Effective Monitoring of Auroral Electrojet Disturbances to Enable Accurate Wellbore Placement in the...

8/11/2019 Effective Monitoring of Auroral Electrojet Disturbances to Enable Accurate Wellbore Placement in the Arctic OTC-24

1/11

OTC 24583

Effective Monitoring of Auroral Electrojet Disturbances to Enable AccurateWellbore Placement in the ArcticBenny Poedjono, SPE, Schlumberger; Stefan Maus, SPE, Magnetic Variation Services; Chandrasekharan Manoj,National Geophysical Data Center

Copyright 2014, Offshore Technology Conference

This paper was prepared for presentation at the Arctic Technology Conference held in Houston, Texas, USA, 10-12 February2014.This paper was selected for presentation by an ATC program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not beenreviewed by the Offshore Technology Conference and are subject to correction by the author(s). The material does not necessarily reflect any position of the Offshore Technology Conference, itsofficers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Offshore Technology Conference is prohibited. Permission toreproduce 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 OTC copyright.

AbstractIn measurement while drilling (MWD), wellbore azimuth is determined relative to the direction of the geomagnetic field.

Converting this magnetic azimuth to a true azimuth requires accurate knowledge of the direction of the geomagnetic field atthe point of measurement downhole. In the Arctic, MWD processing must include corrections for rapid changes in the

geomagnetic field caused by auroral electrojet currents.The auroral zone, those latitudes at which the aurora borealis (or the

northern lights) occurs, is a region where the electric field of the magnetosphere precipitates along magnetic field lines into

the ionosphere. At 100 km above the surface, this electric field drives auroral electrojet currents in the east/west direction,generating the strongest magnetic field disturbances on the planet. The direction of the geomagnetic field in the auroral zone

can change by several degrees in less than an hour.

Data from geomagnetic observatory and variometer stations can be analyzed to characterize the auroral electrojets andcompensate for the disturbance. Knowledge of the spatial structure of the electrojets magnetic signature is essential for

deploying a ground network of monitoring stations in the Arctic. This network provides the real-time geomagnetic

infrastructure essential to support MWD operations, making it the most cost-effective technology available to achieve

accurate wellbore placement in horizontal, relief well, and extended reach drilling, as well as in collision-avoidance

applications.In one case study using historical data from two nearby observatories from 1995 to the present, the disturbance field was

characterized and a time series of maximum disturbances was derived and extrapolated to the year 2020. Maximum

disturbance in the magnetic field was found to lag the maximum of solar activity by approximately two years, predicting the

next maximum in 2015-2019.


2/11

2 OTC 24583

IntroductionDrilling in the Alaskan Arctic poses a number of challenges that demand an advanced approach to wellbore surveying.

Because of both the crowded subterranean environment and the geological complexity of the onshore North Slope of Alaskaand the offshore Beaufort Sea (Fig. 1), precise, real-time wellbore positioning is essential to the success of commercial

development.

Fig. 1Map of the North Slope of Alaska and offshore Beaufort Sea. Sold Federal and State leases shown in orange and active stateleases in shaded orange.

First and foremost, the prevention of accidental intersections with adjacent wellbores is critical because of the associatedhealth, safety, and environmental (HSE) risk, as well as to minimize the consequences of non-HSE risks. By implementing

advanced surveying techniques, ellipses of uncertainty (EOUs) are significantly reduced from conventional measurement-

while-drilling (MWD) surveys. This prevents the overlapping of EOUs and increases the separation factors between adjacentwells, greatly diminishing the probability of a wellbore collision. The complexity of the targeted reservoirs provides another

critical impetus for advanced surveying techniques. It is vital that wellbores are placed with adequate accuracy to ensureproper spacing between injector and producer wells.

To achieve accurate wellbore placement by MWD, a better understanding of geomagnetic disturbance fields is needed. In

this study, we explored the spatial structure of the geomagnetic disturbance field to build an optimal network of real-timemagnetic observatories to support directional drilling in the Arctic.

For the North Sea, a disturbance field correction called interpolated in-field referencing (IIFR) was developed and

published by Williamson et al. (1998, SPE paper 49061). The IIFR technique can be used in the North Sea, which issurrounded by geomagnetic observatories within distances comparable to the length scales of magnetic disturbances there. In

the Arctic, on the other hand, short-scale auroral disturbances and the lack of observatories limit the use of the IIFR

technique. A study for the Norwegian Sea, where one variometer station was used to predict the measurements of anothernearby variometer, was presented by Gjertsen (San Antonio, USA 11-Oct-2012) at the semi-annual meeting of the Industry

Steering Committee for Wellbore Surveying Accuracy (ISCWSA). The study showed that corrections may be possible at

some times and unsuccessful at others. A more comprehensive statistical study using a larger number of stations is therefore

needed.The overriding question in this study was: How close must an observatory be to allow a significant reduction of thedisturbance field at the Arctic drillsite? By analyzing historical data of geomagnetic observatories and variometers from

1995 to 2012, the spatial correlation distances of the disturbance field were inferred for the declination, dip, and total field. It

was further investigated how these distances depend on latitudinal and longitudinal separation.

Why Geomagnetic referencing is neededAccurate wellbore positioning is essential to locate and produce the resources in the Arctic. Unfortunately, the high latitudes

associated with Arctic drilling pose a challenge to standard magnetic surveying techniques. Most notably, the accuracy of

standard MWD surveys can be severely compromised by disturbance fields at high latitudes. The high-inclination limitationand the extensive time requirements of implementation limit the effectiveness of traditional gyroscopic surveys. An accurate

and efficient solution is critical to the success of drilling in the Arctic environments.


3/11

OTC 24583 3

Geomagnetic referencing provides this solution by simultaneously addressing the stringent well-placement requirements

and the challenging surveying environment of Alaskan North Slope operations. Precise real-time positioning is possible by

taking advantage of refinements in the latest development in crustal model processing and improvement in the design ofmagnetic observatories that measure disturbance fields to provide precise, real-time positioning (Fig. 2). The geomagnetic

referencing has smaller EOUs than MWD, as such, the technique has been deemed capable of addressing the challenges to

survey accuracy inherent in high-latitude drilling in the auroral zone. Geomagnetic referencing techniques with correction for

high-disturbance components of Earths magnetic field are particularly important when having to compensate for the effect of

drillstring interference.

Fig. 2The difference between the MWD EOU (in blue) and smaller geomagnetic referencing EOU (in yellow).

Challenges of MWD in the ArcticDirectional drilling requires accurate knowledge of the orientation of the bottomhole assembly (BHA) referenced to vertical

(inclination) and to true north (azimuth). To acquire these critical measurements, wellbore surveying by MWD uses the

direction of Earths gravity and magnetic field as a natural reference frame. Specifically, the hori zontal component of the

geomagnetic field is the key reference when using magnetic north to determine azimuthal orientation of the borehole. Athigher latitudes, the strength of the horizontal component of the geomagnetic field shrinks, which exacerbates any error

sources that accumulate while surveying. This effect has an enormously negative effect on surveying accuracy at high

latitudes.

Based on the smaller horizontal geomagnetic component, there is an increased impact from axial and cross-axialinterference from the drillstring and/or mud effects. BHAs that are magnetically acceptable in lower latitude areas can lead to

significant inaccuracies in the Arctic environment.The geomagnetic field can be divided into three contributions (Fig. 3):

The main field generated by the geodynamo in the Earths core, which is defined, for practical purposes as theinternal field of spherical harmonic degree 1 to 15, excluding time-varying fields with periods shorter than about

2 years

The crustal field caused by magnetic minerals in the crust, which is defined, in practice, as the static internalfield of spherical harmonic degree 16 and higher

Magnetic disturbance fields caused by electric currents in near-Earth space and corresponding mirror-currents induced in the Earth and oceans


4/11

4 OTC 24583

Fig. 3Contribution to the geomagnetic field from the main field (in blue) and crustal field (in green)

If not accounted for, crustal magnetic anomalies have a large impact on survey accuracy due to the systematic error in

nature. The highly active geomagnetic disturbance field in the Arctic further exacerbates the problem. Geomagnetic

referencing allows these challenges to be addressed, thereby improving the accuracy of wellbore positioning while drilling in

the Arctic.

Sources of the magnetic disturbance fieldThe magnetic disturbance field in the Arctic is due to a combination of effects caused by the magnetospheric ring current,auroral electrojets, and secondary induced fields.

Magnetospheric currents

The magnetospheric current systems are fed by charged particles originating in the solar wind. The strongest contribution

is from the ring current, shown in blue in Fig. 4. The ring current increases in strength during magnetic storms, which arecaused by coronal mass ejections from the sun. The field-aligned currents (shown in green in Fig. 4) also have an importanteffect, since they predominantly affect the declination of the magnetic field, leading to errors in the MWD azimuth, if not

corrected for.

Fig. 4Sketch of magnetospheric current systems contributing to the geomagnetic disturbance field in the Arctic.


5/11

OTC 24583 5

Auroral electrojets

The ionosphere is a region from approximately 80 km to 1000 km above the Earth surface. It is much closer to the Earth

than the magnetosphere. Currents in the ionosphere are present even during quiet times and are then caused by tides of theatmosphere. During magnetic storms, a strong electric field is imposed through field-aligned currents (green lines in Fig. 4)

onto the polar ionosphere. This electric field drives strong east/west currents in the auroral region, called auroral electrojets

(Fig. 5). The auroral electrojets cause large magnetic disturbances in the Arctic.

Fig. 5NASA ultraviolet image of the auroral zone in which the electrojets flow (left). The sketch on the right shows the differentcurrents in the ionosphere. Of these, the auroral electrojets (in blue) generate by far the largest magnetic field disturbances in theArctic.

Secondary induced magnetic fields

Finally, any time-varying disturbances in the magnetic field induce electric fields in the Earth and oceans. These electric

fields generate electric currents and secondary magnetic fields. Such induced magnetic fields make up approximately one-

third of the disturbance field. Conductivity in homogeneities within the Earth, as well as the contrast between the solid Earthand oceans, gives rise to complicated spatio-temporal structures of the disturbance field, necessitating real-time

measurements in the vicinity of the drillsite.

Characterization of magnetic disturbance fields in the ArcticWhen setting up a network of observatories to monitor the disturbance field for directional drilling, an important question is

how close the observatories must be to the drillsite and how dense the network is required to be.

Magnetic observatory and variometer station data set

To investigate the spatial correlation of the geomagnetic disturbance field in the Arctic, a statistical study of all available

Intermagnet (http://www.intermagnet.org)magnetic observatories at high latitudes (above 57 geomagnetic latitude) from

1995 to 2012 was undertaken (Fig. 6). Further included in the study were measurements of the Scandinavian IMAGE array(http://www.geo.fmi.fi/image/). There is a scarcity of closely spaced stations, although a few pairs are less than 25 km apart.

For storm-time conditions the closest separation was 87 km.

In the first step, a spline was fitted to each magnetic field component at each observatory and was subtracted from themeasurement (Fig. 7). This procedure was used to remove the main and crustal field contribution, isolating the disturbance

field. The disturbances in Bx, By, and Bz were then transformed into corresponding disturbances of the declination, dip, and

total field.
http://www.intermagnet.org/http://www.intermagnet.org/http://www.intermagnet.org/http://www.geo.fmi.fi/image/http://www.geo.fmi.fi/image/http://www.geo.fmi.fi/image/http://www.geo.fmi.fi/image/http://www.intermagnet.org/


6/11

6 OTC 24583

Fig. 6Total of ninety three Intermagnet magnetic observatories (IMO) and variometer stations (green dots) used in the study.

Fig. 7A spline (in red) was fit and subtracted from each component at each observatory to isolate the disturbance field signal.Shown here is the east component at Irkutsk Observatory.

Correlation distance

The objective was to find out how much of the magnetic disturbance at the drillsite can be removed by subtracting the

disturbance field measured at a nearby observatory. This was investigated by a statistical analysis of all pairs of

observatories. If one considered one of the observatories as a virtual drilling location, then its measurements could simulate

the MWD measurements of magnetic azimuth, dip, and total field. We were therefore interested in the difference between themeasurements at this virtual drillsite and measurements at the other observatory. This difference simulates the commonly

used procedure of using the nearest observatory to correct for the disturbance field at the drillsite. An improvement could be

achieved by using multiple observatories surrounding the drillsite and interpolating between them. This is the IIFR approach

commonly applied in the North Sea. In the Arctic, however, the drillsite is rarely situated between multiple observatories in


7/11

OTC 24583 7

close-enough proximity. No interpolation is then possible, and the IIFR method defaults to simply subtracting the field

measured by the nearest observatory, as simulated here.

For every possible permutation of observatory pairs, the disturbance field at one observatory was subtracted from themeasurements of the other. The difference (residual) is indicative of the remaining disturbance field at the drillsite after

having applied a correction using the nearby observatory. The root mean square residuals (or 1-sigma errors) in the

declination, dip, and total field were then computed and sorted by the distance between the observatory pairs. After fitting a

suitable smoothing function, the result is displayed in Fig. 8 for magnetic storm-time conditions, as defined by a planetary

disturbance index (http://www.ngdc.noaa.gov/stp/geomag/kp_ap.html)of Kp 6. These residual curves can be interpreted asthe 1-sigma disturbance field uncertainty after applying a correction using an observatory at the distance given on the x-axis.

Since the magnetic disturbance field has a long-tailed statistical distribution, peak disturbances during strong magnetic storms

are an order of magnitude larger.A seen in Fig. 8, the declination disturbance is significantly larger than the dip disturbance. This is due to the weakness of

the horizontal field at high latitudes. Consequently, an arbitrary magnetic field disturbance changes the direction of the

magnetic field vector more significantly in the horizontal (declination) than in the vertical (dip) direction.

Fig. 8RMS residual of the difference in the disturbance fields between two observatories as a function of their separation distancefor magnetic storm-time conditions (Kp 6). The closest station pair had a separation of 87 km.

Quantifying the benefit of the disturbance field correction

The ultimate question is how close an observatory must be to the drillsite to allow for a significant reduction in the

disturbance field at the drillsite. In the following, we show that the answer is different for each of the declination, dip, and

total field. Furthermore, it makes a difference whether the observatory is displaced north/south or east/west from the drillsite.

Di ff erences among Declination, Dip, and Total F ield

To allow for a side-by-side comparison of the correction for declination, dip, and total field, the RMS residuals werenormalized by dividing them by the RMS of the uncorrected signal at the drillsite. In other words, one observatory is chosen

as the drillsite, and the other is chosen as the source for the correction. Then the RMS is computed of the difference (residual)and is divided by the RMS of the uncorrected field. A value of 0.0 indicates a complete removal of the magnetic disturbance,

while a value of 1.0 (100%) indicates that the residual is as strong as the uncorrected field. When the corrected residual

exceeds 100%, applying the correction actually makes things worse.Fig. 9 shows the remaining disturbance field error in the total field, dip, and declination as a function of the distance to the

nearest observatory. The correction for the declination holds up somewhat better with distance than the correction for the dip

and total field. Generally speaking, to reduce the disturbance field by 75% requires an observatory within approximately 100

km of the drillsite. Subtracting measurements of an observatory situated more than 600 km from the drillsite introduces an

error which is greater than the uncorrected signal.
http://www.ngdc.noaa.gov/stp/geomag/kp_ap.htmlhttp://www.ngdc.noaa.gov/stp/geomag/kp_ap.htmlhttp://www.ngdc.noaa.gov/stp/geomag/kp_ap.htmlhttp://www.ngdc.noaa.gov/stp/geomag/kp_ap.html


8/11

8 OTC 24583

Fig. 9Remaining error in the total field, dip, and declination after subtracting the disturbance field, plotted against the distance ofthe observatory from the drillsite. A few pairs had separations less than 25 km. A 75% reduction in error requires an observatorywithin about 100 km distance, while an observatory located 600 km apart offers no benefit at all.

Di f ferences between the nor th/south and east/west di rections

Since the auroral electrojets are oriented east/west, one can expect disturbances to have shorter correlation lengths in thenorth/south direction. To investigate this effect, all permutations of pairs of observatories were divided by into a NS group

with a connecting line within 45 of north/south and an EW group within 45 of east/west.

Fig. 10 shows the result for the declination. In this, case, the difference between the two groups is relatively small. Anobservatory displaced to the east or the west offers only a slightly better correction at the drillsite than an observatory to the

north or south.

Fig. 10Remaining error after declination correction. The red line is the average for all directions, while the blue line shows thepairs displaced north/south and the green line shows pairs displaced east/west.

The situation is significantly different for the dip and total field. Fig. 11 shows that using an east/west-displaced

observatory is significantly more effective than using a north/south-displaced one for the dip angle. An observatory 250 km

to the west offers about an equal benefit to one 150 km to the south.


9/11

OTC 24583 9

Fig. 11Remaining error after dip correction. The red line is the average for all directions, while the blue line shows north/southdisplacements and the green line shows east/west displacements.

As shown Fig. 12, the benefit of the correction for the total field also strongly depends on the direction to the observatory.Again, an observatory to the east or west is significantly better than one located to the north or south of the drillsite.

Fig. 12Remaining error total field correction. The red line is the average for all directions, while the blue line shows north/southdisplacements and the green line east/west displacements.

Characterization of the maximum magnetic disturbance field

To characterize the disturbance field, one-minute averaged measurements of the USGS magnetic observatories Barrow

(BRW) and Deadhorse (DED) were used from 1995 to the present. In the first step, a spline was fitted to each magnetic fieldcomponent at each observatory and was subtracted from the measurement. This procedure was used to remove the main and

crustal field contribution, isolating the disturbance field. The residual Bx, By, and Bz values were then transformed into

corresponding residuals of the declination, dip, and total field.

To derive time series of maximum disturbances, data analysis windows of three different lengths were moved over the

residuals of both observatories to determine the running monthly maximum Rm, the running annual maximum Ry, and therunning 4-year maximumR4y. The final maximum was then computed as a weighted average of the maxima as

R = (3 Rm+ 2 Ry+ R4y) / 6

This procedure was repeated separately for the declination, dip, and total field.


10/11

10 OTC 24583

To model the disturbance and predict it outward, a Bezier spline was fitted to each of the time series of the maximum

declination, dip and total field. After the end of 2012, the spline was then extrapolated outward, mirroring the curve for the

previous solar cycle. Fig. 13shows the original time series (blue and red) and the fitted spline (solid black) from 1995 up toits extrapolation to 2020. To provide an envelope of the expected maximum and minimum disturbances, two additional

dashed lines were drawn for 1.8*spline and spline/1.8. Based on Fig. 13, disturbance fields created significant changes in the

declination from 2.5 to 20; Dip from 0.7 to 4; and total field from 300 nT to 1800 nT. Note that the maximum disturbance

in the magnetic field lags the maximum of the solar activity cycle by about 2 years. Therefore, the maximum disturbance in

the magnetic field during the current solar cycle is expected for 2015-2019, while the maximum of the solar sunspot cycle isexpected to occur earlier.

Fig. 13Modulation of the maximum magnetic disturbances in Alaska with the 11 year solar sunspot cycle. The USGS Barrowobservatory is shown in blue and the recently installed USGS Deadhorse observatory is shown in red.

Summary and ConclusionsThis study finds that magnetic disturbances in the Arctic have short correlation lengths. Observatories therefore must be

placed close to the drillsite to offer a significant benefit. Generally, a reduction in disturbance by 75% requires an observatory

within approximately 100 km of the drillsite. Subtracting disturbances measured more than 600 km away offers no benefit

and introduces additional errors that are larger than the uncorrected disturbance field.


11/11

OTC 24583 11

Declination disturbances have larger correlation lengths than dip and total field disturbances. This has important

implications for validation. If a correction is seen to reduce the disturbance field, for example by 50% in MWD

measurements of the dip and total field, then the (invisible) azimuth disturbance is likely to be reduced even more.Correlation lengths are generally shorter north/south than east/west. This anisotropy is particularly pronounced in the dip

and total field. Therefore, latitudinal distance matters more than longitudinal distance between observatory and drillsite. If

possible, observatories should therefore be displaced east/west from the drillsite.

AcknowledgementsThe authors appreciate the permission of Magnetic Variation Services, National Geophysical Data Center of National

Oceanic Atmospheric Administration and Schlumberger for their permission to publish the material contained in this paper.

For the ground magnetometer data we gratefully acknowledge:Intermagnet; USGS, Jeffrey J. Love; Danish

Meteorological Institute; CARISMA, Principal Investigator Ian Mann; CANMOS; the S-RAMP Database, PIs K. Yumoto

and Dr. K. Shiokawa; the SPIDR database; AARI, PI Oleg Troshichev; The MACCS program, PI M. Engebretson;

Geomagnetism Unit of the Geological Survey of Canada; GIMA; MEASURE, UCLA IGPP and Florida Institute ofTechnology; SAMBA, PI Eftyhia Zesta; 210 Chain, PI K. Yumoto; SAMNET, PI Farideh Honary; the institutes who

maintain the IMAGE magnetometer array, PI Eija Tanskanen; PENGUIN; AUTUMN, PI Martin Conners; DTU Space who

operates the Greenland magnetometers; South Pole and McMurdo Magnetometer, PIs Louis J. Lanzarotti and Alan T.Weatherwax; ICESTAR; RAPIDMAG; PENGUIn; British Antarctic Survey; McMac, PI Dr. Peter Chi; BGS, PI Dr. Susan

Macmillan; Pushkov Institute of Terrestrial Magnetism, Ionosphere and Radio Wave Propagation (IZMIRAN); SuperMAG,

PI Jesper W. Gjerloev.

Effective Monitoring of Auroral Electrojet Disturbances to Enable Accurate Wellbore Placement in the...

Documents

Transcript of Effective Monitoring of Auroral Electrojet Disturbances to Enable Accurate Wellbore Placement in the...