Spe 30452 Ms

SPE 30452
Downloaded from http://onepetro.org/SPEATCE/proceedings-pdf/95SPE/All-95SPE/SPE-30452-MS/1967293/spe-30452-ms.pdf by Halliburton Energy Services Group user on 14 May 2021
Reduction of Well-Bore Positional Uncertainty Through Application of a New
Geomagnetic In-Field Referencing Technique
J. P. Russell, SPE, Broken Hill Proprietary Petroleum Ltd, G. Shiells, SPE, Sperry-Sun Drilling Services, and D.J.
Kerridge, British Geological Survey.
Copyrighl 1995. Soclety .f 1'............. EIIflm-•• 1lIe.

Both short term magnetic variations and errors in the
Thl. pap... wo. prepared f.r p........lad.o 01 Ibe SpE Annual TedmkaI Collf.......",4< modeled local magnetic field direction are ignored in this
Exhlbldoo had 10 Dalla•• U.S.A. U-:l5 Odob.... 1995.
surveying technique. Survey errors are potentially large,
Thl. pap... WOI .aeded f.r pr......tadob by m SPE Progl'llDl CommIttee foU.v.iJIg re\'lew.f
Iof.rmad.o <ootalned 10 m abl1nct .obmilled by Ibe .nlbor(.). Cool l. oflbe pap..., OJ giving rise to considerable directional uncertainty. Also the
pr......led, hav. 0.1 bem re\'I.\Wd by lb. SodeI)' of Pdtoleum Eogln md are .nllled 10 accuracy performance of some of the widely used drill-
<orrectloo by lb••nlb.r(.). Th. matorlal, OJ pr 1ed, don Dot o.ceilarUy refled my poilll.o
.r Ih. Sodety of J'dtoleum Eogln...... ll. om or memb..... p.p.... pr......1ed 01 SPE string interference correction methods 1.2 can be sensitive to
mectlDg. are .obJed 10 pobUralloD review by Editorial CoDlmlIteei of lb. SodeI)' of pdtokwo
Englo...... p...... I..Iou to alpy br...trlcted 10 an abllract of 001 more tbao 300 wordo. errors in certain magnetic parameters which are obtained
lUu.tradoo. may 001 be <opled. The ab.tract .houId albtaln alDlplmow adwowiedgm...1 of
wII.... and by wIIom Ibe pap... "pr......led. Write Llbrarbn, SpE. P.O. BOI833836.
from the main-field model and used as input values. Well
RJdlard.oo, TX 75083-3836. U.S.A.. fal 00.1.214·951-9435. bore positional uncertainty tends to be greater for magnetic
surveys than for high accuracy gyro systems. This is due, to
a considerable extent to uncertainty in declination and the
Abstract. Iinfluence of drill string interference.
A new In-Field Referencing (IFR) technique, for measuring Measurement of the local geomagnetic conditions
local geomagnetic-field parameters at, or very close, to the during surveys can significantly reduce survey errors and
well site is described. IFR is shown to enable reduction in directional uncertainty. IFR therefore, has the potential to
the uncertainty associated with the estimates of improve the accuracy of magnetic surveys. Previous authors
geomagnetic field values, normally obtained from main- 3.4.6 have indicated that local measurement of the
field geomagnetic models. This significantly reduces the geomagnetic field can result in lateral position uncertainties
magnitude of certain critical directional uncertainty error for modern magnetic survey tools approaching those
terms which are inputs to survey tool accuracy performance normally reserved for high -accuracy gyros.
models, reducing positional uncertainty for well-planning IFR has been actively discussed for several years by
purposes. several major oil operators and service-contractor
Logistical problems and high unit-cost have precluded IFR companies, but without significant progress. The practice of
development to date, but recent innovative thinking which IFR is not entirely untried, at least on land locations, but
has evolved from dialogue between oil and drilling-service there is a paucity of public domain information describing
companies and the scientific/academic community has practical applications of the technique. There are no
refined the concept and greatly improved the feasibility. reports in the literature of IFR being applied to offshore
drilling activity.
Introduction. It is the aim of this paper to redress this situation and
In-field-referencing (IFR) is the name given to the practice also to describe a recent refinement which could lead to
of measuring the geomagnetic field at, or close to, a drilling wider usage and application of IFR, including offshore
site. drilling projects. This in turn could lead to significant
Magnetic survey tools rely on the principle of measuring drilling cost reductions in high-cost drilling areas such as
the direction of the well bore relative to the direction of the the Gulf of Mexico and the North Sea. This would be
local geomagnetic field. The hole direction can then be achieved via optimisation of directional survey programs,
referenced to a geographic coordinate system from a and also reduction in the quantity of non-magnetic
knowledge of the direction of the geomagnetic field, hardware in BRAs.
relative to true north. This true north reference direction is
normally obtained from charts or from a computerised
model of the geomagnetic main-field.
2 APPLICATION OF A NEW GEOMAGNETIC IN-FIELD REFERENCING TECHNIQUE SPE30452
The Geomagnetic Field. momentary bursts of actiVity can be seen on magnetograms

A description of the geomagnetic field has been provided by as spikes varying randomly around a mean (for 0,1 and F),
earlier oil industry authors (e.g. Ziesling6 ), and detailed Fig. 2 shows a magnetogram for Lerwick observatory, for a
descriptions are found in the geoscience literature (e.g. disturbed period in June 1991. The magnetic disturbances
Stuart & Barraclough 8). A knowledge of the sources and lasted for a period of 4 days, and while discrete spikes are
magnitude of magnetic field variations and magnetic obvious, some of the m~or events lasted several days. This
directional reference errors assists understanding of the means that a magnetic directional survey carried out during
scope of the magnetic survey accuracy problem and the such a disturbed period may have directional error elements
potential global value of the IFR technique described in this which are systematic in nature, and not random. This has
paper. A concise description of the geomagnetic field is, implications for the designers of survey instrument
therefore, uppropriate here. performance models.
The geomagnetic field at any point on the Earth's surface is Crustal anomalies are widespread and may be locally
8
the resultant of fields from three principal sources. The abundant • Geomagnetic disturbances may extend from
main-field originating in the Earth's core, accounts for the polesweU towards the equator, although generally their
about 98.99% of the field strength at most places at most effects diminish with decreasing latitude. The geomagnetic
times. Superimposed on the main-field are fields created by field for any location at any time, calculated from a main-
the magnetisation of rocks in the Earth's crust and by field model alone may, therefore, be significantly in error.
electric currents flowing external to the Earth's surface. A North Sea drilling activity is not alone in potentially
part of the upper atmosphere, the ionosphere, is ionised by suffering from the effects of crustal anomalies, diurnal
absorption of solar electromagnetic radiation, and relatively variations, and geomagnetic disturbances, Errors and
regular diurnal magnetic field variations are caused by uncertainty associated with the main-field model are a
currents flowing there. Charged particles emitted from the global drilling problem wherever magnetic survey tools
Sun form the solar wind which interacts with the main including MWD systems are employed.
geomagnetic field causing irregular and often rapidly·
varying field changes. Energetic events on the Sun's IFR Methodology.
surface, such as solar flares, change the state of the "space Absolute measurements of the geomagnetic field are made
weather" and the result on Earth may be major routinely at magnetic observatories. The standard for
geomagnetic disturbances, usually called magnetic storms. measurement of F is the proton precision magnetometer,
The geomagnetic field may be defined in terms of three and its operation can be automated easily. The modern
magnetic elements: total field strength (F), dip angle (I) standard instrument for 0 and 1 measurements is a non-
and declination (D). A number of estimates of the errors in magnetic theodolite with a fluxgate sensor mounted on its
F, I, and 0 from a main-field model (lGRF) for North Sea telescope. Procedures have been described by Kerridge9 .
areas, are reported in the literature (eg,Wolff & de Warde, For measurement of 0, a true north reference is required.
Zijslinl). Thorogood4 reports that the standard deviations This may be determined by means of astronomical
of variations in F, I and D for the North Sea area are observations or by using a higlhaccuracy north-seeking
300nT, 0.3 deg, and 0.9 deg, respectively. This information gyro mounted on the theodolite.
was abstracted from a British Geological Survey study Accurate geomagnetic measurements are possible at
report which was specifically commissioned for the purpose land locations, provided care is taken to avoid interference
of determining uncertainties associated with the from the rig and drilling hardware, and other man-made
geomagnetic field in the area of the North Sea, These sources of magnetic interference such as underground
values should, therefore, be considered as currently the pipes, overhead cables and buildings. IFR has occasionally
most reliable available, and they are close to the estimates been used on land drilling projects or survey tool field trials
provided by Zijsling6 . Thorogood4 indicates that these at land locations, but it has not so far been possible at
figures represent the combined effect of IGRF model error, offshore drilling rigs.
diurnal variations and crustal anomalies, Larger errors, pf Past thinking in this area focused on placing a
the order of several degrees in D, 0.5 degrees in 1 and more magnetometer; coupled to a North seeking gyro system, on
than 500 nT from the crustal field alone are known to occur the sea bed close to the rig site. The device would be left in
in certain North Sea locations. position during the drilling phase and data would be
The main·field model in 'informed' usage today transmitted from the sea bed to the drilling operation.
(BGGM) is an improvement in many respects, over the There are many problems associated with this approach
IGRF model. However, information on the magnitude and including continuity of power supply over extended periods,
frequency of magnetic disturbances (Fig. 1) in the North logistics and transportation, data transmission and control
Sea area between 1983 and 1994 indicates that the number of sensor calibration shifts. The IFR unit cost would be
of days when significant magnetic disturbance occurred is high, and for North Sea locations at least one unit would
greater than previously estimated. Despite improvement in likely be required per drilling site.
the main-field model, it is therefore probably unsafe to
assume the error values can be reduced.
A common misconception about magnetic disturbances
is that .magnetic storms are short term events where
SPE30452 J. P. RUSSELL, G. SHIELLS, D.J. KERRIDGE 3
A New IFR Technique. Directional Survey Program.

A new IFR technique, described here, potentially obviates A typical well profile for the Liverpool Bay development is
the need for a fixed permanent magnetometer at the drill shown in Fig. 4. The wells tend to have relatively shallow
site. Close to the drilling site, but sufficiently remote to kick-offs in the top-hole section, then build rapidly to high
avoid significant interference, a series of 'spot' or 'snap angle for the long tangent section before falling off towards
shot' measurements of the absolute values of magnetic field the reservoir.
intensity and direction are made. These measurements are The choice of a directional survey program proved
used to establish base-line differences between the rig-site problematical. Using conventional survey program design
magnetometer and one or more fixed magnetometers which philosophy, the long high-angle tangent section would
can be several hundreds of kilometers remote from the drill normally have to be surveyed with a gyro tool to ensure that
site. Permanent geomagnetic observatories such as those drilling and geological objectives were achieved. From the
operated by British Geological Survey at Lerwick, Hartland practical standpoint however, in wells of more than 70
and Eskdalemuir are ideal for this purpose for offshore UK degrees inclination, running wireline gyro surveys inside
waters. An estimate of the absolute field intensity and casing becomes impossible. Specific reservoir features
direction at the rig-site can then be calculated at any other (faulted structures) made it imperative to reduce positional
time, from the observatory data for that time and the base uncertainty during drilling. IIFR applied to MWD and
line information. The new technique involves interpolation EMMS appeared to offer an effective solution to the
of data from one or more locations to another and in order problem. Where reasonably possible, high accuracy gyros
to distinguish it from standard IFR techniques it has been were run to act as a check on IIFR performance.
decided to label it IIFR (Interpolation IFR). Fig. 5 shows lateral positional uncertainty for a number
of different survey programs, applied to the typical well
Application of IIFR to the Liverpool Bay Development. profile. A modified Wolff & de Warde error model was
Liverpool Bay Geomagnetics. Table 1 shows D, I, and F employed to calculate the lateral positional uncertainties.
uncertainties in the BGGM values for the Irish Sea area as For IIFR magnetic survey tools certain of the error
estimated by the BGS. Total uncertainty, including the parameter values (compass reference error and drill-string
uncertainty due to the crustal part of the field and that due interference) were reduced in line with the predictions of
to variations are included. These values are only a little the uncertainties in the geomagnetic field parameters,
lower than North Sea values provided by Thorogood4 • shown in Table 2 (Option 3). A performance model for a
In a study commissioned by Sperry-Sun Drilling drill-string magnetic interference correction method 3 was
Services, from analysis of aeromagnetic data for the specific also applied.
area of Liverpool Bay, BGS concluded there are no large The lateral positional uncertainty curves (Fig. 5)
spatial gradients between the offshore drilling site and indicated that both target objectives could be achieved using
landfall on North Wales near Rhyl (Fig. 3). either Program 2 or 3. The latter involved the use of a top
BGS estimated that by combining modeling of existing to bottom gyro survey inside casing, which due to
aeromagnetic data and taking observations at a suitable operational limitations imposed by inclination was not a
nearby land-fall site, then uncertainty for the crustal practical option. Moreover, it was important to determine
elements at the drilling site could be reduced to 5', 2' and position whilst drilling and prior to setting of the casing.
30 nT for D, I and F respectively. Further improvements in Survey Program 1 involved the use of MWD (BGGM)
the total uncertainty could then be achieved by monitoring during drilling, a north-seeking gyro to maximum
and correcting for daily variation and disturbance fields practicable angle and then an EMMS, with application of
during the drilling phase. corrections for angular misalignment, bias errors, and BHA
Three options were examined. Option 1 was to use data deformation 4 • Program 2 appeared to provide a potentially
from the BGS fixed magnetic observatory at Eskdalemuir, practical solution.
which is about 200 km north of the site, to approximate the This appeared to provide an excellent opportunity to test
external field effects at the drilling site. Option 2 was to use the new I1FR technique. The I1FR technique had not been
data from Eskdalemuir as well as from Hartland applied before and therefore it was decided to test all three
observatory which is about 300 km south-south west of the options.
site. Option 3 was to set up an automatic observatory at the A suitable land site, free from man-made sources of
land fall site to measure and record data during drilling. interference was identified by BGS at the Point of Ayr (Fig.
Since the distance from the land site to the drilling site 3). The 'spot' and variation measurements were carried out
was only about 35 km, and the spatial field gradients are using techniques and equipment developed by BGS. For the
known to be small, the external field variations measured at variation measurements a concrete plinth was laid in the
the fonner should be very close to those that would be ground to provide a stable platform for the fluxgate
observed at the latter. Table 2 shows the estimated total magnetometer sensors during the period of the trial. The
uncertainties in BGGM values that result when each of the equipment was protected bya non-magnetic 'igloo'
three options is used in conjunction with the 'spot' supplied by BGS.
observations. Any of the three options would allow One-minute values of D, I, and F from the monitoring
considerable improvement in error-model parameter values. station at the point of Ayr were transmitted to BGS in
Edinburgh, where data from Eskdalemuir and Hartland
4 APPLICATION OF A NEW GEOMAGNETIC IN.FIELD REFERENCING TECHNIQUE SPE30452
The mean values of the differences in D, I and F between the quality control and diagnostic aspects of magnetic surveys
estimates from the monitoring site data and the BOOM fer and IIFR. This will be the subject of separate publications.
the drilling site are 0.17 deg., 0.05 deg., and 90nT,
respectively. The standard deviation data indicates Discussion
predominantly small geomagnetic field variations during the The new IIFR technique produced good results in this trial,
trial. allowing considerable reduction in uncertainty ci
Table 5 shows the standard deviations of monitor versus geomagnetic parameters at the drilling site. The resulting
IIFR measurements for the trial period. Values are less than reduction of magnetic survey directional uncertainty and
0.01 deg for D and I and less than 10 nT in F. This is lateral positional uncertainty enabled optimisation of the
strongly indicative of the performance capability of the IIFR survey program on the Liverpool Bay Development project.
technique. The wells were drilled successfully with no directional
Table 6 shows standard deviations of D, I, and F during a problems resulting from the application of the IIFR
disturbed day (as in Figure 6). Both the variation data and technique. Geological indicators strongly suggested that the
the differences between monitor and IIFR are in Hne with the drilling objectives had been successfully achieved in all
values shown in the long term data set, suggesting there is wells. IIFR is considered to have been successful in this
no degradation during moderate magnetic disturbances at project.
least. The IIFR technique and monitoring at Point of Ayr were
Although the data distributions are not truly gaussian, carried out simultaneously for a period of4 months. The
these results imply the 99% confidence limits for IIFR, agreement between the two methods is such that virtually
based on this trial are in the order of 0.02 deg, 0.0] deg any 'snap-shot' taken at the point of Ayr could have been
and 12 nT, for D, I, and F respectively. used to establish base-line data for the IIFR method. This
has clear implications for the design of an offshore
Directional Survey Results. Application of I1FR data to monitoring device, which could be small, mobile, have low
magnetic surveys should in theory reduce observed power requirements and would require to be in the vicinity
directional differences between gyro and magnetic surveys. of the drilling rig for just a few hours. Sensor calibrations
The magnitude of improvement in match between such could be checked on a per occasion bases.
surveys is dependent upon the size of the errors associated The results of this study demonstrate that the technique
with the BGGM values, the amount of drill-string is successful in achieving significant reduction in
interference, and the performance of the respective survey geomagnetic uncertainty. However, the comparison between
systems. monitored and interpolated IFR D, I, and F data indicated
The differences between BOOM values and mean that there was some deterioration during periods of magnetic
monitored values (from Table 4) are 0.05 deg, 0.] 7 deg, disturbance. Contact with interested potential end-users in
and 90 nT respectively for D, I, and F. The IIFR corrections the Industry suggests that uncertainties of 0.] deg, 0.05 deg,
to declination are in this case relatively small. Comparison and 50 nT, in D, I, and F are desirable targets for IIFR
of long axis magnetic sensor in this case data with I1FR data performance. Deterioration during periods of magnetic
indicated that drill-string interference fields were disturbance was in fact not severe, and the results reported
predominantly in the range 50 - 100 nT. Observed drill- here are all well within the Industry target. This
string interference corrections were generally of the order ci performance deterioration during magnetic disturbances has
0.5 deg or less. possible implications for latitude dependent limitations to
Table 7 shows the difference between BOOM MWD data the technique.
and IIFR MWD data versus high accuracy gyro runs. In
virtually all cases the match between the magnetic tool and Error Sources. IIFR is a method for removing errors
the gyro is improved by the application of IIFR. The caused by crustal anomalies, external disturbances and from
improvements are also consistent with the magnitude of the the main-field model itself. The interpolation technique
declination accuracy improvement and the apparent levels ci although designed to reduce these errors is in tum likely to
drill-string inteJference. In the case of Well 8, the lack ci suffer errors during implementation. These errors include;
improvement is thought to be due to cross-axial interference
along a considerable length of the section from an adjacent I) An error associated with the application of crustal
well. Standard deviations, however, are not noticeably anomaly correction between the "snap shot" site and the
reduced. This is thought to be due to some residual sensor drilling site.
bias, the effects of which can become magnified on
application of this type of magnetic interference correction. 2) Errors associated with the algorithm used to estimate
There was no overlap of gyro and EMMS surveys, the external variations at the 'snapshot' site from variations
therefore no similar analysis was possible. Comparison measured at one or more distant observatories.
between lIFR EMMS surveys and lIFR MWD revealed
good general agreement, with the rotation-corrected EMMS The error associated with application of the crustal
providing the definitive surveys (Table 8). The major part ci anomaly correction could be significant if the distance
the improvement in the match between the MWD and between the sites is a few tens of kilometers, especially if
EMMS surveys came from application of the rotation there are significant local spatial gradients in Earth's
correction rather than directly from application of IIFR. In all geomagnetic field.
cases drill-string magnetic interference was small. In the current study, the proximity of the 'snap shot' site
At the time of writing this paper much of the complex was indeed within a few tens ofkilometers, but importantly
analysis of the magnetic surveys was ongoing. The data sets a detailed study of aeromagnetic data revealed extremely low
contain much valuable information, particularly on the spatial field gradients across much of the Liverpoo] Bay area.
SPE 30452 J. P. RUSSELL, G. SHIELLS, DJ. KERRIDGE 5
For application of the technique in other locations, either significantly from several hundreds of kilometers to several
similar gradient conditions would have to exist or the 'snap tens of kilometers. Further research is required to quantifY
shot' station would have to be much closer to the drilling the error magnitude in these areas.
site. In offshore applications, great care would have to be
taken to ensure sampling of reference data took place at IIFR Potential. IIFR has the potential to reduce the
sufficient distance from the rig installation to ensure number of gyro survey runs in a survey program, resulting
anomalous free data. Also, although a study of local in cost savings. IIFR should also assist in the elimination
aeromagnetic charts would reveal the presence of field of the dilemma that frequently occurs when gyro and
gradients, sampling program design should be sufficient to magnetic surveys disagree. Errors such as sensor bias, drill-
confirm local spatial gradient conditions. The ideal case string interference, cross axial magnetic interference and
would allow samples to be taken along a transect coincident poor calibration become more easily identifiable. A
in direction and length with the path taken by a well bore. reasonable analogy is that IIFR is like providing a lifelong
This approach would be particularly suitable for extended myope with spectacles for the first time; clarity of vision
reach applications and in areas of high field gradients. becomes remarkable.
The effectiveness of the algorithm has been demonstrated Reduction of the largest single sources of magnetic survey
very well in the present study. This has been possible tool uncertainty may also provide the justification for the
because a monitoring instrument was placed at the 'snap development of a new generation of enhanced performance
shot' site to measure the variations which were then magnetic survey systems, with improved downhole
compared with the predictions produced by interpolation diagnostics.
between Eskdalemuir and Hartland.
The magnitude of the errors associated with the Future Work.
algorithm is dependent on four main things; Detailed analysis of magnetic surveys from the Liverpool
Bay project is ongoing. Further work will include an
a) The latitude of the area which the fixed observatory exploration of the full diagnostic potential of IIFR. The
and the 'snap shot' are located. (This is significant feasibility of an offshore IIFR unit is currently being
primarily because magnetic disturbance activity is examined. The limitations of the technique are currently
generally greater in amplitude and frequency at under investigation and an IIFR accuracy performance model
higher latitudes). is being derived.
b) The separation of the observatory from the 'snap Conclusions.

shot' site. A new technique for estimating geomagnetic field elements
shows good agreement with near-site monitoring
c) The number of observatories used in the measurements. The technique (IIFR) has considerable
interpolation. potential in reducing directional uncertainty in magnetic
survey tools allowing optimisation of directional survey
d) The data quality from the sensors used in the fixed programs. The new technique also offers scope for real time
observatories and 'snap-shot' locations. enhanced, downhole diagnostics.
Development of IIFR capability for offshore drilling projects
To summarise the effects, if the 'snap shot' site is at high should now be much more feasible.
latitude, but close to a fixed observatory, smaller errors
could be expected than from a medium latitude site which is Nomenclature
further from a magnetic observatory, even though the effects D Declination
of disturbance at the higher latitude site would be greater. I Magnetic Inclination
F Total Field Intensity
In an earlier study commissioned by SSDS, BGS observed BGGM British Geological Survey Global Geomagnetic
that where D, I, and F values were predicted for Lerwick, Model
(which is representative of a moderately high latitude EMMS Electronic Magnetic Multishot System.
location) from interpolation of data derived from several I1FR Interpolation In-field-referencing.
European observatories and compared with actual Lerwick BGS British Geological Survey
observatory measurements, then, even during a highly BHPP Broken Hill Proprietary Petroleum
disturbed year (1991) it was possible to predict field values SSDS Sperry-Sun Drilling Services
at Lerwick within the Industry target specification (D< BHA Bottom Hole Assembly.
0.I,F<50nT, 1<0.1) approaching 95% of the time. IGRF International Geomagnetic Reference Field
Judicious use of magnetic storm forecasts could improve this
performance further. Acknowledgments
Errors associated with the I1FR technique can therefore be The authors wish to thank BHP Petroleum Ltd and their
minimised, even in high latitude areas. The distance Joint Venture partners; Powergen (North Sea) Ltd, Lasmo
between 'snap shot' site and observatory should be pIc and Monument Oil & Gas pIc and also Sperry-Sun
controlled (within several hundred kilometers). In addition Drilling Services and the Director of British Geological
data from more than one observatory improves performance. Survey for permission to publish this paper. Special thanks
In low latitude locations, such as the Gulf of Mexico and are due to John Riddick and Toby Clark of BGS, and to
Gulf of Suez, accuracy performance of the IIFR technique Phil Gurden and Patrick Knight of SSDS.
should be at least as good as that observed in the present
study. In extreme high latitudes, separation of the 'snap
shot' site from the observatory may have to be reduced
(, APPLICATION OF A NEW GEOMAGNETIC IN-FIELD REFERENCING TECHNIQUE SPE30452
in cost savings. I1FR should also assist in the elimination

of the dilemma that frequently occurs when gyro and
magnetic surveys disagree. Errors such as sensor bias, drill-
string interference, cross axial magnetic interference and
poor calibration become more easily identifiable. A
reasonable analogy is that IIFR is like providing a lifelong
myope with spectacles for the first time; clarity of vision
becomes remarkable.
Reduction of the largest single sources of magnetic survey
tool uncertainty may also provide the justification ipr the
development of a new generation of enhanced performance
magnetic survey systems, with improved downhole
diagnostics.
Future Work.
Detailed analysis of magnetic surveys from the Liverpool
Bay project is ongoing. Further work will include an
exploration of the full diagnostic potential of IIFR. The
feasibility of an offshore IIFR unit is currently being
examined. The limitations of the technique are currently
under investigation and an IIFR accuracy performance
model is being derived.
Conclusions.
A new technique for estimating geomagnetic field elements
shows good agreement with near-site monitoring
measurements. The technique (IIFR) has considerable
potential in reducing directional uncertainty in magnetic
survey tools allowing optimisation of directional survey
programs. The new technique also offers scope for real
time enhanced, downhole diagnostics.
Development of IIFR capability for offshore drilling
projects should now be much more feasible.
Nomenclature
D Declination
I Magnetic Inclination
F Total Field Intensity
BGGM British Geological Survey Global Geomagnetic
Model
EMMS Electronic Magnetic Multishot System.
I1FR Interpolation In-field-referencing.
BGS British Geological Survey
BHPP Broken Hill Proprietary Petroleum
SSDS Sperry-Sun Drilling Services
BHA Bottom Hole Assembly.
IGRF International Geomagnetic Reference Field
Acknowledgments
The authors wish to thank BHP Petroleum Ltd and their
Joint Venture partners; Powergen (North Sea) Ltd, Lasnto
pic and Monument Oil & Gas pic and also Sperry-Sun
Drilling Services and the Director of British Geological
Survey for permission to publish this paper. Special thanks
are due to John Riddick and Toby Clark of BGS, and to
Phil Gurden and Patrick Knigltt of SSDS.
SPE30452 J. P. RUSSEll, G. SmELLS, D.J. KERRIDGE 7
References
1. Van Dongen, J.C.M., and Maekiaho, L.B.: "Method For
Detennining the Azimuth of a Borehole", U.S. patent no
4.682,421,28.07.1987.
2. Russell, M.K., and Russell, A.W., "Swveying of Boreholes"

U.S. patent no. 4,163.324, 07.08. 1979.
3. Thorogood, J.L.: "Instrument Perfonnance Models and Their
Application to Swveying Operations." paper SPE 18051
presented at the 1988 SPE Annual Technical Conference and
Exhibition, Houston, Oct.2-5. .
4. Thorogood, J.L. and Knott, n.R.: "Swveying Techniques

With a Solid State Magnetic Multishot Device." paper
SPE/lADC 19030, presented at the 1989 SPE/lADC Drilling
Conference, New Orleans, Feb. 28 - Mar. 3.
5. de Lange, J.L. et al: "Accurate Swveying; An Operator's

Point of View." paper IADC/SPE 17213, presented at the
1988 IADC/SPE drilling Conference, DaIlas, Feb. 28-Mar. 2.
6. Zijsling, D.H. and Wilson, R.A. : "Improved Magnetic

Surveying Techniques: Field Experience. paper SPE
1923911, presented at Offshore Europe 89, Aberdeen, Sept. 5 -
8.
7. Wolff, C,J.M. and de Wardt, J. P.: "Borehole Position

Uncertainty - Analysis of Measuring Methods and Derivation
of Systematic error Model." paper SPE 9223 presented at
the 1980 SPE 55th Annual Technical Conference and
Exhibition, Dallas, Sept. 21-24.
8. Stuart, W. F. and Barraclough, D.R. : "Magnetic Reference in

Navigation - The Earth as a Magnet", Journal of Navigation,
v41 p336, no 3. Sept. 1988.
9. Kerridge, DJ.: "Theory of the Fluxgate-Theodolite." British

Geological Swvey Geomagnetism Research Group Report
88/14, 1988.
APPLICATION OF A NEW GEOMAGNETIC IN·FIELD REFERENCING TECHNIQUE SPE30452
Geomagnetic Total Error Crustal Variaticn

Elcmoot COl1lIJcnoot COl1lIJcnoot
Declination 0.67° 0.43° 0.5°
Inclination 0.33° 0.17° 0.28°
Total Field 250 nT 147 nT 202 nT
Table I : Estimated Uncertainties in BOOM Values at the
Liverpool Bay Drilling Site
Geomagnetic Opticn I Opticn 2 Opticn 3

Element
Declination 0.27° 0.18° o.oao
Inclination 0.13° 0.08° 0.03°
Total Field 85 nT 60nT 30nT
Table 2: Estimated Total Uncertainties In BOOM Values
Opticn I: Initial map..sbot at Point of Ayr

Opticn 2 : Initial map..shot, plus data from &kdalemuir
Observatory data from BGS observaticns at
EskdaJemuir and Hartland
Opticn 3 : Initial map-shot, plus set up of automatic
recording staticn at Point of Ayr
Month D (del{) I (del{) F (nT)

Oct 94 0.062 0.027 12.7
Nov 94 0.046 0.022 7.4
Dec 94 0.038 0.Q16 7.1
Jan 95 0.054 0.019 8.2
Table 3 : Monthly Standard Deviations of Hourly

Measurements at Liverpool Bay
Month D (del{) I (del{) F (nT)

Oct 94 -5.941 68.025 48679.4
Nov 94 -5.923 68.Q11 48684.9
Dec 94 -5.891 68.046 48682.9
Jan 95 -5.861 68.044 48679.9
Table 4 : Monthly mean values of monitor measurements.

SPE 30452 J. P. RUSSELL, G. SHIELLS, D.J. KERRIDGE 9
Month D (deg) I (deg) F (nT)

Oct 94 0.006 0.005 5.3
Nov 94 0.008 0.006 3.4
Dec 94 0.007 0.003 5.3
Jan 95 0.005 0.002 4.4
Table 5 : Monthly standard deviations of (monitor-IFR) data.
D (deg) I (deg) F (nT)

Variations 0.069 0.053 22.5
Difference 0.008 0.007 7.2
Table 6 : Standard deviation of one-minute values of 0,

I and F and (monitor-IFR) measurements for the day of
29.10.94.
Mean Differmce (deg) to gyro SD of difference (deg)

survey to gyro survey
Well No BGGM IIFR BOOM IIFR
1 0.86 0.51 0.48 0.33
2 0.57 0.04 0.42 0.51
3 0.31 0.17 0.83 0.46
4 1.04 0.68 1.24 1.18
5 0.45 0.08 0.24 0.24
6 0.54 0.27 0.63 0.57
7 0.55 0.23 0.42 0.29
8 0.32 0.38 1.79 3.47
Table 7 : Comparison of difference between magnetic surveys processed

using BGGM model data and IIFR data respectively, versus north seeking gyro surveys
Divergence FtJ 1000 Ft

Well No. BGGM IIFR
1 10.2 4.21
2 3.54 3.54
3 4.85 3.70
4 2.37 1.43
5 9.1 4.82
6 14.36 2.65
7 11.36 1.81
8 8.82 8.82
Table 8 : EMMS vs. MWD Surveys in 12 1/4" hole Sections
Note: IIFR data has been processed with rotation, BRA sag, and drill-string interference corrections where appropriate.
BOOM data has no rotation corrections applied and uses BOOM D, I and F. BRA sag corrections also applied.
J() APPLICATION OF A NEW GEOMAGNETIC IN·FIELD REFERENCING TECHNIQUE SPE 30452
Fig. 1 : Frequency of Magnetic Disturbances at LerWck Observatory 1983-1994
Number of days when the range of declination exceeded

1 degre.e and 0.5 degrees at Lerwick Observatory
140 ~
120
e: I'l!"
100 "" ~ e:
80
~' "",.
.> 1 deg
60 1" ,& C> 0.5 deg

~
1"
40
lJ - ~ ~ ~ J
20
lJ ~
o
- ~
Number of days when the range of inclination exceeded

0.5 degree and 0.25 degrees
90
80
70
60
50 .> 0.5 deg
40 C> 0.25 deg
30
20
10
0
1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
Number of days when the range of total field intensity

exceeded 400 nT and 200 nT
140 ~
120 ~'
~
100
80
~' ~
~' ~. ""', .> 400 nT
60 C> 200 nT
,& ~
40 rI-' "",.
20
o
lJ J ~ J
1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
~ J~
SPE30452 J. P. RUSSELL, G. SHIELLS, D.J. KERRIDGE II
Geomagnetism Group
EGS Edinbur h LERWICK Lat:60 08 Lon :358 49
Declination in degrees east
I 60min
Inclination in degrees
72.7 J 40min
Total intensity in nT
5014 ! I 300nT
09 10 11 12 13
June 1991
Fig. 2 : Magnetogram for Lerwick Observatory for a Disturbed Period During 1991
Liverpool Bay total intensity anomalies

from BGS aeromagnetic data (nT)
430 -r---~---"'------------r-T-""'--r"r'---'---'
420 ~40
410
Drilling site
$]
, +
390 o
380
370 -+-~-r-l..r--.l.~r-I-r--f--,..--,--yo--r-""'--~~~~-""--r--r--+-.....-,--..,".atr....,..--r---r-~
280 290 300 310 320 330 340

Bastings (km)
Fig. 3 : Map of Study Area, showing Aeromagnetic Contours

SPE 30452 J. P. RUSSELL, G. SHlELLS, D.J. KERRIDGE 13
Scale 1 : 150.00 East=>

·2250 -1650 ·1050 -450 ·150 450
~~-~~-~~-~~-~---',eooo
50100
o
c
-
ei
tn ·150
150
Jl 20" Conductor
3000
2~00
750
1050 1800
1350 t/)
1650 1200 eJ.

1950 '"
----- u:
600
2250
!='
o
C~
2550
9 5/8" .--- o
2850 ~:: ~ Target Honzon 0
3150 ~Bottom Target Horizon .6QO
3450
-450 ·150 750 1350 1950 2550 3150 3750 4350 4950 5550 6150
Scale 1 : 150.00
Fig 4 : Typical Well Profile, Liverpool Bay Development
100
Program 1
00 Program 1 : gyro top hole, EMS & MWD (BGGM)

Program 2
Program 2 : gyro top hole, EMS & MWD (IIFR)
Program 3 : gyro top hole to TO.
80
Program 3
7ll
~
...8 ro ...
Target Tolerance
. .
E.
~
c
~ 50
B
c
:» ....
~
olD
.... ~, .
.!l
.'l .~._ ~.~.~.~ ~ ~ .
30
................ :
.... ,~.",.,. ~.... ,.
, 1
20 ....................." -
".,
10
......
5lD 1,000 1lm 2,000 2lm 3,000 3lm ~,lXXl 4lm 5,000 5lm 6,000 6lm 7.lXXl 7lm
Depth (tt)
Fig. 5 : Lateral Position Uncertainty for 3 survey Programs

14 APPUCATION OF A NEW GEOMAGNETIC IN-FIELD REFERENCING TECHNIQUE SPE 30452
Geomagnetism Group BGS Edinburgh

Liverpool 8ay Development (Monitored Data)
Decl' ation variations
- ~ . 1 30min
Inclination variations
68.1
Total field variations
48741
00 06 12 18 24
Hour CUT)
Date: 29-10-1994 Day number: 302
Geomagnetism Group BGS Edinburgh

Liverpool Bay Development (Interpolated Data)
-~.1
30min
Inclination variations
68.1
15JH~1l
Total field
48770
1
10 o,,1
00 06 12 18 24
Hour (UT)
Date: 29-10-1994 Day number: 302
Fig, 6 : Comparison of Monitored vs, Interpolated Data from Liverpool Bay during a Disturbed Day.
SPE30452 J. P. RUSSELL. G. SHIELLS, D.J. KERRIDGE 15
0.0 0.0 0.0 18.0 -5.1
.~
.
c:<
" . co
ilc
... 2-
m
c
'<
0
co
:.t <
CO
a
.~
I:
'tl
1" ..a 3
0
n
0'
..cr
co
..'"'"
2000T 0.50 dog 1 dig 2000T 0.50 dig I dIg
0.0 0.0 0.0 .e6e•.9 ee.D -5.9
s:.
co
. 'tl
(J
2-
c m
c
'<
o
co
f I:
g
~
oU;
3. f f I:
! ·f f
a
'tl
i" ~
I I I
c
'<
f
:;"
!2-
!'
~. i I
~
~
.f i .. l;
"! z
.3
c
<
.cr
co
..'"'"
I j'
.; ···1:, ...
2DOoT 0.50 dIg 1 dog 2DOoT 0.50 dog tdog
Fig. 7a: Differences in D, I, and F, monitored vs. Interpolated Data, October - November
D.D D.D D.D oUI88l.8 88.D
. ... ,.
>- ..
'"1'''''''' f
;; I .... [ ....J:. ~ ...
oQ ..I •• 1.. I .. ~F' ... 1 ... <.,..
... :r ... !I"
...,
J i
f
... 8,.
I
....t·
II.
.. ... C
It
n
. "T'''''' II
> .,.
3
~:.:-' ..,
It
.~~..
-
..... ~
....
l-
..,..
II ..
loo nT O.SD dIg 1 dIg lDD nr D;5D dIg
D.D D.D D.D 48870.9 68.0 -5.9
. oS
t .
i=-
l"
.F· ..·
....... ?:' ..
f
i
...o
I
Q
.!
!l.
:I...,
'-
.
''1''
I"
""c::
Q
~
tl ...
Ul
Ul
Ul
.~.
zao nT D.SD dog

'<:,'
ZDD nr
··:r··..· . ·•
D.5D dIg 1 dog
Fig. 7b: Differences in D, I and F, Monitored vs. Interpolated Data, December and January.

Spe 30452 Ms

Uploaded by

Copyright:

Available Formats

Spe 30452 Ms

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Spe 30452 Ms

Uploaded by

Copyright:

Available Formats

SPE 30452

Copyrighl 1995. Soclety .f 1'............. EIIflm-•• 1lIe.

The Geomagnetic Field. momentary bursts of actiVity can be seen on magnetograms

A New IFR Technique. Directional Survey Program.

b) The separation of the observatory from the 'snap Conclusions.

in cost savings. I1FR should also assist in the elimination

2. Russell, M.K., and Russell, A.W., "Swveying of Boreholes"

4. Thorogood, J.L. and Knott, n.R.: "Swveying Techniques

5. de Lange, J.L. et al: "Accurate Swveying; An Operator's

6. Zijsling, D.H. and Wilson, R.A. : "Improved Magnetic

7. Wolff, C,J.M. and de Wardt, J. P.: "Borehole Position

8. Stuart, W. F. and Barraclough, D.R. : "Magnetic Reference in

9. Kerridge, DJ.: "Theory of the Fluxgate-Theodolite." British

Geomagnetic Total Error Crustal Variaticn

Geomagnetic Opticn I Opticn 2 Opticn 3

Table 2: Estimated Total Uncertainties In BOOM Values

Opticn I: Initial map..sbot at Point of Ayr

Month D (del{) I (del{) F (nT)

Table 3 : Monthly Standard Deviations of Hourly

Month D (del{) I (del{) F (nT)

Table 4 : Monthly mean values of monitor measurements.

Month D (deg) I (deg) F (nT)

Table 5 : Monthly standard deviations of (monitor-IFR) data.

D (deg) I (deg) F (nT)

Table 6 : Standard deviation of one-minute values of 0,

Mean Differmce (deg) to gyro SD of difference (deg)

Table 7 : Comparison of difference between magnetic surveys processed

Divergence FtJ 1000 Ft

Table 8 : EMMS vs. MWD Surveys in 12 1/4" hole Sections

Fig. 1 : Frequency of Magnetic Disturbances at LerWck Observatory 1983-1994

Number of days when the range of declination exceeded

60 1" ,& C> 0.5 deg

Number of days when the range of inclination exceeded

Number of days when the range of total field intensity

Liverpool Bay total intensity anomalies

280 290 300 310 320 330 340

Fig. 3 : Map of Study Area, showing Aeromagnetic Contours

Scale 1 : 150.00 East=>

1650 1200 eJ.

3150 ~Bottom Target Horizon .6QO

00 Program 1 : gyro top hole, EMS & MWD (BGGM)

Fig. 5 : Lateral Position Uncertainty for 3 survey Programs

Geomagnetism Group BGS Edinburgh

Total field variations

Geomagnetism Group BGS Edinburgh

0.0 0.0 0.0 18.0 -5.1

2000T 0.50 dog 1 dig 2000T 0.50 dig I dIg

0.0 0.0 0.0 .e6e•.9 ee.D -5.9

2DOoT 0.50 dIg 1 dog 2DOoT 0.50 dog tdog

D.D D.D D.D oUI88l.8 88.D

loo nT O.SD dIg 1 dIg lDD nr D;5D dIg

D.D D.D D.D 48870.9 68.0 -5.9

zao nT D.SD dog

You might also like