appropriate tool error models and codes assigned to them. This is important since verifying the accuracy of any survey database before performing an anticollision scan is absolutely crucial.

10 SPE 160784

Tool error coefficients which constitute the tool error codes that define the rate of growth of the ellipsoids of uncertainty (EOUs) for the legacy data were developed to match those provided in the legacy survey data provided. These tool error coefficients would be consistent with the Revision 0 model as defined by the Industry Steering Committee on Wellbore Survey Accuracy (ISCWSA) and described in SPE 67617. The transformation of the legacy tool error codes in this database transfer process were well documented as part of a full management-of-change process for auditing and traceability purposes. It must be noted that the vast majority of error models are based on ISCWSA, but the interpretation of the uncertainty for specific tools and survey correction techniques in directional survey databases usually are specified by the survey contractor or client and are not necessarily the same. During this tool error code transformation process, in the previous drilling contractor’s survey database, there was one custom tool error code which will be defined as “Tool Error Code A” which included geomagnetic referencing and a reduction in the TVD error beyond the standard SAG-corrected model. The lateral reference for this model is based on a study commissioned in 2003 that determined the residual declination error in that specific field area to be 0.34°. To correctly map the Tool Error Code A, an analog tool error code which will be defined as CMAG+CSAG, was created to map this legacy custom model in the directional drilling and planning survey database. CMAG+CSAG incorporated a lateral error terms and coefficients which are effectively a slight downgrade from those of a standard GRS. This was based on the fact that the lateral uncertainties of the legacy survey data grew at a rate that were larger than the standard geomagnetic service tool error model, yet smaller than those of the typical advanced drillstring interference correction algorithms and standard tool error codes. A comparison is shown in Table 2. TABLE 2—COMPARISON OF LATERAL UNCERTAINTIES FOR VAR IOUS TOOL ERROR MODELS

Tool Error Code Description Semi -Major Axis Uncertainty at

Reservoir Entry (m)

Equivalent Lateral Error

(degrees)

MWD+GMAG Geomagnetic Reference Service 48 0.31

CMAG+CSAG Legacy Tool Error Code A 78 0.51

MWD+DMAG+DEC Multistation Drillstring Interference

Correction based on localized crustal data

119 0.77

MWD+DEC Standard MWD surveys based on localized crustal data

163 1.06

Following this, a comprehensive legacy data audit was performed as the final part of the process of incorporating the legacy

survey data in the new database where the new planning based on the new HDGM+MagVAR geomagnetic model would be done. This audit consisted of two fundamental steps:

1. Comparison was made of both the surface location and the bottomhole location (BHL) between the wells from the legacy definitive data.

2. Uncertainty analysis for the well path was performed based upon the comparison of the absolute values of the EOU components at the well TD and a comparison of the growth rates of the EOUs. EOU analysis was done at 95% confidence at 1-dimensionality.

One of the primary objectives of the legacy data audit was the verification that the CMAG+CSAG Tool Error Code would

suitably map the legacy Tool Error Code A. In the analysis phase of this audit, it was observed that the EOUs generated by use of the CMAG+CSAG model were in fact slighter larger than the EOUs generated by the Legacy Tool Error Code A in the order of 2 – 15% for the wells. This is shown by the second to last column (BHL Semi-Major Axis Uncertainty QC) in the below Table 3. As a result, this creates a safety factor in consideration of highly critical anticollision concerns during planning and drilling, based on Oriented Separation Factors which incorporated EOUs. As a result, competent anticollision proximity calculations were performed amongst the new well designs that will be using the improved HDGM+MagVAR tool error model against the legacy wells that are based on the BGGM referencing model which has a lower accuracy.

SPE 160784 11

TABLE 3—LEGACY DATA AUDIT SUMMARY RESULTS

QC refers to the difference in value between the drilling contractor’s drilling database for the legacy data and that for the new

drilling database with the improved GRS. Color code definitions for Table 3 are: • Red: EOU for that axis is more than 5% smaller than that for legacy data • Yellow: EOU for that axis is within 0% to 5% smaller than that for legacy data • Green: EOU for that axis is within 0% to 5% larger than that for legacy data • Blue: EOU for that axis is greater than 5% larger than that for legacy data

Highly Accurate High-Definition Geomagnetic Model V alidated The MagVAR model was validated in two ways: 1. By comparing the model prediction with ground measurements in the field as shown in Table 4 2. By comparing the predicted dip and total field with measurement-while-drilling (MWD) surveys at another field, farther north For further validation, MWD surveys were provided for three wells at the other field, north of the Project location. Since no gyro runs were taken, these MWD surveys can only verify the total field and dip; not the declination. The result of the comparison is shown in Fig. 11.

In Fig. 11, the residuals of the MWD surveys against the global BGGM and HDGM models are shown in green and red, respectively. The residuals against the local MagVAR model are shown in blue. Displayed on the x-axis is the measured depth. Wells start on land at a measured depth of MD=0 and then extend offshore to the east. Where the total field residuals (left plot) are negative, the models predict a stronger field than measured downhole. There may be some contamination by surface structures because the mean total field residuals go to zero in larger distances of the platform (large MD). This comparison shows that the MagVAR solution is in good agreement with the MWD surveys. In particular, the residuals are significantly smaller than when using either of the two global models.

TABLE 4—MagVAR AND GROUND MEASUREMENTS

No Lat Lon Date Measured (nT) MagVAR Residual (nT) 1 53.25986 143.2062 5/20/2002 54447.1 54434.8 12.3

2 52.29514 143.2151 5/21/2002 54492.1 54463.5 29.2

3 53.30969 143.2088 5/21/2002 54484.8 54471.7 13.1

4 53.30592 143.1998 5/22/2002 54486.3 54466.3 20.0

5 53.24569 143.2321 5/23/2002 54448.0 54428.4 19.6 Note: Had to subtract 155 nT in order to get their magnetic model (based on the NGDC public domain aeromagnetic data) to agree with these ground measurements.

12 SPE 160784

Fig. 11—Residuals of MWD surveys of three wells in an oilfield in the Far East Coast of Russia against the global BGGM and HDGM models and the local MagVAR model.

Finally, the MagVAR model was compared with the results of the earlier study. The earlier study report gives a table of

magnetic field predictions for surface and downhole points at the Project location. A comparison with the MagVAR model is shown in Table 5. TABLE 5—COMPARISON OF MagVAR MODEL FOR PROJECT LOCA TION WITH THE PREDICTIONS OF A EARLIER STUDY in 2003

Declination Dip Total Field Earlier MagVAR Difference Earlier MagVAR Difference Earlier MagVAR Difference

Landpad P1

0.05 0.29 -0.24° 0.06 0.09 -0.03° 47 -39 87 nT

P2 0.05 0.28 -0.23° 0.05 0.08 -0.03° 56 -41 97 nT P3 0.03 0.48 -0.45° 0.03 0.08 -0.05° 67 -46 113 nT P4 0.02 0.51 -0.49° 0.03 0.09 -0.06° 69 -2 71 nT P5 0.00 0.49 -0.50° 0.01 0.10 -0.09° 75 102 -27 nT

Offshore P1

-0.04 0.25 -0.29° 0.05 0.09 -0.03° 59 9 51 nT

P2 -0.06 0.26 -0.32° 0.05 0.08 -0.03° 56 -2 58 nT P3 -0.09 0.27 -0.36° 0.05 0.08 -0.03° 54 5 49 nT P4 -0.10 0.32 -0.42° 0.04 0.07 -0.03° 55 30 25 nT P5 -0.11 0.35 -0.45° 0.03 0.06 -0.03° 56 62 -6 nT

The differences for the dip are small, and the differences for the total field are moderate. The large differences in the

declination are surprising. The earlier study essentially predicted no declination anomaly at the Project location, while the MagVAR model shows a large declination anomaly of up to 0.5°. Judging from the lack of detail of the maps in the earlier study, this is due to the poor spatial resolution of their model. The difference in resolution between both models is illustrated in Fig. 12.

SPE 160784 13

Fig 12—Comparison of earlier study and MagVAR model s at the Project location. Shown here is the declin ation anomaly against BGGM at mean sea level. The earlier model appears to hav e very poor resolution, completely missing a signif icant declination anomaly at the Project location. Conclusions This study, employing the latest advances in geomagnetic referencing technology, finds substantial differences from a previous geomagnetic field study of 2003. The MagVAR model, based on redigitized and reprocessed RAMS data, resolves a significant declination anomaly of about 0.5°, whereas the earlier declination anomaly at the Project location is close to zero. The high-resolution MagVAR model agrees well with ground measurements and MWD surveys without applying any adjustments, whereas 150 nT had to be subtracted in the earlier study to reach agreement with ground shots. This was partly due to using poor-quality input data and incomplete representation of intermediate to long wavelengths of the geomagnetic field. This new study rectifies these deficiencies and reduces geomagnetic referencing uncertainties on the far east coast of a Russian Island.

Essential and accurate wellbore positioning allows the industry to push the drilling envelope, and to do so with significant time and cost savings. The high-resolution geomagnetic technique presented here offers substantial gains in accuracy, serving the needs of upcoming developments in Russia’s Far East.

Acknowledgments The authors appreciate the permission of Schlumberger and Magnetic Variation Services for their permission to publish the material contained in this paper. We thank them for their contributions and for ensuring that the operations were safely and successfully executed. References Maus S., M. C. Nair, Poedjono B., et al. 2012. High Definition Geomagnetic Models: A New Perspective for Improved Wellbore Positioning.

IADC/SPE Paper 151436-PP presented at the IADC/SPE Drilling Conference and Exhibition, San Diego, California, USA, 6–8 March. Poedjono B., Montenegro D., Clark P., et al. 2012. Successful Application of Geomagnetic Referencing for Accurate Wellbore Positioning in a

Deepwater Project Offshore Brazil. IADC/SPE Paper 150107-PP presented at the IADC/SPE Drilling Conference and Exhibition, San Diego, California, USA, 6–8 March.

Poedjono B., Adly E., Terpening M., and Li X. 2010. Geomagnetic Referencing Service—A Viable Alternative for Accurate Wellbore Surveying. Paper SPE 127753-PP presented at the IADC/SPE Drilling Conference, New Orleans, Louisiana, USA, 2–4 February.

SI Metric Conversion Factors Bbl × 1.589 873 E–01 = m3 ft × 3.048* E–01 = m hp × 7.460 43 E–01 = kW

*Conversion factor is exact.