Winter 2013/2014

Ice Core Science

Seeking Sweet Spots in Shales

Rotary Sidewall Coring

Autonomous Ocean Monitoring

Oilfield Review

14-OR-0001

Oilfield Review AppSchlumberger Oilfield Review iPad† app for the Newsstand is available free of charge at the Apple† iTunes† App Store.

Oilfield Review communicates advances in finding and producing hydrocarbons to oilfield professionals. The free Oilfield Review Apple iPad app for accessing content is part of the Newsstand and allows access to both new and archived issues. Many articles have been augmented with richer content such as animations and videos, which help explain concepts and theories beyond the capabilities of static images. The app offers access to several years of archived issues in a compact format that retains the high-quality images and content you’ve come to expect from the print version of Oilfield Review.

Download and install the app from the iTunes App Store by searching for “Schlumberger Oilfield Review” from your iPad or scan the QR code below, which will take you directly to the iTunes site.

†Apple, iPad, and iTunes are marks of Apple Inc., registered in the U.S. and other countries.

Announcing the new

Oilfield Review iP

ad app

Ten years ago, when drilling into a shale reservoir, it was unlikely that a drilling engineer would have had much time for a geophysicist, and to be frank, there wasn’t much the geophysicist could have told him that would have helped. But as the commercial success of shale reservoirs has developed through horizontal drilling and multistage hydraulic fracturing, geophysicists have found that, in many cases, they do in fact possess the information that can lead to optimized production.

So, what is the role of a geophysicist? The industry would like geophysicists to produce maps of “commercially pro-ducible hydrocarbons.” Unfortunately we are not there yet. We can produce maps of the subsurface structure and give some indications of rock properties, but these results come from interpretation of incomplete sets of information.

However, when it comes to unconventional reservoirs, the ability to identify and high grade shales is essential to optimizing completions and subsequent production. To this end, seismic data plays a role that typically goes beyond the role it plays in conventional reservoirs. Geophysicists can use seismic data to map parameters that, while not direct measurements of commercially producible hydrocar-bons, can certainly be interpreted to give a qualitative indication of likely subsequent production. These parame-ters are rock properties, in situ stress, presence of natural fractures and reservoir geometry. All of these have a direct impact on the likely effectiveness of hydraulic fracturing.

Geophysicists predict rock properties using inversion of prestack seismic data to estimate and map, among other things, the brittleness of rocks, which is a measure of how easily they break or shatter. Brittleness can vary based on the local environment—for example, with temperature. (Take a bar of chocolate out of the refrigerator and drop it on a hard floor; now leave it out in the sun for 10 minutes before you drop it. The material is the same, but in one instance it is brittle and shatters, and in the other, it doesn’t.) Understanding the spatial variations in brittle-ness in a shale reservoir is valuable because the hydraulic fracture engineer is likely to be far more successful work-ing in an area of brittle shale than in ductile shale.

The azimuthal response of seismic data (the changing response as a function of the direction that sound moves as it passes through the rock) can lead to insights about the in situ stress in the rock related to natural fault stress and can help an engineer understand the likely direction of fracture growth. Azimuthal variations in the seismic data may also give indications of the presence and geometry of natural fractures that are likely to increase intrinsic per-meability of the rock.

Unconventional Reservoir Sweet Spots from Geophysics

1

Seismic data has always been used to help geologists understand reservoir geometry and map natural barriers in the rock, and it serves the same purposes in characterizing unconventional resources.

By using this information, it is possible to gain a better understanding of the sweet spots in a shale, significantly improving the likelihood of success in production of hydro-carbons (see “Seeking the Sweet Spot: Reservoir and Completion Quality in Organic Shales,” page 16).

But the geophysicist isn’t finished yet. In the last five years, it has become common practice to monitor in real time the fracturing of the rocks caused by hydraulic stimu-lation. The challenge is one that the seismic geophysicist is familiar with: recording seismic energy and resolving the source of the sound. Although microseismic monitoring doesn’t yet have the track record of surface seismic map-ping, the technique is rapidly improving and represents an opportunity to close the loop on the interpretation of the surface seismic data. If we make predictions of rock prop-erties and “fracability” from surface seismic data, then stimulate the rock and measure what actually happens, we have the opportunity to expand our understanding of the information within the seismic data; armed with this understanding, geophysicists will indeed get closer to being able to produce maps of commercially producible hydrocarbons.

Dave MonkDirector of GeophysicsApache CorporationHouston, Texas, USA

Dave Monk, who holds a PhD degree in physics from The University of Nottingham in England, is the Director of Geophysics and one of only two Distinguished Advisors at Apache Corporation. Based in Houston, he is responsible for seismic activity including acquisition and processing in Argentina, Australia, Canada, Egypt, the North Sea and the US. He started his career on seismic crews in Nigeria and has subsequently been involved in seismic processing and acquisition in locations worldwide. Author of more than 100 technical papers or articles and a number of patents, Dave received best paper awards from the SEG in 1992 and 2005 as well as one from the Canadian SEG in 2002, and was recipient of the Hagedoorn Award from the European Association of Exploration Geophysics in 1994. Dave received hon-orary membership in the Geophysical Society of Houston and life membership in the SEG and is the immediate Past President of the SEG.

www.slb.com/oilfieldreview

Schlumberger

Oilfield Review

1 Unconventional Reservoir Sweet Spots from Geophysics

Editorial contributed by Dave Monk, Director of Geophysics, Apache Corporation

4 Drilling Through Ice and into the Past

Climatologists, chemists, physicists and engineers have developed drilling units capable of retrieving ice that has been isolated from the rest of the world for more than a million years. Their aim is to discover how and when the Earth’s climate has changed over those millennia.

16 Seeking the Sweet Spot: Reservoir and Completion Quality in Organic Shales

For best results, wells in shale reservoirs target production sweet spots where reservoir quality and completion quality are high. Determining a sweet spot is an integral part of the exploration and development effort. Operators are using results from advanced interpretation techniques of surface seismic data to develop drilling and completion programs.

Executive EditorLisa Stewart

Senior EditorsTony SmithsonMatt VarhaugRick von Flatern

EditorRichard Nolen-Hoeksema

Contributing EditorsKate MantleGinger OppenheimerRana Rottenberg

Design/ProductionHerring DesignMike Messinger

Illustration Chris LockwoodMike MessingerGeorge Stewart

PrintingRR Donnelley—Wetmore PlantCurtis Weeks

Oilfield Review is published quarterly and printed in the USA.

Visit www.slb.com/oilfieldreview for electronic copies of articles in English, Spanish, Chinese and Russian. A free iPad® app is available for download.

© 2014 Schlumberger. All rights reserved. Reproductions without permission are strictly prohibited.

For a comprehensive dictionary of oilfield terms, see the Schlumberger Oilfield Glossary at www.glossary.oilfield.slb.com.

About Oilfield ReviewOilfield Review, a Schlumberger journal, communicates technical advances in finding and producing hydrocarbons to customers, employees and other oilfield professionals. Contributors to articles include industry professionals and experts from around the world; those listed with only geographic location are employees of Schlumberger or its affiliates.

On the cover:

A geologist serves to indicate the scale in this photograph of an organic-rich shale outcrop. Dark circles in an image log (left) positively identify points where a large-volume sidewall coring tool extracted samples from a well drilled in the Marcellus Shale. In a hori-zontal well drilled in a different uncon-ventional reservoir (middle), sweet spots identified from gas shows on the mud log (blue curve) correlate with proximity to strong seismic attribute readings (pink and red clouds that proj-ect out of the page). (Outcrop photo-graph courtesy of Aaron Frodsham.)

2

Winter 2013/2014Volume 25Number 4

ISSN 0923-1730

3

30 Rotary Sidewall Coring—Size Matters

Sidewall core analysis is a cost-effective method of directly determining petrophysical and geophysical rock properties. Until recently, one of the main limitations of sidewall cores has been their small size. A new rotary coring tool provides cores that are large enough for experiments and studies without core sample size limitations.

40 A New Platform for Offshore Exploration and Production

The oil and gas industry has long capitalized on remote sensing platforms for exploration and production. A mobile, remotely controlled sensor platform has been developed to provide persistent coverage over an offshore area. Powered by wave motion and sunlight, it can be fitted with numerous sensors that continuously monitor ocean parameters and collect and transmit real-time data to support offshore exploration and production activities.

Hani Elshahawi Shell Exploration and Production Houston, Texas, USA

Gretchen M. Gillis Aramco Services Company Houston, Texas

Roland Hamp Woodside Energy Ltd. Perth, Australia

Dilip M. Kale ONGC Energy Centre Delhi, India

George King Apache Corporation Houston, Texas

Andrew Lodge Premier Oil plc London, England

Advisory Panel

Editorial correspondenceOilfield Review 5599 San FelipeHouston, TX 77056United States(1) 713-513-1194Fax: (1) 713-513-2057E-mail: editorOilfieldReview@slb.com

SubscriptionsCustomer subscriptions can be obtained through any Schlumberger sales office. Paid subscriptions are available fromOilfield Review ServicesPear Tree Cottage, Kelsall RoadAshton Hayes, Chester CH3 8BHUnited KingdomE-mail: subscriptions@oilfieldreview.com

Distribution inquiriesMatt VarhaugOilfield Review 5599 San FelipeHouston, TX 77056United States(1) 713-513-2634E-mail: DistributionOR@slb.com

51 Contributors

52 Coming in Oilfield Review

53 New Books

54 Defining Directional Drilling: The Art of Controlling Wellbore Trajectory

This is the twelfth in a series of introductory articles describing basic concepts of the E&P industry.

56 Annual Index

4 Oilfield Review

Drilling Through Ice and into the Past

That the Earth’s climate is changing is irrefutable. The future course, rate and ultimate

effects of that change are less clear. Climatologists, glaciologists and engineers are

retrieving ice cores from the Greenland and Antarctic ice sheets and from glaciers in

temperate climes in an effort to learn from the past what the future may hold.

Mary R. AlbertDartmouth CollegeHanover, New Hampshire, USA

Geoffrey HargreavesUS Geological SurveyDenver, Colorado, USA

Oilfield Review Winter 2013/2014: 25, no. 4. Copyright © 2014 Schlumberger.For help in preparation of this article, thanks to Jay Johnson, Ice Drilling Design and Operations group, Madison, Wisconsin, USA; Nature A. McGinn and Julie M. Palais, US National Science Foundation (NSF), US Antarctic Program, Arlington, Virginia, USA; and Mark Twickler, Institute for the Study of Earth, Oceans and Space, University of New Hampshire, Durham, USA. Mary R. Albert acknowledges NSF support through award PLR-1327315.Isopar is a mark of ExxonMobil Corporation.

1. Alley RB: The Two Mile Time Machine: Ice Cores, Abrupt Change, and Our Future. Princeton, New Jersey, USA: Princeton University Press, 2000.

2. Committee on Abrupt Climate Change, National Research Council: Abrupt Climate Change: Inevitable Surprises. Washington, DC: The National Academies Press, 2002.

Lüthi D, Le Floch M, Bereiter B, Blunier T, Barnola J-M, Siegenthaler U, Raynaud D, Jouzel J, Fischer H, Kawamura K and Stocker TF: “High-Resolution Carbon Dioxide Concentration Record 650,000–800,000 Years Before Present,” Nature 453, no. 7193 (May 15, 2008): 379–382.

Brook E: “Paleoclimate: Windows on the Greenhouse,” Nature 453, no. 7193 (May 15, 2008): 291–292.

3. Langway CC Jr: “The History of Early Polar Ice Cores,” Cold Regions Science and Technology 52, no. 2 (January 2008): 101–117.

4. Dansgaard W: “The O18-Abundance in Fresh Water,” Geochimica et Cosmochimica Acta 6, no. 5–6 (December 1954): 241–260.

5. Langway CC Jr: “Willi Dansgaard (1922–2011),” Arctic 64, no. 3 (September 2011): 385–387.

6. Bentley CR, Koci BR, Augustin LJ-M, Bolsey RJ, Green JA, Kyne JD, Lebar DA, Mason WP, Shturmakov AJ, Engelhardt HF, Harrison WD, Hecht MH and Zagorodnov V: “Ice Drilling and Coring,” in Bar-Cohen Y and Zacny K (eds): Drilling in Extreme Environments: Penetration and Sampling on Earth and Other Planets. Dramstadt, Germany: Wiley-VCH (August 2009): 221–308.

> Quelccaya ice cap, 1977. (Photograph courtesy of Lonnie Thompson, The Ohio State University, Columbus, USA.)

Winter 2013/2014 55

Climatologists need to look back hundreds to many thousands of years to learn how the Earth’s climate conditions have changed over those years. Doing so helps them better understand Earth’s climate processes so they can make pre-dictions of what is to come. In areas of ice sheets where snow does not melt but piles up over many hundreds of thousands of years, the resulting kilometers-thick ice forms an archive of clues to past climate.

Although the science of interpreting climate from ice cores is less than 70 years old, climatolo-gists have made some remarkable discoveries.1 For example, ice core science enabled the revela-tion that climate can change abruptly, in less than 10 years, and the realization that the carbon dioxide [CO2] composition of the atmosphere is higher now than it has been in more than 800,000 years.2

The first drills for retrieving ice cores for scien-tific use were designed by the US Army Corps of Engineers in the 1950s. These drills, the design of which originated from concepts associated with geologic drilling, were used to drill a number of intermediate-depth and deep ice cores both in Greenland and Antarctica.3 When the US Army created Camp Century in Greenland in the 1960s, army engineers built a new electromechanical drill for retrieving the first deep, continuous core to bedrock; for that core, Chester Langway, Jr., who was responsible for scientific analysis of the cores, formed an international team that included US, Danish and Swiss scientists to conduct an array of measurements on the core. Ice core sci-ence rapidly evolved in many nations, and even today, international, interdisciplinary endeavors continue to be a hallmark of ice core science.

Danish scientist Willi Dansgaard performed work that led to international collaboration on the analysis of the Camp Century ice core. Dansgaard made a discovery in the 1950s that enables ana-lysts today to decipher the information etched into these ancient records. Dansgaard developed instrumentation that could rapidly measure the seasonal variations in climate conditions over short time intervals by measuring variations of stable oxygen isotope ratios, such as 18O/16O in ice cores. Dansgaard applied this technique to analy-sis of the 1,390 m [4,560 ft] long core recovered at Camp Century in 1966 (right).4 Analysis of other chemical species, by a wide range of scientists from around the world, has since been performed on ice cores to extract such weather and climate information from dust, the results of volcanic activity, snow accumulation rate and, for natural and anthropogenic-related markers, a host of chemical tracers in the ice sheets.5

This article describes the process of drilling through Arctic and Antarctic ice sheets and gla-ciers in tropical climates, the techniques used to retrieve intact ice cores and the way ice cores are stored and analyzed. Case histories include the results of efforts to capture cores from the Eemian interglacial period in Greenland, the West Antarctic Ice Sheet (WAIS) and the Quelccaya ice cap in Peru.

Building an Ice Drilling RigAs scientists sought to acquire cores from greater depths in the thick ice sheets of Greenland and Antarctica, the equipment to do so has evolved to meet the challenges unique to those environments.6 One of the most recent iterations in the development of ice drilling rigs is the electromechanical deep ice sheet coring (DISC) drill. The DISC drill, with its directional

> The first Camp Century ice core. Results from the analysis of the Camp Century ice core showed that past climate conditions could be derived from ice cores. The graph shows the amount, in parts per thousand (0/00), by which the ratio of stable oxygen isotopes 18O/16O (δ18O) varies as a function of depth and age along the 1,390-m length of the ice core. Low δ18O values (blue shading) are associated with low temperatures at the time, and high values (purple shading) are associated with warm temperatures. The large deviation of δ18O values at around 1,100 m corresponds to the change from the last glacial to the current interglacial period. Various past climate events (2 to 5e) are also identified. These results demonstrate that ice core drilling and the oxygen isotope method are viable ways of reconstructing some past climatic conditions. [Adapted from “The History of Danish Ice Core Science,” University of Copenhagen, Centre for Ice and Climate, Niels Bohr Institute, http://www.iceandclimate.nbi.ku.dk/about_centre/history/ (accessed June 5, 2013).]

1930 optimum

Postglacialoptimum

23

4

5a5b5d

5e

5c

LittleIce Age

MedievalWarm Period

1,000

800

600

400

300

200

100

50

30

20

10

10

20

50

100

200

500

1,000

2,000

5,000

10,000

20,000

50,000

100,000

–45 –40 –35 –30

1,100

1,200

1,300

1,330

1,360

Dept

h, m

Age

in y

ears

bef

ore

1968

δ18O, 0/00

6 Oilfield Review

drilling capability, was designed and built by the Ice Drilling Design and Operations (IDDO) group at the Space Science and Engineering Center at the University of Wisconsin-Madison, USA. The drill was developed to retrieve cores in deep ice by incorporating, among other fea-tures, the ability to do the following:• collect ice cores from depths to 4,000 m

[13,000 ft]• capture ice cores of more than 98 mm [3.9 in.]

diameter• maintain 5° or less borehole inclination• collect replicate cores using directional

drilling• sample and record depth, drill rotation speed,

torque, WOB, fluid temperature and core barrel acceleration 10 times per second.7

Designers also sought to reduce overall project duration by optimizing the balance between trip time and coring time; in deeper projects, moving the bit in and out of the hole is a larger contributor

to overall project time than is coring on the bot-tom. The DISC unit is able to drill longer cores than were possible using earlier drills, thus is able to reach its depth objectives in fewer trips.

The DISC drill consists of a drill sonde, drill cable, drill tower, winch, surface power supply and control system. The modular drill sonde includes a cutter head assembly, core barrel, screen section, motor pump section and instru-ment section. The cutter head assembly, which has four replaceable cutters, incorporates a core barrel to protect the captured core (above). The cutter head, which cuts an annular ring of ice to produce the core, includes four core dog cages, or pawls, that break the core at the end of the cor-ing run and keep it from slipping out the bottom of the barrel as it is brought to the surface. The cutter head assembly also includes buttons, or shoes, located on the bottom face of the cutter head. The buttons serve to limit the penetration of the cutters by setting the pitch of the cutters.8

The motor pump section of the DISC sonde contains two motors and a drill fluid pump that can operate in temperatures to –50°C [–58°F] and pressures to 40 MPa [5,800 psi]. One motor drives the pump and the other drives the core barrel and cutter head assembly.9 Ice cuttings are collected in the screen section, which consists of a housing made of the same tubes as the core bar-rel fitted with screens in its center. Drilling fluid carries the ice chips created during coring opera-tions up into the annulus of the core barrel to screens in the assembly, which brings them to the surface along with the core (next page, top).

The drill sonde is composed of antitorque, instrument and motor sections for motor control, data acquisition, power conditioning and commu-nications. The system includes two motors, which operate independently and are controlled through a closed-loop current control system. Because motor torque in these systems is propor-tional to current, torque is controlled by moder-ating power to the motors.

Sensors within the sonde measure the tem-perature of the electronics, drilling fluid, cutter motor, pump motor and motor fluid. Because the instrument section must remain pressure sealed, sensors monitor the pressure between two redun-dant seals and each end cap of the assembly.10

The upper section of the sonde includes the cable’s mechanical, electrical and optical fiber terminations. Rotary joints allow the drill sonde to rotate relative to the cable. The cable physi-cally supports the drill, supplies power to it and enables communication between the sonde and the surface. The DISC drill cable includes a cen-tral king wire, fiber-optic cables, copper wires and an outer wrapping of galvanized steel wires to provide the mechanical strength to lower and raise the sonde (next page, bottom).11

Unlike oil and gas drilling unit arrangements, the axis of the DISC drill drum is parallel to the drill cable as it runs through a spooling device to the tower. The configuration allows the winch to be located at the base of the tower, resulting in a smaller footprint than those in which the cable runs perpendicular off the winch to the tower. Because of the tower-winch configuration, once the drill is on the surface, it must be laid down and the core barrel disconnected from the sonde. Rig workers then lift and rotate the core barrel 180° to allow the core to be pushed from the top of the barrel onto a processing tray. Removed from the core barrel, the cores are usually cut from their original 3.5-m [12-ft] lengths into 1-m [3-ft] lengths; they are then stored in a freezer for transportation to an archival storage and

> Ice core cutter head. The ice core cutter creates an annulus between the ice and the sonde. As the tool moves downward, it captures a continuous column of ice for retrieval to the surface. Using this system, drillers have reached depths of about 3,800 m [12,500 ft] and can retrieve a core 12.2 cm [4.8 in.] in diameter and 4 m [13 ft] in length. The rotating core barrel consists of a series of mechanically connected tubes; the barrel can be fitted with a fiberglass sleeve, which helps keep fractured cores intact. Core dogs, which pivot into and break the ice when the drill is lifted, hold the core in the core stabilizer and cage (photograph) as the cutter is brought to the surface. Core shoes are small buttons on the bottom face of the cutter head that limit the penetration of the cutter blades. The vertical distance between the bottom surface of the shoes and the cutter tips sets the pitch, or rate of penetration, of the drill. (Adapted from Mason et al, reference 8.)

Core dog

Cutter

Core barrel segment

Core cage and stabilizer

Sleeve

Ice core

Shoe

Threaded connectionto screen section

Winter 2013/2014 7

research facility. The DISC drill was field tested in Greenland and the designers implemented necessary modifications prior to the drill team’s work in Antarctica.12

Similar to those used in the oil field, drilling fluids in ice coring serve multiple functions; in addition to lifting ice chips to the screens, drill-ing fluids used in ice drilling create a hydrostatic pressure that prevents the borehole from collaps-ing. Ice boreholes are not geopressured, but ice is plastic and will flow into the borehole in response to vertical and shear stresses imposed on well-bore walls. Vertical stress, or glaciostatic pres-sure, is caused by the overburden weight of the ice; shear stress, or glaciodynamic stress, is caused by glacier flow over rock.13

The density of ice core drilling fluids is designed to be as close as possible to the density of the ice being drilled; in the past, drillers used n-butyl acetate as drilling fluid. But driven by health concerns for personnel, project leaders at the WAIS Divide site in central West Antarctica opted for a mixture of about three parts Isopar K fluid to one part hydrochlorofluorocarbon. Fluid handling systems for ice coring contain a tank with measuring devices, valves, pumps and cen-trifuges to recover ice chips from the screens before returning the fluid to the system.

Development of new ice drilling technology in the US is driven by the Long Range Science Plan, which was the result of scientific community planning organized by the Ice Drilling Program Office (IDPO).14 The IDPO oversees engineering

7. Shturmakov AJ, Lebar DA, Mason WP and Bentley CR: “A New 122 mm Electromechanical Drill for Deep Ice-Sheet Coring (DISC): 1. Design Concepts,” Annals of Glaciology 47, no. 1 (2007): 28–34.

8. Mason WP, Shturmakov AJ, Johnson JA and Haman S: “A New 122 mm Electromechanical Drill for Deep Ice-Sheet Coring (DISC): 2. Mechanical Design,” Annals of Glaciology 47, no. 1 (2007): 35–40.

9. Mason et al, reference 8.10. Mortenson NB, Sendelbach PJ and Shturmakov AJ:

“A New 122 mm Electromechanical Drill for Deep Ice-Sheet Coring (DISC): 3. Control, Electrical and Electronics Design,” Annals of Glaciology 47, no. 1 (2007): 41–50.

11. Shturmakov AJ and Sendelbach PJ: “A New 122 mm Electromechanical Drill for Deep Ice-Sheet Coring (DISC): 4. Drill Cable,” Annals of Glaciology 47, no. 1 (2007): 51–53.

12. Johnson JA, Mason WP, Shturmakov AJ, Haman ST, Sendelbach PJ, Mortensen NB, Augustin LJ and Dahnert KR: “A New 122 mm Electromechanical Drill for Deep Ice-Sheet Coring (DISC): 5. Experience During Greenland Field Testing,” Annals of Glaciology 47, no. 1 (2007): 54–60.

13. Aber JS, Croot DG and Fenton MM: Glaciotectonic Landforms and Structures. Amsterdam: Springer Netherlands (1989): 155–168.

14. The Long Range Science Plan was created by the US National Science Foundation to set goals and offer direction and logistical support for US ice coring and drilling science and to support ice drilling technology development and infrastructure.

> Screen section. The screen section filters ice chips produced by the cutters from the drilling fluid as it is circulated through the drill. The section also provides a compartment in which to collect and store the ice chips for transport to the surface. The ice chip screen is designed for maximum filter area and minimum pressure drop. A modular, interchangeable screen cartridge was developed for the DISC drill for speed and ease of cleaning during drilling operations. The DISC screen and barrel design are modular so that any number of screen cartridges can be used. Check valves control the direction of drilling fluid flow. The check valve assembly is connected to the screen section below the screens and held in place by a spring-loaded locking ring. The check valve assembly supports the weight of the screen cartridges that are above it and employs a set of double door check valves to allow the fluid-chip slurry that is pumped up from the cutters to enter the inside of the screen cartridge stack, where chips are filtered and collected from the drilling fluid; the filtered fluid is discharged into the wellbore. A concentric array of 12 openings allows one-way backflow of clear fluid to drain and bypass the screens in the opposite direction down through the drill as the tool is tripping out of the borehole. (Adapted from Mason et al, reference 8.)

Connection to cutter head

Connection to cable

Screen barrel segment

Check valve assembly

Fluid discharge ports

Screen cartridge

Double door check valve

> DISC drill cable. The DISC drill cable is designed primarily for coring conditions at the WAIS site and engineered for requirements for weight, size and breaking strength. The cable is sized to fit on the winch spool that lowers the device into the core hole. The spool and cable are light enough to be handled by available cranes and shipping methods. The breaking strength is specified at 142 kN [31,900 lbf], which is greater than that of the mechanical fuse at the top of the sonde and less than the total winch pulling force. The cable is designed with void filler material around cable parts and outer layers that are impervious to drilling fluids used at the WAIS site. The cable has an operational life of five years. (Adapted from Shturmakov and Sendelbach, reference 11.)

King wire

Six optical fibers

Nylon buffer

Eight copper-clad steel wires

High-density polyethylene belt

High-density polyethylene belt

22 galvanized improved plow steel strength-member wires

36 galvanized improved plow steel strength-member wires

60 copper wires boundwith aluminum polymer tape

8 Oilfield Review

> Drilling a replicate core. The replicate drilling sonde (bottom) is a modified ice drilling and coring sonde with reduced diameter core barrel and screen sections and a lower actuator module that applies pressure against the borehole wall to initiate a sidetrack wellbore from the uphill, or high, side of the parent wellbore (top). Lower actuators are fitted with disk wheels (not shown) to reduce friction along the borehole wall. Upper actuators keep the sonde from spinning while the core is being cut by preventing transfer of torque to the sonde. [Adapted from Souney J: “Replicate Ice Coring System,” In-Depth 6, no. 2 (Fall 2011): 7.]

Antitorque section

Lower actuators

Instrument sectionPump and motor section

Cutter

Reduced diametercore barrel

Upper actuators

Vertical

Vertical

Sidetrackwellbore

Increased diameterfrom reaming

Uphill sideof borehole

30-m maximum

Parentborehole

Maximum20°

performed by IDDO.15 Scientists in the ice core community have long wished for replicate cores from scientifically significant specific depths in ice sheets such as those at which abrupt climate changes have occurred. A replicate core is a core from a sidetrack wellbore that has been drilled nearly parallel to and very near a previously retrieved core so that the two will have exact depth and layer matches. In the last five years, engineers at IDDO have found a way to realize this possibility. In 2012, engineering adaptations to the DISC drill helped scientists recover repli-cate ice cores from multiple targeted depths at the WAIS drillsite. Because scientists wished to continue to deploy gravity-driven sensors in the

borehole below the depth at which the replicate core was taken, the replicate core had to be taken from the uphill, or high, side of the hole.

To meet this requirement, the replicate cor-ing technique uses actuators placed along the sonde that apply pressure to the sidewall of the main wellbore, which causes the wellpath to deviate (left). The deviated section becomes a separate borehole within 30 m [100 ft] of the point at which lateral forces are first applied. The sidetrack exits from the high side of a slightly deviated main wellbore. Once the sidetrack is established, cores are taken from a borehole that is drilled nearly parallel to the main wellbore. The replicate coring sonde includes actuators on its upper end that act as antitorque devices to keep the sonde from spinning; on the lower end of the tool, actuators with disk wheels allow the sonde to move smoothly along the deviated sec-tion. Coring is performed in repeated trips, each of which can capture a 10.8-cm [4.25-in.] diame-ter core, until the desired length of core section is acquired.16

Preparing the TakeAlthough the practices involved in drilling and coring ice may be comparable to those used in oilfield operations, rock and ice in place behave differently. Unlike rock, ice is plastic and flows downward and laterally (below left). Therefore to ensure true depth correlation of strata, drill-ers must site their equipment at the top of an ice structure, or dome. Additionally, ice compo-sition differs depending on burial depth, and as a result, the ice must be handled accordingly. Glaciers are composed of ice containing chemi-cal impurities and air bubbles. From about 600 to 1,200 m [1,970 to 3,940 ft], the cored ice is usually brittle when it is extracted from the borehole. Because the pressure of the air bub-bles trapped in the ice is greater than the bond between the ice crystals, the ice core may spon-taneously fracture and sometimes shatter. Deeper than about 1,200 m, the pressure and temperature of the ice force the air bubbles into clathrate hydrates, making them part of the ice crystal structure, and ice instability ceases to be an issue.17

Technicians electronically measure the length of cores brought to the surface and feed the measurements into a computer program so they may be tallied with measurements from pre-vious cores brought up from the same wellbore. They then evacuate drilling fluid from around the core as it is pulled from the core barrel. Residual drilling fluid is then removed from the core in a

> Ice flow. Because ice is plastic, it flows downward and outward (blue arrows) from the summit of a dome. Therefore, ice cores taken from the center of a dome (horizontal black lines) retain a true depth-age correlation. The black lines represent layers that become thinner with depth as they are compressed by increasing overburden weight.

Summit

Ice

Bedrock

Winter 2013/2014 9

15. Albert M, Twickler M and Bentley C: “A New Paradigm for Ice Core Drilling,” Eos Transactions, American Geophysical Union 91, no. 39 (September 28, 2010): 345–346.

16. “Replicate Ice Coring System,” US Ice Drilling Program, http://www.icedrill.org/equipment/replicate-coring-system.shtml (accessed July 6, 2013).

17. Clathrate hydrates are solids in which molecules, of air in this case, occupy cages in molecular crystals of hydrogen-bonded water molecules.

drying booth, and the cores are then bagged, boxed and shipped. Brittle ice is captured in net-ting to minimize breakage, and when it does break into many pieces, scientists can still dis-cern a great deal from it as long as the mass of the core is preserved in stratigraphic order.

Many of the cores from the major ice sheets of Greenland and Antarctica are now shipped to the US National Ice Core Laboratory (NICL) in Denver. Managed by the US Geological Survey and funded by the US National Science Foundation (NSF), the NICL stores more than 17 km [11 mi] of ice cores from 34 drillsites at a storage temperature of –36°C [–33°F] (right).

The NICL area for examining cores is main-tained at –24°C [–11°F]. The ice core to be ana-lyzed is cut lengthwise, or slabbed. Slabs are further divided into sections to be distributed to scientists for various types of studies (below right). For example, because sections cut from the center of the core are least likely to have been contaminated by drilling fluids or other out-side materials during capture, transportation and storage, scientists at the laboratory typically designate those sections for chemical analysis. Other sections are cut to size specifications or from certain locations within the core for bubble and layer counting, imaging or gas analysis using mass spectrometers.

Paleoclimatologists seeking information about past climates use proxy data gleaned from natural resources such as tree rings and ocean bottom sediment. The records they construct from these sources are paleoproxy records—indirect natural records of past climate or meteo-rologic variability. From the isotopic and chemical composition of ice and dust in ice cores, scientists are able to estimate past regional aver-age air temperatures, atmospheric circulation variations, precipitation amounts, atmospheric composition, solar activity and volcanic erup-tions. Proxy data include a variety of chemical species, stable isotopes, radioisotopes, dust com-position, snow accumulation rate, volcanic ash and sulfur, which scientists use to determine past climate conditions.

> Cores in storage. The National Ice Core Laboratory in Denver serves as a center for preparation and storage of ice cores. The laboratory currently contains more than 17 km of ice cores from around the world.

> Dividing the work. In the laboratory, technicians section the core for specific types of analysis. In this instance, sections DD17 and DD18 were used to determine stable isotopes (H and O) in water. Thin section DDVTS was used for crystal and fabric analysis for size, shape and axes orientation of ice crystals; these 10-cm [4-in.] sample sections are taken every 20 m [65 ft]. Sections DD02 and DD06 were used for beryllium-10 isotope analysis. Sections DD03, DD04 and DD05 were designated for chemical analysis. DD07 and DD09 were archived. DD08 was used for gas analysis, with samples taken every 10 to 50 cm [4 to 20 in.] depending on climate signature and time interval. Kerf is the width of the cut, which is dictated by the width of the saw blade and represents how much material is sacrificed during sectioning.

DDO3

DD

O2

DD

O6

DDO7 DDO8 DDO9

DDO4

WAIS 2011 Cut PlanKerf (blue line) = 2 mm

DDO5

Thin secton (DDVTS)

3 cm × 3 cm

1.2

cm ×

2.8

cm

1.2

cm ×

2.8

cm

3 cm × 3 cm 3 cm × 3 cm

DD17 DD18

10 Oilfield Review

Evidence in the IceSince Dansgaard’s work in the 1950s, use of radioisotopic ratios—primarily hydrogen 2 [δ2H], or deuterium [δD], and oxygen 18 [δ18O]—has further developed, and the ratios are common ice core proxies.18 Isotopes are atoms of the same element with the same number of protons but an unequal number of neutrons. As with all oxygen atoms, the 18O isotope has 8 protons. However, rather than the 8 neutrons of stable oxygen 16 [16O], which makes up about 99.8% of all oxygen

atoms, 18O has 10 neutrons. Because 18O is heavier than 16O, water molecules made up of hydrogen and 16O [1H2

16O] evaporate more read-ily than do molecules containing 18O [1H2

18O]. The resulting vapor contains a high ratio of light-to-heavy water molecules. As an air mass cools, the heavier molecules condense more readily and fall from the clouds as snow and rain. Thus the oxygen isotopic ratio of rain and snow is strongly related to condensation temperature. If the temperature of the air continues to fall, the

condensation will contain decreasing concentra-tions of heavy molecules, resulting in depletion of 18O relative to precipitation that had previ-ously condensed in a warmer environment. As a consequence, past warming and cooling trends have had a large influence on the heavy-to-light oxygen isotope ratio (18O/16O, or δ18O) records within the ice core.19

Scientists also take into account other factors that might affect proxy values, and understand-ing relationships between proxies and climate factors is important to core analysis. Scientists, who gather data about past temperature, mois-ture source regions and hydrology from stable isotopes present in the snow and ice, have recently been tracking trace elements in the ice to assess the past and current contributions from anthropogenic and volcanic sources.20

Chemicals and dust found in ice cores also provide proxies that signal past atmospheric cir-culation, volcanic eruptions, wind speed and tro-pospheric turbidity. Evidence of a volcanic eruption in the form of ash layers and sulfate detected through chemical analysis and other tests can help scientists set dates of ice core lay-ers.21 Ion concentrations of certain chemicals in the ice reveal changes in atmospheric conditions and the causes driving those changes.22

Scientists interpret dust layers in ice cores to infer changes in climate and wind in the area near where the core is captured. Dust layers also help scientists mark instances of atmospheric turbidity; they then use this information to assign a date to the core. Dust concentration correlates well with δ18O composition in glacial ice. Scientists have learned to interpret the value of δ18O in glacial ice and in planktonic foraminifera in sea sediments as a measure of the amount of the Earth’s water that is frozen in ice; plotting these data reveals the occurrence and duration of ice ages.23 Paleoclimatologists use the oxygen isotope–to-dust concentration correlation to bet-ter understand the causes of ice ages by studying dust in ice that was buried deep enough to docu-ment climate variations in years before, during and after numerous past ice ages.24

Analysis of ions and trace elements has typi-cally required technicians to progressively remove the potentially contaminated outer portion of the core under extremely clean conditions. This method has served researchers well but provides low resolution of 10 to 20 cm [4 to 8 in.] per sam-ple, and because this process is labor intensive and time-consuming, datasets are often discon-tinuous. Scientists streamlined the process through the development of continuous ice core

> Ice formation. Recently fallen snow layers are 70% air by volume but are compacted under succeeding layers of snow. Beneath the annual snowfall, which may range from 1 to 200 cm [0.4 to 80 in.] per year, the snow becomes firn, which resembles granular ice with interstitial air decreasing from about 60% to 10% with depth. Deeper than about 60 to 120 m [195 to 390 ft], the firn becomes glacial ice, with air remaining as bubbles within the ice matrix. As the burial process continues, bubble volume is further reduced and the ice becomes clear.

Snow70% air

Firn60% air

Firn10% air

Glacial ice2% as

air bubbles

Winter 2013/2014 11

melting systems that reduced sample preparation time and increased sample resolution while pro-viding continuous and coregistered data for a large suite of elements. These systems use inline continuous flow analysis (CFA) techniques or couple the melter to an ion chromatograph and inductively coupled plasma and field mass spec-trometers. These innovations provided continu-ous measurements of isotopes in meltwater and in air trapped within ice core bubbles.

While the chemical and isotopic analysis of the ice matrix yields proxy evidence of past environmental conditions, ancient air trapped in bubbles within the ice provides the only direct samples of past atmospheres. Bubble for-mation results from the process of snow deposi-tion, compaction and transition to ice at depth. In the extremely cold locations of Greenland and Antarctica where snow melt is rare, snow-fall progressively piles up over many thousands of years, creating kilometers-thick ice sheets. As the snow continues to accumulate on the surface, the increasing overburden compresses the underlying snow. Snow that is more than one year old and still porous is called firn (pre-vious page). With depth, the pore spaces between crystals in the firn become com-pressed. At a depth of 60 to 120 m [195 to 390 ft], the remaining pore space exists as bubbles in the matrix, which has become solid ice; this is known as close-off depth. Because air in the pore space can diffuse through the firn, the air trapped in the bubbles is younger than the ice in which it is enclosed.

A combination of in situ firn air gas measure-ments, measurements of gases in the bubbles in the ice, glaciological measurements and model-ing is used to determine the difference between the age of the gas and that of the ice at pore close-off at a given site. Below pore close-off depth, the gases age at the same rate as the ice in which they are trapped. Measurement of the gas composition with depth of the core allows scien-tists to determine changes in past atmospheric

composition for various gases, including changes in methane [CH4] and carbon dioxide [CO2] lev-els.25 The air trapped in the bubbles deep in the polar ice sheets provides the only opportunity for direct measurements of the chemical composi-tion of the ancient atmosphere.

Getting the Dates Right In addition to correcting for the age difference between the air trapped in ice and the ice itself, scientists face the task of depth-age correlation in ice cores. They accomplish this by comparing profiles of chemical species that exhibit seasonal variation at the time of deposition and gases of known past atmospheric composition, correlat-ing core depth to the depth of volcanic deposition for known eruptions and, for some locations, by visually counting layers. Visual stratigraphy, which relies on differences in brightness, tex-ture, air bubbles and color between core layers, is a direct means of correlating depth and age at sites with high snow accumulation rates where melt has not occurred and where dust serves to enable layer identification (right).

Because counting layers visually is not always possible, scientists most often use dating meth-ods that compare chemical variations and gas composition profiles. In 2003, scientists dated 50 m [165 ft] of ice at Siple Dome, Antarctica, by sending a camera with an LED down the well-bore. Image brightness was filtered digitally. The results of the depth-age relationship derived using digital imaging were close to those derived by manually counting layers and by electrical conductivity measurement (ECM) in a core from a nearby location.26

Direct current ECM, one of two methods ana-lysts use in the process of electrical stratigraphy, measures the low-frequency conductivity of cores. In this process, laboratory workers pull two electrodes of relatively high potential differ-ence along the surface of a prepared slab and measure the current flowing through the core. The measurements are digitized at every millime-

> Visual stratigraphy. Annual layers are clearly visible in this ice core sample. Summer layers (arrows) appear lighter because they contain less dust. (Photograph courtesy of the University of Colorado Boulder, USA.)

ter along the length of the core; these data are stored along with other information such as depth, time of recovery, ice temperature and locations of breaks and fractures. Because the ECM conductivity measurement is a reflection of the acidity of the ice, it is a direct indicator of the volcanic activity influence on the chemistry of the core. Scientists interpret ECM measurements to reveal a stratigraphy of volcanic eruptions,

18. δD = {[(2H/1H)sample – (2H/1H)VSMOW] (2H/1H)VSMOW} × 1000, where (2H/1H)sample is the ratio of deuterium to ordinary hydrogen in a sample corresponding to a particular datum, and (2H/1H)VSMOW is the ratio of deuterium to ordinary hydrogen in Vienna Standard Mean Ocean Water (VSMOW).

19. In the 1960s, the Vienna Standard Mean Ocean Water was developed for the isotopic composition of freshwater. Scientists studying ice cores use the standard to estimate the temperature of condensation at the time the snow fell.

20. Osterberg EC, Handley MJ, Sneed SB, Mayewski PA and Kreutz KJ: “Continuous Ice Core Melter System with Discrete Sampling for Major Ion, Trace Element, and Stable Isotope Analyses,” Environmental Science & Technology 40, no. 10 (May 2006): 3355–3361.

21. Scientists have commonly used chemical testing to detect sulfate in ice cores and to detect preindustrial volcanic activity. However, because rising volumes of anthropogenic sulfates create background signals that obscure the chemical signal from natural sources, the technique is less accurate for post-Industrial Revolution samples.

22. Osterberg et al, reference 20.23. Planktonic foraminifera are single-celled shelled

animals that live on the surface of the ocean. When they die their shells fall to the seabed. Depending on their species, planktonic foraminifera, which can be differentiated by their shells, flourish in various ocean waters, from the warmer surface to the colder depths. Therefore, scientists can use the planktonic foraminifera

remains found in the strata of ocean floor to infer the ocean’s temperature at the time a sediment layer was laid down.

24. Miocinovic P, Price PB and Bay RC: “Rapid Optical Method for Logging Dust Concentration Versus Depth in Glacial Ice,” Applied Optics 40, no. 15 (May 20, 2001): 2515–2521.

25. Bender M, Sowers T and Brook E: “Gases in Ice Cores,” Proceedings of the National Academy of Sciences of the United States of America 94, no. 16 (August 5, 1997): 8343–8349.

26. Hawley RL, Waddington ED, Alley RB and Taylor KC: “Annual Layers in Polar Firn Detected by Borehole Optical Stratigraphy,” Geophysical Research Letters 30, no. 15 (August 2003): HLS1-1–HLS1-3.

12 Oilfield Review

which can be used to date ice cores. They use these findings, along with chemical dating, to establish depth-age relationships (above).27 Scientists also use these measurements to deter-mine depth correlations between cores; these correlations may be used to determine or clarify annual layers that are difficult to discern because of droughts.28

A second method for electrical stratigra-phy—dielectric profiling (DEP)—employs high-frequency, alternating current to measure ice conductivity. Dieletric profiling conductivity is an indication of the amount of acid present in the ice, but unlike the ECM method, the DEP mea-surement may be influenced by chemicals such as ammonium and chloride. In the DEP method,

whole ice cores are placed between curved elec-trodes, lending the method several advantages over ECM. DEP conductivity tests may be per-formed without touching the ice core and without removing the core from the plastic shipping sheath, which makes the method particularly useful on unstable, brittle core sections.

Ice core depth-age correlation is also affected by ice flow and base rock deformation. Ice flow around basal deformities can cause melting, fold-ing and other ice sheet deformations. These events can affect how scientists interpret dates and in some cases can destroy the physical record.

Because of these and other difficulties, scien-tists must sometimes indirectly establish a depth-age relationship of a core. In 1968, the Byrd ice core in Antarctica was drilled to bed-rock. But because the top 88 m [290 ft] of the core were damaged or missing, scientists could not establish a correlation by counting layers. Chronology was established instead by first iden-tifying the horizon at 97.8 m [321 ft] below the surface as a layer created by volcanic activity known to have occurred in 1259 CE. Mean annual accumulation at the Byrd site was 1.12 cm [0.44 in.] per year for the 709-year period prior to 1968.29 The timescale for the remainder of the core was established using ECM. Because mea-surements were sparse in the brittle zone from 300 to 800 m [980 to 2,600 ft], the measurements were fitted with linear functions and the depth-age relationship obtained by integrating the layer-thickness profile from surface to depth. The timescale for older sections of the core was sub-sequently adjusted by correlating measurements of methane concentration in the Byrd ice core with those in layer-counted chronologies from Greenland ice cores.30

Researchers may also extend depth-age rela-tionships established from chemical and visual studies from ice cores over larger, adjacent geo-graphic areas by applying ice-penetrating radar that uses time domain electromagnetic pulses. Radar reflections received at the antennae are caused primarily by conductivity contrasts in the ice that indicate distinct snowfalls (left). By extrapolating radar-determined isochrones from a dated ice core to the geographic area of inter-est, scientists can determine the lateral extent of key stratigraphic layers in places that are distant from the ice coring site.

Looking Back at the FutureMany climatologists consider capturing an ice core from the last interglacial period—the Eemian, which lasted from 130,000 to 115,000 years ago—

> Ice-penetrating radar. This 150 km [95 mi] long section of radar data collected around the North Greenland Ice-Core Project (NGRIP) drillsite in Greenland shows fairly flat bedrock (dark line at bottom) at a depth of about 3 km [2 mi] and undulating ice layers. The shape of these layers is created by variations in basal melt rates. Where the layers dip down, the basal melt rate is highest. (Photograph used with permission from the Center for Remote Sensing of Ice Sheets, University of Kansas, Lawrence, USA.)

Bedrock

Electrical stratigraphy graph. Using the Greenland Crete and Camp Century ice cores, technicians calculated the volcanic stratigraphy of the last 10,000 years from the size of the direct current electrical conductivity method (ECM) peaks (red lines). ECM responds to the acidity of ice, which varies with acidic input from volcanic activity. Scientists can date the ice by matching the dates of these peaks with those of known volcanic eruptions. (Adapted from Wolff, reference 27.)

500

400

300

200

100

–8,000 –6,000 –4,000 –2,000 2,00000

Volc

anic

aci

d fa

llout

, kg/

m2

Year, CE

Winter 2013/2014 13

to be crucial to understanding the Earth’s current climate warming trend. The Eemian interglacial period was a warm period similar to that which the Earth may be trending toward today. Although the warming then was not caused by anthropogenic emissions, results of the warming on the glaciers and ice sheets do provide clues to climate pro-cesses and may help scientists improve predic-tions about the future. For example, some climate models suggest the Greenland ice sheet will disap-pear if today’s apparent warming trend continues; proxy records from the Eemian interglacial period provide a test of that hypothesis.31

Until recently, efforts to extract an ice core with the complete record from the Eemian period have been unsuccessful. That hurdle was recently overcome at the North Greenland Eemian Ice Drilling (NEEM) site when scien-tists extracted a 2,540-m [8,330-ft] ice core. The NEEM project, an international collaboration led by investigators at the Niels Bohr Institute, University of Copenhagen, Denmark, took from 2008 to 2012 to acquire the core. The top 1,419 m [4,656 ft] are from the current Holocene inter-glacial period. The glacial ice below that can be matched to the glacial ice of the North Greenland Ice-Core Project (NGRIP). Below NGRIP depth, the Greenland Ice Core Chronology 2005 extended timescale can be used to a depth of 2,206.7 m [7,239.8 ft], which correlates to 108,000 years before present, assuming 1950 as present.32

Deeper than 2,206.7 m, annual layering in the Eemian ice core becomes more difficult to dis-cern because the ice near the bottom of the ice sheet is folded. This lower section does, however, contain zones with relatively high stable isotope values of H2O [δ18Oice], which, as a proxy for con-densation temperature, indicates the ice is from the Eemian interglacial period (right). This con-clusion is supported by the fact that ice deeper than 2,537 m [8,323 ft] is near bedrock and has low δ18O values, which indicate it is from the glacial period that preceded the Eemian.33

The NEEM ice core is the first ice core record from the entire Eemian period. Scientists will continue their studies in an attempt to further decode the folded ice; however, it is clear that Greenland during the Eemian was about 8°C [14°F] warmer than it is today. From the analysis of the core, scientists have concluded that melt-ing occurred at the edge of the ice sheet and the flow of the entire ice mass caused the ice sheet to lose mass and become reduced in height. Although the ice sheet was shrinking at a rate of

> Observed NEEM records. The observed records of isotopes δ18Oice, δ18Oatmosphere and δ15N along with traces of CH4 and N2O in parts per billion volume (ppbv) and air content from 2,162 m [7,093 ft] and deeper are plotted here on the NEEM depth scale. Each zone (0 to 6) represents a section of the NEEM ice core record. Symbols mark the start (diamond) and end (square) of each zone. There is no discontinuity between Zones 4 and 5, but spikes of CH4, N2O and air content occur in Zone 4 (shaded blue), which indicate a period of surface melting or wet surface conditions. For comparison, the NGRIP data are plotted as light grey curves on the NGRIP depth scale on top of the plot. The NEEM and NGRIP depth scales are synchronized between the NEEM depths of 2,162 and 2,207.6 m [7,093 and 7,242.8 ft]. (Adapted from NEEM Community members, reference 32.)

δ15 N

, 0 /00

δ18 O

atm

, 0 /00

N2O

, ppb

v

CH4,

ppbv

Air c

onte

nt, m

L/kg

δ18Oice

δ18Oatm

δ15N

CH4

N2O

Air content

NEEM depth, m

NGRIP depth, m

δ18 O

ice,

0 /00

–32

–36

–40

–44

70

80

90

100

2,200 2,300 2,400 2,5002,250 2,350 2,450

0.2

0 1 2 3 4 5 6

0.3

0.4

0.5

0.5

–0.5

1

0

400

350

300

250

2,000

1,000

800

600

400

2,800 3,000 3,200

27. Wolff E: “Electrical Stratigraphy of Polar Ice Cores: Principles, Methods, and Findings,” in Hondoh T (ed): Physics of Ice Core Records. Sapporo, Japan: Hokkaido University Press (2000): 155–171.

28. Taylor K, Alley R, Fiacco J, Grootes P, Lamorey G, Mayewski P and Spencer MJ: “Ice-Core Dating and Chemistry by Direct-Current Electrical Conductivity,” Journal of Glaciology 38, no. 130 (1992): 325–332.

29. Langway CC Jr, Clausen HB and Hammer CU: “An Inter-Hemispheric Volcanic Time-Marker in Ice Cores from Greenland and Antarctica,” Annals of Glaciology 10 (1988): 102–108.

30. Neumann TA, Conway H, Price SF, Waddington ED, Catania GA and Morse DL: “Holocene Accumulation and Ice Sheet Dynamics in Central West Antarctica,” Journal of Geophysical Research: Earth Surface 113, no. F2 (June 2008): F02018-1–F02018-9.

31. Wilhelms F, Schwander J, Mason B, Augustin L, Azuma N, Hansen SB, Fitzpatrick J and Talalay PG: “Ice Core Drilling Technical Challenges,” International Partnerships in Ice Core Sciences white paper, http://www.isogklima.nbi.ku.dk/nyhedsfolder/engelske_nyheder/centre-people-to-antarctic-2013/IPICS_Technical_Challenges.pdf/ (accessed January 15, 2014).

32. NEEM Community members: “Eemian Interglacial Reconstructed from a Greenland Folded Ice Core,” Nature 493, no. 7433 (January 24, 2013): 489–494.

33. NEEM Community members, reference 32.

14 Oilfield Review

about 6 cm [2.5 in.] per year, it did not disappear, and the research team estimates the volume of the ice sheet was not reduced by more than 25% during the warmest years of the Eemian period.34 This may indicate that high sea levels during the Eemian period are primarily attributable to the collapse of the West Antarctic Ice Sheet (WAIS).

From the Deep SouthAt a field camp 1,045 km [650 mi] from the mag-netic South Pole, engineers and scientists have recently recovered an ice core that dates 68,000 years into the past. The WAIS Divide Ice Core Project provides southern hemisphere cli-mate and greenhouse gas records that are of comparable time resolution and duration to the

Greenland ice cores. This ice core allows scien-tists to compare environmental conditions between the northern and southern hemispheres with greater detail than before and allows them to study the levels of greenhouse gases present in ancient atmospheres.

Researchers are using the ice core to under-stand the history of the WAIS to provide further insights into past atmospheric composition and abrupt climate change and to investigate the biological signals contained in deep Antarctic ice cores. Because the WAIS Divide core has an order of magnitude less dust than the Greenland ice core has, scientists expect it to provide them with a more detailed atmospheric CO2 record than was possible from Greenland ice. Many

34. “Greenland Ice Cores Reveal Warm Climate of the Past,” University of Copenhagen, Niels Bohr Institute (January 22, 2013), http://www.nbi.ku.dk/english/news/news13/greenland-ice-cores-reveal-warm-climate-of-the-past/ (accessed October 23, 2013).

35. Thompson LG: “Ice Core Evidence for Climate Change in the Tropics: Implications for Our Future,” Quaternary Science Reviews 19, no. 1–5 (January 2000): 19–35.

other gases (both greenhouse and nongreen-house) and their isotopes are being measured at unprecedented precision and resolution.

The research team recovered the ice core from ice that is more than 3,460 m [11,300 ft] thick; they stopped drilling just 50 m [165 ft] above bedrock to avoid contaminating water at the bottom of the ice that has remained isolated from the environment for at least 100,000 years. Because snow falling at the WAIS Divide rarely melts, each of the past 40,000 years can be identi-fied in individual layers of ice (above). Deeper than that depth, individual annual layers are not as readily identifiable, but the core contains a higher time resolution record than any previously recovered cores. Results from the analysis of this

>West Antarctic Ice Sheet core. Because snowfall at the WAIS Divide rarely melts, ice layers for the past 40,000 years are unbroken, and their divisions are visible and easily counted. The ice also contains much less dust than other ice sheets do. A dark ash layer, however, in this 2 m [6.5 ft] long core section is clearly visible. (Photograph courtesy of Heidi A. Roop, WAIS Divide Science Coordination Office, University of New Hampshire.)

36. Thompson LG, Mosley-Thompson E, Davis ME, Zagorodnov VS, Howat IM, Mikhalenko VN and Lin PN: “Annually Resolved Ice Core Records of Tropical Climate Variability over the Past ~1800 Years,” Science 340, no. 6135 (May 24, 2013): 945–950.

37. Fischer H, Severinghaus J, Brook E, Wolff E, Albert M, Alemany O, Arthern R, Bentley C, Blankenship D, Chappellaz J, Creyts T, Dahl-Jensen D, Dinn M,

Frezzotti M, Fujita S, Gallee H, Hindmarsh R, Hudspeth D, Jugie G, Kawamura K, Lipenkov V, Miller H, Mulvaney R, Parrenin F, Pattyn F, Ritz C, Schwander J, Steinhage D, van Ommen T and Wilhelms F: “Where to Find 1.5 Million Yr Old Ice for the IPICS ‘Oldest Ice’ Ice Core,” Climate of the Past 9 (November 5, 2013): 2489–2505.

Winter 2013/2014 15

formed in the high Andean plain of southern Peru. The snow that became the ice that formed the cores originated to the east of the Quelccaya ice cap; this snow was also affected by El Niño weather effects originating in the west. Because El Niño is a temporary climate change that is driven by sea surface temperatures, the chemical signature in the Quelccaya ice cap is a proxy for sea surface temperatures in the equatorial Pacific Ocean over the past 1,800 years (left).36 Chemical and isotopic records from these tropical ice cores provide historical evidence of the nature of cli-mate change in the lower latitude regions of the planet and a context for current changes.

Perfecting the Tools Although early ice coring drills were based on concepts from geologic drilling, current state-of-the-art units include advances not attempted in rock drilling. For example, the DISC drill’s ability to retrieve replicate cores from the high side of the borehole, while leaving the original hole accessible for borehole logging studies, is an innovation that is unique to ice core drilling.

Enabled by a number of advances in technol-ogy, the relatively young science of using ice cores to understand past climates and environments has yielded societally relevant and important discoveries. As it matures, the science is certain to provide climatologists with increasingly clear insight into the future of Earth’s climate.

These technological advances in ice core drilling may also help answer a question that has plagued scientists for decades. Currently, ice cor-ing experts are working to choose the optimal surface location in the East Antarctic Ice Sheet from which to drill into 1.5 million–year-old ice.37 Proxy records from such an ice core may help solve the mystery behind a climate transition that analyses from marine sediments indicate took place between 900,000 and 1.2 million years ago. Before this Middle Pleistocene event, the time between Earth’s warming periods and ice ages was about 41,000 years; since the climate change during the Pleistocene, the time between these temperature extremes has been about 100,000 years. To this date, the cause of this tran-sition is unknown, and scientists hope the answer is in the air bubbles and chemistry of this now reachable ancient ice. —RvF

core have resolved a scientific debate concerning relationships between climate change in the northern and southern hemispheres and con-firmed that the WAIS is very sensitive to condi-tions in the southern oceans. Scientists expect additional significant results from further analy-sis now that drilling of the ice core is complete.

To the Top of the Mountain Ice coring is not restricted to extreme polar climes and has been performed in high-altitude, low-latitude glaciers. The wellbores in these trop-ical locations are shallower than those drilled in ice sheets at high latitudes, but these ice cores effectively reveal details of the Earth’s tropical climate history. Tropical ice cores have been retrieved from such geographically disparate

> Identifying climate events. Decadal averages of δ18O, net accumulation, insoluble dust, ammonium and nitrate in the Quelccaya Summit Dome ice core from the Quelccaya ice cap of Peru allowed scientists to identify specific climatological periods (shading). The asterisk on the dust profile indicates the 1600 CE eruption of Huaynaputina in Peru. (Adapted from Thompson et al, reference 36.)

1,200

1,100

1,000

900

800

700

600

500

400

300

1,300

1,400

1,500

1,600

1,700

1,800

1,900

2,000–20 –18 –16 0

0.8 1.2 1.6 0 50 100

0 100 2002 4 6Ye

ar, C

E

Early LittleIce Age

*

Late LittleIce Age

Currentwarm period

Nitrate, ppbDust, 104 ppb/mL

Accumulation, mwater equivalent/year

Ammonium, ppb

δ18O, 0/00

Medievalclimateanomaly

locations as the Himalaya and the Andes moun-tain ranges.

In 2003, scientists from The Ohio State University, Columbus, USA, retrieved ice cores 1.92 km [1.19 mi] apart from two wellbores—Quelccaya Summit Dome and Quelccaya North Dome—drilled to bedrock in the Quelccaya ice cap of the Peruvian Andes. Researchers analyz-ing the cores found that each of the 1,800 years spanned by the cores was clearly defined by alter-nating dark and light layers. The dark layers are tinted by dust accumulated during dry seasons; the light layers are the result of snowfall during wet seasons.35

In addition to the unprecedented clarity of the annual layers, the Quelccaya ice cap cores are important to climatologists because they were

16 Oilfield Review

Seeking the Sweet Spot: Reservoir and Completion Quality in Organic Shales

Placement of horizontal wells in shale reservoirs can be a costly and risky business

proposition. To minimize risk, operators acquire and analyze surface seismic data

before deciding where to drill.

Karen Sullivan GlaserCamron K. MillerHouston, Texas, USA

Greg M. JohnsonBrian ToelleDenver, Colorado, USA

Robert L. KleinbergCambridge, Massachusetts, USA

Paul MillerKuala Lumpur, Malaysia

Wayne D. PenningtonMichigan Technological UniversityHoughton, Michigan, USA

Oilfield Review Winter 2013/2014: 25, no. 4. Copyright © 2014 Schlumberger.For help in preparation of this article, thanks to Alan Lee Brown, Raj Malpani, William Matthews, David Paddock and Charles Wagner, Houston; Helena Gamero Diaz, Frisco, Texas; and Ernest Gomez, Denver.sCore is a mark of Schlumberger.

In the late 20th century, E&P geoscientists began to consider shales in a new light. Although production from shales had been established in the early 1800s, operators considered shale for-mations mainly as source rocks and low-perme-ability seals for conventional reservoirs. However, during the 1980s and 1990s, operators showed that the proper application of horizon-tal drilling combined with multistage hydraulic fracturing could make organic shales produc-tive, spurring the exploitation of organic shales as self-sourced reservoirs.1 Despite the success-ful development of the Barnett and Haynesville shales in the US, the industry quickly realized that not all shales are viable targets for eco-nomic hydrocarbon production, and operators sought technologies that could identify appro-priate targets for development.

Shale formations that offer the best potential require a unique combination of reservoir and geomechanical rock properties; such formations are relatively rare. Organic shales have extremely small pore size and ultralow matrix permeability, which makes these unconventional resource plays fundamentally different from most conven-tional reservoirs.2 Furthermore, because hydro-carbon migration paths tend to be short, productive zones of shale reservoirs may be con-fined to a certain area within a basin or restricted to a stratigraphic interval.

The two factors that determine the economic viability of a shale play are reservoir quality and completion quality. Good reservoir quality (RQ) is defined for organic shale reservoirs as the ability to produce hydrocarbons economically after hydraulic fracture stimulation. Reservoir quality is

a collective prediction characteristic that is largely governed by mineralogy, porosity, hydrocarbon saturation, formation volume, organic content and thermal maturity.

Completion quality (CQ), another collective prediction attribute, helps predict successful res-ervoir stimulation through hydraulic fracturing. Similar to RQ, CQ largely depends on mineralogy but is also influenced by elastic properties such as Young’s modulus, Poisson’s ratio, bulk modulus and rock hardness. Completion quality also incor-porates factors such as natural fracture density and orientation, intrinsic and fractured material anisotropy and the prevailing magnitudes, orien-tations and anisotropy of in situ stresses.

To be successful in today’s shale plays, opera-tors drill horizontally within reservoir strata that possess superior RQ and CQ. Stimulation treat-ments are most effective when the induced frac-tures remain propped open, thereby exposing the reservoir to a large fracture surface area and allowing fluids to flow from the reservoir to the wellbore, effectively raising the reservoir’s sys-tem permeability.3

Operators judge the quality of a hydraulic fracture completion design based on a postjob evaluation of data from sources such as micro-seismic monitoring of hydraulic fracturing, flowback tests and initial production to deter-mine how effectively and efficiently the reser-voir was stimulated.

Ideally, an operator places horizontal wells within shale intervals with favorable geologic characteristics and high RQ and CQ and without geohazards.4 Retrospective studies have demon-strated that this strategy would have resulted in

1. Boyer C, Kieschnick J, Suarez-Rivera R, Lewis RE and Waters G: “Producing Gas from Its Source,” Oilfield Review 18, no. 3 (Autumn 2006): 36–49.

Boyer C, Clark B, Jochen V, Lewis R and Miller CK: “Shale Gas: A Global Resource,” Oilfield Review 23, no. 3 (Autumn 2011): 28–39.

2. Nelson PH: “Pore-Throat Sizes in Sandstones, Tight Sandstones, and Shales,” AAPG Bulletin 93, no. 3 (March 2009): 329–340.

3. System permeability refers to the overall permeability of the effective reservoir volume and is the sum of contributions from matrix permeability and natural fracture permeability. Matrix permeability in shale reservoirs ranges from 0.1 to 1,000 nD. Natural and induced fractures are necessary for wells in these formations to be economically productive.

4. Miller C, Waters G and Rylander E: “Evaluation of Production Log Data from Horizontal Wells Drilled in Organic Shales,” paper SPE 144326, presented at the SPE North American Unconventional Gas Conference and Exhibition, The Woodlands, Texas, USA, June 14–16, 2011.

Cipolla C, Lewis R, Maxwell S and Mack M: “Appraising Unconventional Resource Plays: Separating Reservoir Quality from Completion Effectiveness,” paper IPTC 14677, presented at the International Petroleum Technology Conference, Bangkok, Thailand, February 7–9, 2012.

Winter 2013/2014 17

18 Oilfield Review

as much as a tenfold increase in production (below).5 Determining where the best RQ and CQ coincide is therefore an exploration effort, and the best technique to enhance the exploration effort, before drilling the initial well, is interpre-tation of surface seismic data. Recent studies have indicated that seismic interpretation is use-ful for defining production sweet spots within organic shale plays.

In this article, we describe a systematic and strategic approach for using surface seismic data to identify reservoir sweet spots in shale resource plays, starting with basinal and regional RQ and progressing toward local RQ and CQ. Case studies from the Arkoma, Delaware and Williston basins in the US demonstrate how reflection seismic data provide the key to deter-

mining where a resource play may exist and where RQ and CQ are highest.

Mudstone CharacteristicsGeologists define shales as mudstones that exhibit fissility—the ability to split easily, like a deck of playing cards, into individual laminae. The oil and gas industry typically considers resource plays as gas- or liquid-producing “shales.” However, it would be more accurate to call them mudstones or mudrocks, because these “shales” often are not fissile.

Mudstones dominate the sedimentary record and make up roughly 60% to 70% of Earth’s sedimentary rocks.6 They are fine-grained sedi-mentary rocks composed of clay- and silt-size par-ticles with diameters of less than or equal to

62.5 micron [0.00246 in.].7 These small particle sizes result in low permeability; poor sorting—the mixing of various grain sizes—can further reduce both permeability and porosity.

Mudstones have a complex mix of organic mat-ter and clay minerals—illite, smectite, kaolinite and chlorite—along with quartz, calcite, dolomite, feldspar, apatite and pyrite. Geologists with Schlumberger recently introduced the sCore ter-nary diagram mudstone classification scheme, which is built on relationships established between core and log, using clay, QFM (quartz, feldspar and mica) and carbonate as the corner points. The sCore diagram defines 16 classes of mudstones and can classify a sample as an argilla-ceous (clay-rich), siliceous or carbonate mud-stone. This classification scheme allows geologists and engineers to examine empirical relationships between mineralogy and factors that influence the RQ and CQ of mudstones by overlaying points that include indications of RQ, CQ or both (next page).8 Productive mudstones most sought by oil compa-nies tend to be dominated by nonclay minerals, principally silicates and carbonates, and therefore lie in the lower part of the diagram, away from the clay point; higher RQ and CQ rocks are near the edges of the triangle.9

Several factors control the physical proper-ties of mudstones: the mineralogy and propor-tions of grains, the fabric of the originally deposited muds and the postdepositional pro-cesses—such as resuspension, redeposition, dia-genesis, bioturbation and compaction—that convert mud into rock.10 Mudstones tend to be highly heterogeneous, and this heterogeneity can vary horizontally and vertically, originating from the sequence of depositional environments and tectonic regimes that prevailed as the mud strata stacked up through geologic time.

An individual layer of mud, called a lamina-tion, is typically about a millimeter thick. Laminations stack up to form laminae sets, called beds. Beds in turn stack up to form bed sets that group together into members and then into geo-logic formations. The mineral and organic com-position of each layer depends on the sequence

> Best 12-month production results. This 50-mi2 [130-km2] area of the Barnett Shale play in northwest Tarrant County, Texas, USA, shows the first year’s gas production for more than 650 horizontal wells. Black dots represent surface locations of well pads, which may service multiple wells. Areas of warm colors (top of scale) are production sweet spots and areas of cool colors (bottom of scale) are not. (Adapted from Baihly et al, reference 5.)

120

280

200

360

440

520

600

Gas

prod

uctio

n, M

Mcf

0 km

0 mi 2

2

N

5. Baihly JD, Malpani R, Edwards C, Han SY, Kok JCL, Tollefsen EM and Wheeler CW: “Unlocking the Shale Mystery: How Lateral Measurements and Well Placement Impact Completions and Resultant Production,” paper SPE 138427, presented at the SPE Tight Gas Completions Conference, San Antonio, Texas, November 2–3, 2010.

6. Aplin AC, Fleet AJ and Macquaker JHS: “Muds and Mudstones: Physical and Fluid-Flow Properties,” in Aplin AC, Fleet AJ and Macquaker JHS (eds): Muds and Mudstones: Physical and Fluid-Flow Properties. London: The Geological Society, Special Publication 158 (1999): 1–8.

7. A micron, or micrometer, is equal to one millionth of a meter or one thousandth of a millimeter. Its abbreviation

is μ, μm or mc. In Imperial units, a micron equals 3.937 × 10–5 in.

8. For more on the sCore classification scheme: Gamero-Diaz H, Miller C and Lewis R: “sCore: A Mineralogy Based Classification Scheme for Organic Mudstones,” paper SPE 166284, presented at the SPE Annual Technical Conference and Exhibition, New Orleans, September 30–October 2, 2013.

9. Loucks RG and Ruppel SC: “Mississippian Barnett Shale: Lithofacies and Depositional Setting of a Deep-Water Shale-Gas Succession in the Fort Worth Basin, Texas,” AAPG Bulletin 91, no. 4 (April 2007): 579–601.

Passey QR, Bohacs KM, Esch WL, Klimentidis R and Sinha S: “From Oil-Prone Source Rock to Gas-Producing Shale Reservoir—Geologic and Petrophysical

Characterization of Unconventional Shale-Gas Reservoirs,” paper SPE 131350, presented at the CPS/SPE International Oil and Gas Conference and Exhibition in China, Beijing, June 8–10, 2010.

Lash GG and Engelder T: “Thickness Trends and Sequence Stratigraphy of the Middle Devonian Marcellus Formation, Appalachian Basin: Implications for Acadian Foreland Basin Evolution,” AAPG Bulletin 95, no. 1 (January 2011): 61–103.

10. Aplin AC and Macquaker JHS: “Mudstone Diversity: Origin and Implications for Source, Seal, and Reservoir Properties in Petroleum Systems,” AAPG Bulletin 95, no. 12 (December 2011): 2031–2059.

Winter 2013/2014 19

> sCore classification tool. In the clockwise direction, the corners of the sCore ternary diagram (top left) are clay, carbonate and quartz plus feldspars plus micas (QFM). The diagram defines 16 classes of mudstones based on mineralogy. The mudstones (top right) sought by oil companies tend to have less than 50% clay. In the Wolfcamp Shale (middle), siliceous mudstones exhibit high RQ and CQ. In the Eagle Ford Shale (bottom), carbonate mudstones have high RQ and CQ. In these examples, RQ is directly proportional to effective porosity and CQ is inversely proportional to the stress gradient of the minimum in situ principal compressive stress.

Carbonate-richsiliceous mudstone

Silica-richcarbonate mudstone

Silica-dominatedlithotype

QFM 20

20

50

50

80

80

80

50

20

10

80

50

20

10

10

10

20

50

80

10

20 50 80 10

Clay Clay

Carbonate

Shale plays

Barnett

Marcellus

Haynesville

Fayetteville

Eagle Ford

Wolfcamp

QFM Carbonate

Mixedsiliceousmudstone

Argillaceous-siliceousmudstone

Mixedmudstone

Argillaceous-carbonatemudstone

(marl)

Mixed argillaceousmudstone

Clay-dominatedlithotype

MixedcarbonatemudstoneCarbonate-

siliceousmudstone

Clay

-rich

silice

ous m

udsto

ne

Silic

a-ric

h

argil

laceo

us m

udsto

ne Carbonate-rich

argillaceous mudstone

Clay-rich

carbonate mudstone

Carbonate-dominatedlithotype

sCore Classification

QFM QFM

80

50

20

10

80

50

20

10

20

50

80

10

20

50

80

10

10 20 50 80 10 20 50 80

Clay Clay

Carbonate

High RQHigh CQ

Carbonate

Effective porosity, fraction

Incr

easi

ng R

Q

0.150

0.120

0.090

0.061

0.031

0.001

Minimum stress gradient, psi/ft

Incr

easi

ng C

Q

0.62

0.65

0.68

0.71

0.74

0.77

Wolfcamp Shale

QFM QFM

80

50

20

10

80

50

20

10

20

50

80

10

20

50

80

10

10 20 50 80 10 20 50 80

Clay Clay

High RQHigh CQ

CarbonateCarbonate

Minimum stress gradient, psi/ft

Incr

easi

ng C

Q0.83

0.85

0.88

0.90

0.93

0.95

Effective porosity, fraction

Incr

easi

ng R

Q

0.080

0.064

0.048

0.033

0.017

0.001

Eagle Ford Shale

20 Oilfield Review

or history of geologic conditions of the area through time. Geologists use the principles of stratigraphy to decipher this geologic history.11

Layering has a particular effect on some rock properties: It is a fabric that causes anisotropy.12 A rock is anisotropic if its properties vary with direction.13 A consequence of layering is that the composition, size, shape, orientation, packing and sorting of the particles in the layer tend to vary more quickly perpendicular to, rather than parallel to, layers. As a result, rock properties tend to vary with direction. They are different if measured layer parallel than if measured layer perpendicular. Another aspect of rocks that can lead to anisotropy is the presence of roughly par-allel open fracture networks, which can control the efficiency of reservoir stimulation. Because anisotropy is observable in seismic data, geophys-

icists are able to characterize it for geologists and engineers to use in their various geologic, geomechanical and fluid-flow models of the pro-spective reservoir (below).

Mudstones play an important role in a petroleum system. Their small grain sizes and sorting characteristics contribute to their char-acterization as low-porosity rocks with low to ultralow permeability and high fluid-displace-ment entry pressures. Consequently, when mud-stones are in the correct stratigraphic and structural location and configuration, they form the seals that cap and delimit conventional hydrocarbon reservoir geometries.

Some mudstones are characterized as organic rich, and these have been viewed historically as the source rocks that, through secondary migra-tion, supply hydrocarbons to adjacent and remote

conventional and unconventional continuous res-ervoirs. These same organic-rich mudstones have also proved to be self-sourcing reservoirs, yielding hydrocarbons that have been expelled and under-gone primary migration to be stored within the source rocks themselves.14 For example, the Eagle Ford Shale in south Texas, USA, is a mudstone that sources the prolific Austin Chalk fractured reservoir, which has been explored and produced for more than 80 years. Now, operators recognize the Eagle Ford Shale itself as a reservoir capable of producing oil, condensate, wet gas and dry gas that simply never left the source rock.15

Not all mudstones contain sufficient hydro-carbons to be considered potential reservoir rocks. Mudstones are defined as organic rich if their total organic carbon (TOC) concentration is greater than 2 weight %.16 The preservation and

>Mudstone layering at many scales. Layering may be observed in photographs of outcrop, core and thin section. The Eagle Ford Shale outcrop (left) is in Lozier Canyon in Terrell County, Texas. The images of core (plain and ultraviolet light, center) and thin section (original and close-up, right) are of lower Eagle Ford Shale from the BP-Schlumberger Lozier Canyon number 1 well. The 2-ft [0.6-m] core length was cut from depths 226 to 228 ft [68.9 to 69.5 m]. The thin section is of fossiliferous siliceous-calcareous mudstone and has a mineralized fracture running along its right side, which has been stained with potassium ferricyanide and alizarin red S dye to distinguish carbonate minerals. In the close-up view from the thin section, there is evidence that the fracture propagated, stopped and restarted along a different path. (Outcrop photograph courtesy of Karen Sullivan Glaser. Core and thin section images courtesy of Schlumberger and BP America Incorporated.)

0 mm 5

0 mm 0.5

0 m

0 ft 3

1

226

227

Core

leng

th, f

t

Plain Light Ultraviolet Light0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

1.9

2.0

Winter 2013/2014 21

richness of organic matter depend on the relative rates of its production, dilution and destruction (right).17 Inorganic matter deposited at the same time as organic matter will dilute organic matter concentration. Destruction of organic matter occurs through bacterial consumption, by oxida-tion reactions at shallow depths and by deeper thermally activated reactions, which transform part of the organic matter into oil and gas before it ultimately changes to graphite, or dead carbon. The primary portion of organic matter in source rock is kerogen, which is insoluble in common organic solvents; the other portion is bitumen, which is soluble.

Kerogen has petrophysical characteristics that differ significantly from those of the mineral con-stituents in shale, and these characteristics affect the overall bulk properties of the reservoir rock. For example, depending on kerogen type and maturity, the density of kerogen can vary from 1.1 to 1.4 g/cm3, considerably less than the bulk den-sity of its shale host rock.18 Consequently, the bulk density of organic-rich shales appears lower (as if the shale has a higher porosity) than that of shales containing lower concentrations of kerogen.

The distribution of kerogen varies from iso-lated particles dispersed through the mudstone matrix to lenses and sheets aligned with mud-stone laminae. Investigators have found that kerogen particles contain secondary porosity that likely formed during thermal maturation.19 This organic porosity occurs as nanopores, defined as less than 1 micron in diameter.

Kerogen fabric affects the physical properties of organic-rich mudstones. When the organic con-tent is high and the kerogen forms interconnected layer-parallel networks through the mudstone

frame, the organic porosity may be sufficient for hydrocarbon storage and for providing permeabil-ity to gas and liquid hydrocarbon in an otherwise extremely low-permeability matrix.20

> Organic matter. The thin section (left), which has been stained with potassium ferricyanide and alizarin red S dye on its left side, is of calcareous pelletal mudstone. In the close-up view (right), the layer is composed of planktonic foraminifera (white and pink), flattened fecal pellets (reddish brown) and organic matter (black). (Core and thin section images courtesy of Schlumberger and BP America Incorporated.)

0 mm 5

0 μm 200

11. Neal J, Risch D and Vail P: “Sequence Stratigraphy— A Global Theory for Local Success,” Oilfield Review 5, no. 1 (January 1993): 51–62.

12. Fabric refers to the spacing, arrangement, distribution, size, shape and orientation of the constituents of rocks, such as minerals, grains, organic matter, porosity, layering, bed boundaries, lithology contacts and fractures. Fabric elements contribute to the anisotropy of materials when the elements have preferential orientation along crystallographic axes, fractures and elongated and flat particles.

13. For more on permeability anisotropy: Ayan C, Colley N, Cowan G, Ezekwe E, Wannell M, Goode P, Halford F, Joseph J, Mongini A, Obondoko G and Pop J: “Measuring Permeability Anisotropy: The Latest Approach,” Oilfield Review 6, no. 4 (October 1994): 24–35.

For more on elastic anisotropy: Armstrong P, Ireson D, Chmela B, Dodds K, Esmersoy C, Miller D, Hornby B, Sayers C, Schoenberg M, Leaney S and Lynn H: “The Promise of Elastic Anisotropy,” Oilfield Review 6, no. 4 (October 1994): 36–47.

For more on the anisotropy of electrical properties: Anderson B, Bryant I, Lüling M, Spies B and Helbig K: “Oilfield Anisotropy: Its Origins and Electrical Characteristics,” Oilfield Review 6, no. 4 (October 1994): 48–56.

14. Primary migration refers to the flow of newly generated hydrocarbon fluids within source rocks. Secondary migration refers to the flow of free hydrocarbon fluids away from source rocks toward adjoining or distant reservoir rocks.

15. Martin R, Baihly J, Malpani R, Lindsay G and Atwood WK: “Understanding Production from Eagle Ford–Austin Chalk System,” paper SPE 145117, presented at the SPE Annual Technical Conference and Exhibition, Denver, October 30–November 2, 2011.

16. Boyer et al (2006), reference 1. Loucks and Ruppel, and Lash and Engelder, reference 9. The bulk volume fraction or percent of TOC in rock is

roughly twice its weight fraction or percent. A concentration of 2 weight % [0.02 kg/kg] TOC in rock is approximately equivalent to 4 bulk volume % [0.04 m3/m3] TOC. The exact calculation depends on the density and maturity of the organic matter and the bulk density of the host rock.

17. Bohacs KM, Grabowski GJ Jr, Carroll AR, Mankiewski PJ, Miskell-Gerhardt KJ, Schwalbach JR, Wegner MB and Simo JA: “Production, Destruction, and Dilution—The Many Paths to Source-Rock Development,” in Harris NB (ed): The Deposition of Organic-Carbon-Rich Sediments: Models, Mechanisms, and Consequences. Tulsa: Society of Sedimentary Geology, SEPM Special Publication 82 (2005): 61–101.

For more on source rock geochemistry: McCarthy K, Rojas K, Niemann M, Palmowski D, Peters K and Stankiewicz A: “Basic Petroleum Geochemistry for Source Rock Evaluation,” Oilfield Review 23, no. 2 (Summer 2011): 32–43.

18. The density of kerogen increases as organic carbon matures from immature, generative organic carbon to overmature, nongenerative organic carbon. For more on kerogen: Jarvie DM, Jarvie BM, Weldon WD and

Maende A: “Components and Processes Impacting Production Success from Unconventional Shale Resource Systems,” Search and Discovery Article 40908, adapted from an oral presentation at the 10th Middle East Geosciences Conference and Exhibition, Manama, Bahrain, March 4–7, 2012.

Okiongbo KS, Aplin AC and Larter SR: “Changes in Type II Kerogen Density as a Function of Maturity: Evidence from the Kimmeridge Clay Formation,” Energy & Fuels 19, no. 6 (November 2005): 2495–2499.

19. Loucks RG, Reed RM, Ruppel SC and Jarvie DM: “Morphology, Genesis, and Distribution of Nanometer-Scale Pores in Siliceous Mudstones of the Mississippian Barnett Shale,” Journal of Sedimentary Research 79, no. 12 (December 2009): 848–861.

Curtis ME, Cardott BJ, Sondergeld CH and Rai CS: “Development of Organic Porosity in the Woodford Shale with Increasing Thermal Maturity,” International Journal of Coal Geology 103 (December 1, 2012): 26–31.

20. Wang FP and Reed RM: “Pore Networks and Fluid Flow in Gas Shales,” paper SPE 124253, presented at the SPE Annual Technical Conference and Exhibition, New Orleans, October 4–7, 2009.

Ambrose RJ, Hartman RC, Diaz-Campos M, Akkutlu IY and Sondergeld CH: “Shale Gas-in-Place Calculations Part I: New Pore-Scale Considerations,” SPE Journal 17, no. 1 (March 2012): 219–229.

Curtis ME, Sondergeld CH, Ambrose RJ and Rai CS: “Microstructural Investigation of Gas Shales in Two and Three Dimensions Using Nanometer-Scale Resolution Imaging,” AAPG Bulletin 96, no. 4 (April 2012): 665–677.

22 Oilfield Review

In addition, kerogen fabric affects the elastic and mechanical properties of reservoir rocks.21 Generally, mudstones that contain intercon-nected kerogen within their frame are character-ized by lower elastic moduli and higher ductility compared with those mudstones that have iso-lated kerogen particles dispersed through their matrix. Kerogen content distributed parallel to the laminae may profoundly affect the anisotro-pic elastic and mechanical properties of mud-stones.22 These effects will be enhanced if, in addition to creating secondary porosity within kerogen, hydrocarbon generation and charging of the kerogen-rich laminae cause overpressuring, a condition that is conducive to creating layer-par-allel microcracks, which strike parallel to layers and open perpendicular to them.23 Because matrix permeability in shale reservoirs is excep-tionally low, ranging from 10–7 to 10–3 mD, natu-ral fractures play a significant role in reservoir completions and hydrocarbon production.

Natural fractures contribute to the perfor-mance of hydraulic fracture stimulations by pro-viding planes of weakness and conduits for fluid flow.24 As planes of weakness, natural fractures may dictate the propagation and development of induced fracture networks, especially if the in situ stress anisotropy is low.25 As conduits for fluid flow, these fractures may enhance the extent of effective reservoir volume drained by the wellbore, and they may admit high-pressure fluids, which could cause permanent shear slip-page along their fracture planes and increase fracture aperture and conductivity.

Local RQ sweet spots in prospective mud-stone reservoirs often contain natural fractures that provide flow pathways connecting matrix porosity and storage to hydraulic fractures and the well. Natural fractures may also affect CQ through the geometry of stimulation-induced hydraulic fracture networks, which tend to become more widespread and complex when pre-existing natural fracture networks are oriented at an angle to the present-day principal horizon-tal stress.26 When mudstone reservoirs lack natu-ral fractures, operating companies must rely on hydraulic fracture stimulations to create induced fracture networks that connect production from the reservoir matrix to the wellbore. Therefore, natural fractures, which can increase both RQ and CQ, are a seismic exploration target in the hunt for sweet spots in shale reservoirs.

Through the analysis of seismic attributes, geo-physicists detect and characterize fracture networks. This process takes advantage of the volume-averaged response from the entire reservoir interval contain-ing an open natural fracture system.27

Numerous fracture detection methods use seis-mic attributes. When natural fractures align in a consistent strike orientation, they cause elastic properties and seismic attributes to vary with azi-muth, including velocity and reflection ampli-tude.28 Geophysicists observe these variations from analysis of 3D surface seismic surveys that have been acquired along multiple azimuths.29 Azimuthal analysis of shear waves (S-waves) has proved to be a good fracture detection method.30 Analysis of seismic waveform scattering, which was often treated as noise in the past, may also reveal information about fracture orientation and spacing through frequency analysis.31 In addition, combinations of attributes such as reflection strength and seismic variance—the variation between seismic samples—may be blended, or superimposed, to expose subtle structural features that have fracture systems associated with them.32

Regional- or Basin-Scale Sweet SpotsDuring the initial years of the current surge in activity in shale plays, some operators were able to develop shale plays on the basis of hydrocar-

bon shows on mud logs that were recorded in shales encountered while drilling conventional reservoirs within a basin. The regions within these basins where organic shales are thermally mature were already known to the industry; therefore, for many of the shale plays in North America, it was unnecessary for operators to investigate the plays’ thermal maturity.

Because of the successful development of the Barnett Shale in the Fort Worth basin in north-central Texas, operators widened their search for shale gas beyond North America to basins that are less explored. In certain basins around the world, few wells have been drilled, and operators lack the level of understanding of the strati-graphic and structural framework to anticipate where potential shale resource plays exist. In these basins, initial exploration for potential shale reservoirs relies on evaluating preexisting 2D seismic surveys and additional structural data from remote sensing analyses and outcrop stud-ies of surface geology.

Geoscientists evaluate these data to establish the structural framework of the major basinal

21. Suárez-Rivera R, Deenadayalu C and Yang Y-K: “Unlocking the Unconventional Oil and Gas Reservoirs: The Effect of Laminated Heterogeneity in Wellbore Stability and Completion of Tight Gas Shale Reservoirs,” paper OTC 20269, presented at the Offshore Technology Conference, Houston, May 4–7, 2009.

22. Vernik L and Landis C: “Elastic Anisotropy of Source Rocks: Implications for Hydrocarbon Generation and Primary Migration,” AAPG Bulletin 80, no. 4 (April 1996): 531–544.

Vernik L and Milovac J: “Rock Physics of Organic Shales,” The Leading Edge 30, no. 3 (March 2011): 318–323.

Sayers CM: “The Effect of Kerogen on the Elastic Anisotropy of Organic-Rich Shales,” Geophysics 78, no. 2 (March–April 2013): D65–D74.

23. For more on layer-parallel microcracks: Lash GG and Engelder T: “An Analysis of Horizontal Microcracking During Catagenesis: Example from the Catskill Delta Complex,” AAPG Bulletin 89, no. 11 (November 2005): 1433–1449.

24. Miller C, Hamilton D, Sturm S, Waters G, Taylor T, Le Calvez J and Singh M: “Evaluating the Impact of Mineralogy, Natural Fractures and In Situ Stresses on Hydraulically Induced Fracture System Geometry in Horizontal Shale Wells,” paper SPE 163878, presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, Texas, February 4–6, 2013.

25. Weng X, Kresse O, Cohen C, Wu R and Gu H: “Modeling of Hydraulic Fracture-Network Propagation in a Naturally Fractured Formation,” SPE Production & Operations 26, no. 4 (November 2011): 368–380.

Kresse O, Cohen C, Weng X, Wu R and Gu H: “Numerical Modeling of Hydraulic Fracturing in Naturally Fractured Formations,” paper ARMA 11–363, presented at the 45th US Rock Mechanics/Geomechanics Symposium, San Francisco, June 26–29, 2011.

26. Miller et al, reference 24.27. For more on fracture detection using reflection

seismology: Aarre V, Astratti D, Al Dayyni TNA, Mahmoud SL, Clark ABS, Stellas MJ, Stringer JW, Toelle B, Vejbæk OV and White G: “Seismic Detection of Subtle Faults and Fractures,” Oilfield Review 24, no. 2 (Summer 2012): 28–43.

28. For more on elastic anisotropy: Armstrong et al, reference 13.

29. For more on azimuthal seismic anisotropy analysis: Barkved O, Bartman B, Compani B, Gaiser J, Van Dok R, Johns T, Kristiansen P, Probert T and Thompson M: “The Many Facets of Multicomponent Seismic Data,” Oilfield Review 16, no. 2 (Summer 2004): 42–56.

30. Hardage B: “Fracture Identification and Evaluation Using S Waves,” Search and Discovery Article 40792, adapted from five Geophysical Corner columns by B Hardage in AAPG Explorer 32, no. 4–8 (April–August 2011).

31. Burns DR, Willis ME, Toksöz MN and Vetri L: “Fracture Properties from Seismic Scattering,” The Leading Edge 26, no. 9 (September 2007): 1186–1196.

32. High seismic variance occurs where seismic data vary rapidly, such as when crossing faults or stratigraphic boundaries.

33. For more on petroleum system modeling: Al-Hajeri MM, Al Saeed M, Derks J, Fuchs T, Hantschel T, Kauerauf A, Neumaier M, Schenk O, Swientek O, Tessen N, Welte D, Wygrala B, Kornpihl D and Peters K: “Basin and Petroleum System Modeling,” Oilfield Review 21, no. 2 (Summer 2009): 14–29.

Peters KE, Magoon LB, Bird KJ, Valin ZC and Keller MA: “North Slope, Alaska: Source Rock Distribution, Richness, Thermal Maturity, and Petroleum Charge,” AAPG Bulletin 90, no. 2 (February 2006): 261–292.

Peters K, Schenk O and Bird K: “Timing of Petroleum System Events Controls Accumulations on the North Slope, Alaska,” Search and Discovery Article 30145, adapted from an oral presentation at the AAPG International Conference and Exhibition, Calgary, September 12–15, 2010.

Higley DK: “Undiscovered Petroleum Resources for the Woodford Shale and Thirteen Finger Limestone–Atoka Shale Assessment Units, Anadarko Basin,” Denver: US Geological Survey Open-File Report 2011–1242, 2011.

34. Boyer et al (2006), reference 1.35. US Department of Energy Office of Fossil Energy and

National Energy Technology Laboratory: “Modern Shale Gas Development in the United States: A Primer,” Washington, DC: US Department of Energy, April 2009.

36. The shorter wavelength portions of the seismic signals are scattered enough to become incoherent and cancel themselves out.

Winter 2013/2014 23

stratigraphic units, including the locations of major fault zones and other tectonic features. Once they complete this analysis, basin analysts can use the structural framework as input to petroleum system modeling to determine if organic shale formations might be thermally mature, and if so, where in the basin these may occur.33 When this information is combined with regional mapping of available TOC data, regional- or basin-scale sweet spots may be identified, enabling operators to select optimal locations for drilling initial vertical pilot wells in the next phase of exploration.

Local or Operating Area Sweet SpotsPetroleum system modeling predicts the location and characteristics of basin-scale sweet spots, including the distribution of kerogen content, its thermal maturity and the pore pressure within the prospective interval. However, these predic-tions can be confirmed only by drilling a pilot well. Core and logging measurements from the vertical pilot well provide data to update the modeling and determine whether the pilot well has intersected a sweet spot. Engineers can cat-egorize local sweet spots by analyzing RQ and CQ using the newly acquired core and log data.

Local sweet spots of high RQ have one or more of three properties. Local sweet spots may have high matrix porosity that contains signifi-cant amounts of free gas, which may be produced at high rates during initial production, allowing for rapid payout of a horizontal appraisal well.

In addition, sweet spots may have significant concentrations of kerogen. Kerogen-rich sweet spots also contain large volumes of adsorbed gas, which is stored mainly on kerogen surfaces.34 This adsorbed gas contributes to sustained production as pressure drops during reservoir depletion, long after the free gas has been consumed.

Local RQ sweet spots may also have dense networks of open microfractures. Similar to high-porosity sweet spots, densely fractured sweet spots contain free gas that is produced during the early production of a well. In addition, microfrac-tures also increase the system permeability within a shale reservoir.

The best RQ sweet spots include all three properties—increased porosity, kerogen and microfracturing—which in turn affect various attributes of seismic data through their effect on rock properties. Increased porosity and the pres-ence of fractures typically cause decreases in seis-mic velocity and increased attenuation of high frequencies. Concentrations of kerogen can also

lower the elastic moduli and density of mud-stones, but to a lesser degree. Changes in certain seismic attributes associated with these rock properties may be used to identify RQ sweet spots.

Correlating Frequency Anomalies to

Production BehaviorIn the Arkoma basin of southeast Oklahoma, USA, gas production has been established from the Woodford Shale, a Late Devonian–Early Mississippian age organic-rich mudstone. Its mineralogy is primarily quartz and illite, with small quantities of pyrite and dolomite. Porosity ranges from 3% to 9% and TOC ranges from 1 to 14 weight % [0.01 to 0.14 kg/kg].35

An operator targeting Woodford Shale gas production had drilled six vertical wells within a 4-mi2 [10-km2] area. The wells’ production rates varied widely. In a 2.5-year period, cumulative gas production per well ranged from 18 to 372 MMcf [0.51 to 10.5 million m3] with average cumulative production from the five lowest pro-ducing wells of 40 MMcf [1 million m3]. The operator had acquired a 3D seismic survey over the field and requested that Schlumberger ana-lysts interpret the data to determine why the production was so variable and to locate areas of potentially higher production.

The 3D seismic data provide far greater cover-age of the reservoir interval than could be achieved by vertical or horizontal well data. The 3D seismic data were initially interpreted to locate faults and any other geohazards within the area, but the observed faulting and fracturing associated with fault damage zones could not explain the production history or the well-to- well variability.

Geophysicists analyzed the data for seismic attributes that would reveal RQ sweet spots. They identified a seismic frequency attribute that at certain frequencies corresponded with areas of higher production. These seismic sweet spot anomalies were areas in which the dominant seismic frequency proved to be relatively low, apparently the result of scattering of waves from networks of natural fractures or microfractures.36 The anomalies appeared as isolated patches within the field, and the team interpreted them to represent areas of increased porosity and microfracturing within the shale reservoir. Productive wells were located within these anomalous areas, while the underperforming wells were not. The well with the highest produc-tion is situated within a large anomaly (above). At the time of the study, this well had produced nine times the average production of the other

> Reservoir quality sweet spots in a shale reservoir. Six vertical wells (red dots) were drilled into the Woodford Shale in the Arkoma basin in southeast Oklahoma. Their cumulative gas production after approximately 2.5 years through June 2009 varied widely. Interpretation of a 3D seismic dataset revealed faulting (black). However, the wells’ proximity to faults, which often is associated with fracture density in the faults’ damage zone, did not explain the production variation. An analysis of seismic frequencies in the dataset revealed a frequency attribute that interpreters identified with RQ sweet spots (dashed red outlines) when this attribute was strong. Gas production correlated with the size and strength of the seismically identified sweet spots.

18 MMcf

23 MMcf

Untestedanomaly

22 MMcf

80 MMcf

61 MMcf

372 MMcf

N

0 km

0 mi 2

2

Freq

uenc

y an

omal

y

Strong

Weak

24 Oilfield Review

five wells combined. This observation is in line with the tenfold increase observed in the Barnett Shale for wells located within sweet spots.37

In another shale play, an operator was devel-oping a combination fractured carbonate and gas shale unconventional reservoir in the Delaware basin of southern New Mexico and western Texas, USA. The company had drilled a number of hori-zontal boreholes at the interface between the carbonate and the underlying shale. Production from these wells varied significantly.

Schlumberger geophysicists analyzed a 3D seis-mic volume to help determine the location and extent of potential RQ sweet spots and define their geologic nature. The geophysicists performed prestack azimuthal inversion and several fre-quency-related studies. The results of these sepa-rate investigations converged on the same locations within the shale reservoir as potential RQ sweet spots. These sweet spots manifested themselves through specific frequency-related seismic attri-bute anomalies that were also coincident with zones of S-wave anisotropy. The team interpreted these areas to be volumes of enhanced microfrac-turing in the upper portion of the gas shale (left).

The operator drilled three horizontal wells along the carbonate/shale interface in the hope of encountering fractures within the carbonate formation and zones of high gas content in the shale. Production rates from these wells appeared to be directly related to the magnitude and size of the frequency anomalies and S-wave anisotropy. Well A was drilled across the top of a gentle anti-clinal feature, where high seismic variance indi-cated the presence of faulting along the crest of the fold. At the time of the study, Well A was the best producer, with an average production rate of 64 MMcf [1.8 million m3] of gas per month. Well B was drilled near a smaller seismic frequency anomaly and its monthly production rate was 28 MMcf [0.79 million m3], less than half that from Well A. Well C did not penetrate a frequency anomaly and its monthly production rate was a poor 7 MMcf [0.2 million m3].

The team believed that the frequency anoma-lies highlighted zones within the shale that con-tained more microfractures than other locations. The concentration of microfractures at the crest of the anticline is consistent with the tectonic extension the layers experienced during anti-cline formation. Other evidence suggests that this fracturing did not extend through the full shale thickness. Zones of enhanced microfractur-ing within the shale were encountered by Well A and, to much lesser degree, by Well B, and resulted in the improved production observed in both wells compared with that in Well C.

> Fracture detection with seismic frequency attributes. A seismic fence diagram composed of seismic cross sections and a horizon slice shows a frequency-related seismic attribute. The horizon slice is also blended with the seismic variance attribute (grayscale); only high variance values are shown. The fence diagram (inset) is formed from seismic sections along the trajectories of Wells A, B and C (dark blue). The horizon slice, which is taken along the top of the formation immediately below the shale reservoir, is curved by a gentle anticline. Along the anticlinal crest, the seismic variance and frequency attributes are high. Average monthly gas production rates, shown above each well’s lateral, illustrate how each well’s production rate corresponds to its proximity to strong frequency anomalies.

Freq

uenc

y an

omal

y

Weak

Strong

Variance

HighLow

64 MMcf 28 MMcf 7 MMcf

High variance fault indicator

Anticlinalcrest

Well A Well B Well C

Frequencyanomaly

No faults

Horizon along topof next deeper

formation

Seismic fence diagram

> Gas shows encountered during the drilling of Well A (black line). A seismic section (background) is shown in a perspective view looking down and into it. The section is parallel to the trajectory of Well A and cuts through the 3D volume of the frequency attribute. High values of the frequency attribute (red and pink) appear as clouds coming out of the section. Gas chromatograph readings (blue curve), obtained from the mud log, are shown along the horizontal portion of Well A. Perforation cluster locations (cyan diamonds) align with the mud log depth points (small red triangles below the log curve). Gas shows from the mud log were strong when the wellbore was near high values of the seismically derived frequency attribute.

Freq

uenc

y an

omal

y

Weak

Strong

Frequencyanomaly highs

Farther fromanomaly

Close to anomaly

Mud loggas shows

Strong shows Weak shows

Winter 2013/2014 25

37. Baihly et al, reference 5.38. Sturm SD and Gomez E: “Role of Natural Fracturing in

Production from the Bakken Formation, Williston Basin, North Dakota,” Search and Discovery Article 50199, adapted from a poster presentation at the AAPG Annual Convention and Exhibition, Denver, June 7–10, 2009.

39. Pitman JK, Price LC and LeFever JA: “Diagenesis and Fracture Development in the Bakken Formation, Williston Basin: Implications for Reservoir Quality in the Middle Member,” Denver: US Geological Survey Professional Paper 1653, 2001.

Pollastro RM, Roberts LNR and Cook TA: “Geologic Assessment of Technically Recoverable Oil in the Devonian and Mississippian Bakken Formation,” in US Geological Survey Williston Basin Province Assessment Team (ed): Assessment of Undiscovered Oil and Gas Resources of the Williston Basin Province of North Dakota, Montana, and South Dakota, 2010, Denver: US Geological Survey Digital Data Series DDS–69–W (2011): 5-1–5-34.

40. A continuous petroleum system is one that displays little if any buoyancy, or gravity, segregation of the reservoir fluids. Generated oil or gas migrated directly into reservoir storage within the source rock or adjacent formations. This contrasts with conventional petroleum systems in which generated oil or gas migrated from source rocks into traps that lie beneath a reservoir seal. Conventional reservoirs display distinct fluid contacts, which are products of gravity segregation.

Examination of the gas shows encountered during the drilling of Well A also supported this interpretation (previous page, bottom). The stron-gest gas shows coincided with strong seismic fre-quency anomalies. Where the frequency anomalies were weaker, gas shows were not as strong.

In another location within the same Delaware basin study area, the operator drilled two hori-zontal wells from a vertical pilot well. The wells were drilled from east to west, hydraulically stim-ulated in multiple stages and monitored for induced microseismicity. The team was able to correlate the microseismic event locations to areas where seismic frequency anomalies were strongest (above). It was evident that high levels of frequency anomaly corresponded to RQ sweet spots or, more specifically, to zones of high poros-ity and increased microfracture density. In addi-tion, these zones appeared to have favorable CQ.

Associating Anisotropy to Production PatternsThe Bakken Formation is an oil-producing petro-leum system. Its stratigraphy represents deposi-tion in a restricted, shallow water environment that existed throughout most of the Williston

basin, which covers portions of Alberta, Saskatchewan and Manitoba in Canada and Montana, North Dakota and South Dakota in the US.38 The Bakken Formation is of Late Devonian–Early Mississippian age and lies unconformably above the Late Devonian Three Forks Formation and conformably below the Early Mississippian Lodgepole Limestone.39 The Bakken Formation has been subdivided into lower, middle and upper members. The middle member is the reservoir and is a mixed clastic-carbonate interval consist-ing of dolomitic sandstones, dolomites and lime-stone. The upper and lower members are organic-rich shales that serve as the seal and source for hydrocarbons.

The model for the Bakken Formation is one of a continuous petroleum system.40 The organic-rich upper and lower Bakken shale members have 8 to 10 weight % [0.08 to 0.1 kg/kg] TOC and are source rocks that generated oil that had migrated locally into reservoirs hosted by the adjacent middle Bakken member and the under-lying Pronghorn, which includes the Sanish Sand member of the Three Forks Formation. Because of the relatively closed nature of this petroleum

> Comparison of microseismicity and frequency attribute anomalies indicating zones of good CQ. This perspective view looks down into a west-to-east seismic section. The seismic section is fully opaque, showing all values of the frequency attribute. Two horizontal wells (black curves) were kicked off from a vertical pilot well in the east. Strong values of the seismic frequency attribute within the 3D seismic volume and limited to the upper portion of the shale reservoir are shown as clouds (tan to red). Microseismic events (dots), color-coded by stimulation stage, tend to occur where values of the frequency anomaly are high (white ovals). This relationship suggests that strong values of this frequency attribute may also indicate zones of good CQ.

Freq

uenc

y an

omal

y

Weak

Strong

N

S

EW

system, overpressuring occurs in deeper parts of the basin, where the most hydrocarbon genera-tion has taken place. Pore space and fractures within the upper and lower Bakken shale mem-bers also provide reservoir storage.

26 Oilfield Review

Natural fractures can occur locally in the Bakken Formation, and where their intensity is sufficiently high, such as along the Antelope anticline in North Dakota, they can affect pro-duction. In general, the fractures are vertical to subvertical, bed bounded and partially to totally filled by quartz, calcite or, rarely, pyrite cements. Some vertical microfractures appear to be expul-sion, or fluid release, fractures that form when fluid pressures exceed the prevailing minimum principal compressive stress, allowing oil to migrate from the source rocks into adjacent res-ervoir members.

The RQ (porosity and permeability) of the middle member, along with the degree of over-pressuring, plays a large role in determining the productivity of the Bakken Formation. The ability to predict where better reservoir quality occurs dramatically increases the chance of success in this play.

For this reason, an E&P company operating in the Williston basin contracted with Schlumberger, whose geophysicists reprocessed a proprietary 3D multiazimuth seismic survey over an area within the Bakken play of North Dakota. The target res-ervoir horizon was in the middle Bakken member. The company wanted to base drilling locations on

patterns of initial production and seismic attri-butes, which are both affected by characteristics of reservoir geology. The company hoped to move away from drilling wells based on geometric pat-terns—lease boundaries or the Public Land Survey System—which ignores geologic heteroge-neity, and to take a deliberate approach to locate, orient and drill infill horizontal wells into highly productive reservoir locations.41

Geoscientists constructed a calibrated geo-logic model that was constrained by all available data, including well logs, borehole image logs and core samples. Geophysicists processed the 3D

> Rocks under stress. Randomly oriented soft—malleable and yielding—fabrics (top left, blue) within a host matrix (tan) may open in any direction in an isotropic stress field; soft fabrics may include pores, kerogen particles and microcracks. Under an anisotropic stress field (top right), such fabrics will be preferentially squeezed in the maximum compressive stress direction (orange arrows) and have their shapes modified less in the other principal stress directions. The N–S oriented maximum compressive stress (σHmax, bottom left) causes incident SW–NE polarized S-waves (gray arrows) to split into N–S polarized fast S-waves (brown arrows) and W–E polarized slow S-waves (gold arrows). In addition, incident P-waves (green arrows) resolve into P-waves that are fastest (red arrows) parallel to the N–S maximum compressive stress and slowest (blue arrows) perpendicular to it; the sinusoid (bottom right) shows the full azimuthal P-wave velocity variation.

NESW

Fast S-wave, N–S

Slow S-wave, W–E

σHmax

σV

σHmin

Isotropic Stress Field Anisotropic Stress Field

Azimuth

Fast P-wave, N–S

Slow P-wave, W–E

Incident P-wave

Incident P-wave

Incident S-wave

P-w

ave

velo

city

East

North South

West

Winter 2013/2014 27

41. Johnson GM and Miller P: “Advanced Imaging and Inversion for Unconventional Resource Plays,” First Break 31, no. 7 (July 2013): 41–49.

For more on the Public Land Survey System: “US Topo Quadrangles—Maps for America.” http://nationalmap.gov/ustopo/ (accessed January 17, 2014).

42. Johnson GM and Dorsey J: “Modeling Overburden Heterogeneity in Terms of Vp and TI for PSDM, Williston Basin, U.S.A.,” Expanded Abstracts, 80th SEG Annual Meeting, Denver (October 17–22, 2010): 4062–4065.

43. For more on OVT processing: Stein JA, Wojslaw R, Langston T and Boyer S: “Wide-Azimuth Land Processing: Fracture Detection Using Offset Vector Tile Technology,” The Leading Edge 29, no. 11 (November 2010): 1328–1337.

44. Johnson and Miller, reference 41.

seismic data to account for horizontal variability and anisotropy of seismic velocities in the strata above the reservoir.42 Seismic processors sorted the seismic data into offset vector tile (OVT) gathers, in which traces share similar source-to-receiver offset and azimuth.43 Using high-resolu-tion, multiazimuth OVT tomography, the processors modeled seismic velocities and anisotropy and used them for prestack depth migration (PSDM) of the OVT gathers. If there was disagreement between seismic picks of for-mation-top depths from PSDM and those from well data, the velocity and anisotropy model parameters were readjusted, and the tomography and PSDM steps were repeated until there was acceptable agreement between the geologic model and PSDM image.

Once the geologic model and PSDM image matched, subsequent processing could focus on the seismic anisotropic effects at middle Bakken reservoir depths that appeared to result from ori-ented geologic fabrics or stress anisotropy (previ-ous page). The geophysicists used the fitted elliptical anisotropy from traveltimes (FEATT) workflow to find the fast and slow compressional-wave (P-wave) velocities and directions at the reservoir level.

The FEATT workflow starts by converting the PSDM OVT gather from depth to two-way travel-time. The analyst or an automated routine picks residual traveltimes across common offset-azi-muth time horizons, converts the traveltimes to interval velocities and fits an ellipse to the veloci-ties. The ellipse’s major and minor axes and their orientations provide estimates of the fast and slow P-wave velocities and directions (right).

Following application of the FEATT workflow, the geophysicists applied amplitude variation with offset and azimuth (AVOAZ) analyses to esti-mate S-wave velocity anisotropy.44 The AVOAZ analysis of S-waves may provide higher vertical resolution of the anisotropy variation than P-wave

> Azimuthal anisotropy. The seismic data were sorted into offset vector tiles (OVTs) and converted to depth by conventional migration (top left) and by anisotropic prestack depth migration (PSDM) and tomography (top right). The latter process reduced the waviness in the data attributable to overburden effects and produced datasets appropriate for azimuthal anisotropy analysis. In both panels, the yellow zigzag line gives the azimuth distribution in the OVT, and offset increases from left to right. The PSDM OVT data (cyan box) were converted from depth to time (bottom left), and a horizon (red) was selected for fitted elliptical anisotropy from traveltimes (FEATT) analysis (bottom right). In this example, the seismic processors selected the minimum number of three points (red) required to fit an ellipse; in practice, processors use many more than three. Processors converted the residual moveout at each azimuth to P-wave velocity (the radius of the radial plot) and fitted a FEATT ellipse (blue points, black points and radii) to the input points. The ellipse yielded a fast P-wave velocity azimuth of 114.24° with a slow-to-fast P-wave velocity ratio of 0.974, or P-wave velocity anisotropy of 2.6%. (Adapted from Johnson and Miller, reference 41.)

0

180

Trav

eltim

eDe

pth

Offset

P-w

ave

velo

city

, m/s

3,000

3,200

3,400

3,600

3,800

Offset

Input pointsFitted pointsEllipse definition

PSDM OVT Gather

Conventional Migration of OVT Advanced PSDM Migration of OVT

0

30

60

90

120

150

180

210

240

270

300

330

Major axis 114.24°Aspect ratio 0.974

Azim

uth,

deg

ree

28 Oilfield Review

velocity anisotropy methods because of its sensi-tivity to interface contrasts rather than to the average cumulative response of overlying strata.45

The present-day Bakken Formation maximum in situ principal horizontal compressive stress direction determined from the hydraulic fracture stimulations is generally NE–SW.46 Vertical natu-ral fractures observed in wells within the area of investigation were oriented NW–SE, in the pres-ent-day minimum in situ horizontal compressive stress direction. The fractures tended to be min-eralized, had permeabilities in the microdarcy to nanodarcy range and were not believed to be contributing to production.47 In addition, the RQ of the Bakken Formation in the area of investiga-tion was poor to fair, which explains the low pro-duction rates.

The team compared the seismic anisotropy results to the first 90 days of production from wells across the field. Areas of low production correlated to those having weak P- and S-wave anisotropy, and areas of high production were associated with strong anisotropy (left). Anisotropy was weak to the west and strong in the east, which helped explain why the eastern part of the field was more productive than the western part. Along with production improve-ment from west to east in the area of interest, the anisotropy orientation changed from NW–SE in the west to NE–SW in the east. An improvement in matrix properties is one explanation for this change; in addition, geophysicists speculate that this change in anisotropy direction represents a change in natural fracture orientation from one side of the field to the other. In the east, NE–SW oriented fractures would be parallel to the pres-ent-day maximum principal stress direction and would tend to be open (left).

In addition, anisotropy appeared strong in close proximity to source and reservoir rock con-tacts. From bottom to top, strong anisotropy occurred around the upper Three Forks–lower Bakken contact, the lower Bakken–middle Bakken contact, the middle Bakken–upper Bakken contact and through the upper Bakken member into the lower Lodgepole Limestone (next page). This result indicates that anisotropy from 3D multiazimuth surface seismic data may be used to delineate the areal and depth distribu-tion of sweet spots and future drilling targets.

> Production sweet spots. A seismic horizon through the middle Bakken member shows the slow-to-fast S-wave velocity ratio derived from AVOAZ inversion. The black arrows represent the relative magnitude of the estimated S-wave anisotropy; the arrow directions provide the fast S-wave vector azimuth from the inversion. The colored circles mark the average location of long horizontal wells and show the initial 90 days of oil production within the mapped area. To the west, the production was low to moderate, and S-wave velocity anisotropy is weak (blue to purple); the fast S-wave direction tends to be NW–SE. In the east, the production was higher, anisotropy is stronger (yellow to red) and the fast S-wave direction has a SW–NE trend, which is consistent with the present-day regional maximum in situ principal compressive stress direction. Initial production tends to be higher where the anisotropy is stronger. Analysts interpret the anisotropy to be associated with production sweet spots that are potential targets for drilling. (Adapted from Johnson and Miller, reference 41.)

3,000

12,000

0

0

m

ft

10,000

20,000

30,000

40,000

50,000

70,000

80,000

60,000

Firs

t 90

days

of o

il pr

oduc

tion,

bbl

Anisotropic Isotropic

0.8 1.0Slow/fast shear velocity ratio

N

> Volumes of high anisotropy. This view of S-wave velocity anisotropy within the middle Bakken member is looking down and north. The orange and red clouds are volumes of low ratios of slow-to-fast S-wave velocity, equivalent to high anisotropy, extracted from 3D seismic data between the upper and lower Bakken members. The anisotropy is strong in the east and south and weaker in the northwest. The blue surface underneath the clouds is from within the lower Bakken member and shows the ant-tracking seismic attribute (black to white), which accentuates traces of faults and fractures. (Adapted from Johnson and Miller, reference 41.)

Slow

/fas

t she

ar v

eloc

ity ra

tio

Anisotropic

Isotropic1.0

0.8

Top of theupper Bakken

Top of thelowerBakken

Top of themiddleBakken

1,500 m4,900 ft 1,5

00 m

4,900

ft

N

45. Hall SA and Kendall JM: “Constraining the Interpretation of AVOA for Fracture Characterization,” in Ikelle L and Gangi A (eds): Anisotropy 2000: Fractures, Converted Waves, and Case Studies. Tulsa: Society of Exploration Geophysicists (2001): 107–144.

46. Sturm and Gomez, reference 38.47. Sturm and Gomez, reference 38.

Winter 2013/2014 29

The Value of Seismic DataThese examples of using surface seismic data to understand patterns of production have been ret-rospective rather than prospective. Operators continue to test and appraise the identified sweet spots with new wells.

An increasing number of operators are acquir-ing and analyzing 3D surface seismic data during the early stages—exploration, pilot and appraisal phases—in the operating cycle of organic shale plays. Suitably analyzed and interpreted seismic data have proved to be invaluable for guiding the

placement of initial wells within a frontier shale basin, appraisal wells within a prospective shale basin and infill wells as part of a field develop-ment program in a mature basin. —RCNH

> Slow-to-fast S-wave velocity ratio near the middle Bakken member boundaries. The S-wave velocity ratio was calculated from an AVOAZ inversion for a pair of crossing seismic sections. The red rectangle (top) shows the middle Bakken reservoir interval displayed in the main figure (bottom left). The vertical black dashed line marks the intersection of the inline and crossline seismic sections and the approximate location of a vertical well. The relative shear velocity ratio within the middle Bakken Formation at this location is higher (blue to purple) in the center and lower (green to yellow) at the formation boundaries, indicating that anisotropy increases from the formation’s center to its boundaries. The log plot (bottom right) shows two tracks. Track 1 (left) displays well log traces of bulk density (pink), P-wave slowness (red), P-wave impedance (blue) and tops of geologic formations and members. Track 2 (right) shows the slow-to-fast S-wave velocity ratio (blue) from the AVOAZ inversion result along the well trace in the main display; it also shows formation and member tops. There is a difference in resolution between the well logs and the inversion result. Locations of tops are sharp and clear in the well log display and not as clear in the inversion display because of surface seismic resolution limitations. (Adapted from Johnson and Miller, reference 41.)

Isotropic

Slow-to-fast shearvelocity ratio

Anisotropic0.820.84

0.86

0.88

0.90

0.92

0.94

0.96

0.981.0

MiddleBakken

30 m

120 m

Bulk Density

P-Wave Slowness

P-Wave Impedance

1.5 g/cm3

μs/ft

3

147

15,700 (ft/s)×(g/cm3) 60,500

24

Isotropic

Slow-to-Fast ShearVelocity Ratio

Anisotropic

Sanish Sand

Three Forks

Sanish Sand

Three Forks

False Bakken

Scallion

Upper Bakken

Middle Bakken

False Bakken

Scallion

Upper Bakken

Middle Bakken

Lower Bakken Lower Bakken

X,235

Dept

h, ft

X,240

X,245

X,250

X,255

Low

erLo

dgep

ole

Bakk

en

30 Oilfield Review

Rotary Sidewall Coring—Size Matters

A formation sample, acquired from downhole, provides a wealth of information

about rock properties not readily available elsewhere. Rotary sidewall coring is

one alternative for acquiring downhole rock samples, but in the past, the small

sample size often limited laboratory evaluation. A new rotary coring tool, with

features that improve sidewall coring operations, addresses core size limitations

by offering larger diameter samples.

Abhishek AgarwalSugar Land, Texas, USA

Robert LarongaClamart, France

Larissa WalkerShell Appalachia Exploration Sewickley, Pennsylvania, USA

Oilfield Review Winter 2013/2014: 25, no. 4. Copyright © 2014 Schlumberger.For help in preparation of this article, thanks to Chad Albury, Ryan Chapman, Lenishan Fernando and Farouk Hamadeh, Perth, Western Australia, Australia; Joe Loman, Dacey McManus and Chris Tevis, Houston; William Murphy, e4sciences, Sandy Hook, Connecticut, USA; Tim Sodergren, Salt Lake City, Utah, USA; and Benjamin Wygal, Natchez, Mississippi, USA.CST, FMI, GPIT, MDT, MSCT, TerraTek, TerraTek HRA, TerraTek TRA and XL-Rock are marks of Schlumberger.HRSCT and RSCT are marks of Halliburton. MaxCOR and PowerCOR are marks of Baker Hughes.

Petrophysics is the branch of geology concerned with determining physical properties and behavior of reservoir rocks. Some petrophysicists specialize solely in interpreting remotely acquired measure-ments in the form of logs; others prefer having a piece of the reservoir rock along with logs. Those with a preference for analyzing rock samples pre-fer large ones over small ones. Using larger rock

samples, geologists and engineers can perform more experiments and more easily determine geo-logic, geomechanical and petrophysical proper-ties. Because of the global proliferation of unconventional resource developments, the option for taking representative samples without the expense and loss of efficiency associated with con-ventional coring has arrived at a crucial time.

Winter 2013/2014 3131

In the oil and gas industry, operators use two main methods to acquire rock samples from the subsurface: cutting whole core with the drilling assembly and taking sidewall samples from the borehole. Sidewall cores can be further divided into percussion and rotary cores. Drill cuttings, pieces of rock ground up by the drill bit and cir-culated back to the surface, are another source of downhole samples, but they may be unreli-able for determining downhole rock properties because of the damage they incur during drill-ing and because of depth uncertainty related to circulation.

Whole cores, also referred to as conventional cores, are continuous sections of reservoir rock cut with a hollow coring bit (above).1 As the bit penetrates strata, a cylindrical section of the rock passes through the bit and remains inside a core barrel, which is part of the drilling bottom-hole assembly. Whole cores are usually cut in multiples of 9-m [30-ft] lengths. In deepwater appraisal wells, it is not uncommon to cut as much as 81 m [270 ft] of core on a single trip in the well. Longer coring intervals come with the added risk of jammed core barrels and damaged core sections.

After an interval has been cored, the drilling crew pulls the pipe back to the surface and retrieves the core barrels; the core barrels can also be extracted from the downhole assembly by mechanical means, which allows the bottomhole hardware to remain downhole for continued coring. Technicians take great care to avoid dam-aging the sample inside the carrier after it is retrieved back to the surface. Conventional coring operations often provide the best rock samples for testing, analyzing and evaluating reservoirs. Compared with normal drilling operations, how-ever, the conventional coring process is time consuming and expensive.

Sidewall cores (SWCs) are plugs of rock taken perpendicular to the wellbore; they are usually acquired by tools attached to wireline. Sidewall cores can be recovered relatively quickly from downhole and may cover multiple zones of interest in the same wireline descent; they provide a cost-effective alternative to conventional cores. One convenience of SWCs is that they are taken after logs have been run, allowing petrophysicists to be selective—they can pick core depths based on interpretations from openhole logs. Sidewall cores also offer an alternative means for petrophysicists

to acquire core data should conventional coring operations fail. Because of their small size relative to conventional core, there is a chance that SWCs taken from a heterogeneous formation may not have properties that are representative of the for-mation at a reservoir scale. The rock from which the SWC is taken may also lack crucial features that geologists need to analyze the reservoir, espe-cially in laminated sand-shale sequences, organic shales and fractured reservoirs.

Two types of sidewall coring are available—percussion and rotary. Percussion coring tools, first introduced in the 1930s, use hollow bullets fired into the formation using an explosive charge. Rotary SWCs are acquired with a tool that uses a small bit to cut plugs from the side of the borehole. Rotary coring tools were devel-oped primarily to address shortcomings of per-cussion SWCs.

This article focuses on the development of sidewall coring and reviews developments in the rotary sidewall coring technique. Case studies

1. For more on whole cores and core analysis techniques: Andersen MA, Duncan B and McLin R: “Core Truth in Formation Evaluation,” Oilfield Review 25, no. 2 (Summer 2013): 16–25.

> Conventional coring. Used with a drilling rig, a bit with an open throat (top left) cuts conventional cores. The cores are retrieved at the surface, boxed (left) and shipped to a laboratory for analysis. At the laboratory, technicians typically cut the core lengthwise (middle), referred to as slabbing, to access representative rock for testing. Tests, such as compressive strength measurements along the red lines shown on the slabbed core, may be performed at this stage. After testing the slabbed core, technicians usually cut plugs for additional evaluation. A biscuit cut, which is a slice of core material (right), is an option to taking core plugs from slabbed sections; these sections of core material provide alternative testing options at the core laboratory. (Photographs courtesy of Tim Sodergren.)

Coring Bit

Boxed Cores

Slabbed Cores

Biscuit Cut

1 in.1 ft

32 Oilfield Review

demonstrate the application of the XL-Rock tool, a recently introduced large-volume rotary side-wall coring service, in an unconventional resource play in the northeast US and a deepwa-ter well offshore Australia.

Getting the CoreThe advent of rotary drilling systems in the 1800s made possible the practice of whole core drilling. The origin of core drilling is attributed to French civil engineer Rudolph Leschot, who mounted diamonds on a hollow, circular drill bit to acquire rock samples.2 The practice of core drilling has remained a mainstay in mineral exploration.

The first coring operation was performed in Pennsylvania, USA, in the 1860s to locate coal seams and measure their thickness.3 Wireline geophysical logging—introduced by Conrad Schlumberger in the 1920s—was originally referred to as “electrical coring,” and was first envisioned as a tool for coal exploration. Geologists consider conventional cor-

ing in oil and gas wells, a technique practiced frequently even today, crucial in the initial phase of field development.

Unfortunately, coring can impact drilling effi-ciency because of the time required to cut and recover whole cores. Depending on the coring objective and cost limitations, E&P companies may deem conventional coring nonessential. In such cases, they may turn to percussion and rotary sidewall coring to gain valuable reservoir rock information. Of the two methods, percussion coring has been the most common; however, in some environments, especially hard rock reser-voirs, deepwater exploration and unconventional resource plays, petrophysicists prefer rotary cor-ing tools. Petrophysicists use SWCs to validate log responses and provide empirical petrophysical and geophysical properties. Core points are determined based on interpretation of logging data, and gamma ray or spontaneous potential (SP) logs are used for depth correlation between openhole logs and core depths.

The first percussion coring tool was intro-duced in the 1930s.4 Today, all major wireline ser-vice companies offer percussion coring tools, also referred to as core guns. Core guns such as the Schlumberger CST chronological sample taker appear similar to the original percussion coring tools introduced almost a century ago; however, improvements to core gun hardware have resulted in the reliable and cost-effective sys-tems in use today.

Core guns have hollow, barrel-shaped bullets mounted on a carrier (above). The bullets are forcefully ejected from the carrier into the bore-hole wall by means of an explosive charge. Each bullet is fired sequentially by application of elec-trical power from the surface after the tool is posi-tioned at the desired zone. Bullets are attached to the gun body by means of steel cables, which facili-tate extraction of cores from the sidewall. After a bullet is embedded in the formation, the wireline operator uses the weight of the gun and the force applied by the logging winch to work the core free.

> Percussion sidewall coring. Core bullets are fired from a core gun (A) using explosive charges behind each bullet. Various types of bullets are available, including combo bullets (B), which are for medium hard to soft formations. A groove around the top of the bullet accommodates a cutting ring (C) that is held in place with a snap ring (not shown). The type of formation dictates which cutting ring is employed. In soft formations, larger cutting rings reduce bullet penetration. The hardened steel, hard rock bullet does not use a cutting ring. Cables attached to each bullet (D) help extricate the bullet from the formation after it has been shot. Guns are retrieved to the surface with core bullets attached (E), and technicians remove core samples and place them in bottles. The bottles are labeled, boxed and sent to a laboratory for analysis. (Photograph of retrieved core gun courtesy of Benjamin Wygal.)

Combo Bullets Hard Rock Bullet

7/8 in. 11/16 in. 11/16 in.

Cutting Rings

0 2in.

Steel cables

A

C

B

D

E

Winter 2013/2014 33

The engineer selects core bullet geometries and explosive charge strength according to the properties of the formation to be cored. Hard for-mations, those with porosity less than 15%, gener-ally require hard rock bullets and larger explosive charges. Cores taken in hard formations tend to shatter on impact from the bullet, potentially resulting in empty core barrels when the gun is retrieved at the surface. Soft formations are eas-ier to sample, although in unconsolidated forma-tions, the barrels may become so deeply embedded that they cannot be extracted. Samples taken from unconsolidated formations have a tendency to wash out of the bullet because of the turbulent effects of the mud on the exposed core barrels as they are retrieved through the mud column. For sampling in unconsolidated formations, the wire-line operator attaches cutting rings to the bullets, which help decrease bullet penetration; the use of smaller explosive charges may also help improve sample recovery.

Cutting rings decrease penetration for bullets shot into soft formations and create a hole that has a larger diameter than that of the bullet; the larger diameter hole facilitates core extraction. The ring is designed to snap off and remain in the newly created hole. If the bullet embeds too deeply in the formation and cannot be worked free, the operator can break the retaining cables by applying tension with the logging winch. Hard formations do not normally require cutting rings because depth of penetration is rarely a concern, and the rings may impede penetration and limit sample length.

After the cores are shot, the gun is pulled to the surface, where field technicians free the core barrels from the carrier by removing the cables; then they separate and sort the barrels. A technician uses a plunger press to push each sample from the barrel into a sample bottle, seals each bottle and marks it with the sample depth. The sealed cores are then transported to a laboratory for analysis. Prior to shipping the cores, operators may use UV lights on location to identify the presence of hydrocarbons in the core samples.

Although percussion core guns offer a cost-effective and rapid means of acquiring samples from the formation, the process lends itself to potential problems. The impact of the bullet with the formation can, and usually does, damage the core.5 Both hard and soft rocks tend to shatter upon the bullet’s impact, which alters the rock properties of the sample. Cores taken in uncon-solidated formations may be compacted by the impact of the bullet, and mudcake from the inner borehole wall may be injected into the rock matrix of the sample, altering rock properties.

Percussion SWCs provide reliable information on grain size distribution, mineralogy, sedimen-tary features and hydrocarbon residuals.6 However, material alterations caused by the impact of the core bullet may distort porosity measurements. Such distortion is well documented, and laborato-ries that specialize in analyzing SWCs have devel-oped empirical relationships to correct for some of these effects; final analyses, however, are often an approximation.7 Permeability is not typically mea-sured in percussion core samples but is empiri-cally derived from porosity and mineralogy.

Taking percussion SWCs may also create operational issues. Because of the potential for leaving cutting rings, broken cables and core bar-rels in the wellbore during coring, many opera-tors attempt no further logs after cores are taken without first making a cleanout, or conditioning, trip. In addition, the bullets may damage the borehole wall, which can affect measurements made by pad contact tools. Debris from percus-sion cores is not much of a concern when cores are taken on the last logging trip in the well because the debris can be ground up and circu-lated out of the well during the cleanout and before casing operations commence.

In many parts of the world, high-angle and horizontal wells are becoming more common than conventional vertical wells.8 In these types of wells, gravity no longer assists the extrication of a core from the sidewall, and conveyance of core guns in these types of wells relies on drillpipe, tractor or coiled tubing. The use of explosives in percussion coring further complicates operations with these conveyance techniques. Despite well-established safety procedures, service companies must consider the fact that percussion SWCs are explosive devices and must be handled with care. Because radio silence is required during parts of the operation, use of explosives in deepwater exploration makes percussion sidewall coring pro-hibitive as well. For these reasons and others, per-cussion sidewall coring is almost never attempted in high-angle and horizontal wellbores or in deep-water applications.

Frequency of percussion sidewall coring var-ies geographically; for many wells in the south-eastern US, geologists take SWCs whenever and wherever logs indicate potential hydrocarbons. In other regions, especially in hard rock reser-voirs where sample recovery is usually poor and properties measured on percussion cores are unrepresentative of actual rock properties, side-wall coring is rarely attempted. Operators and service companies recognized the limitations imposed by percussion sidewall coring but for

almost fifty years accepted these limitations. The situation changed in the 1980s when the first rotary coring tools were introduced.

Rotary Cores to the RescueRotary sidewall coring tools have miniature, dia-mond-tipped drill bits, which are small versions of those used for conventional coring operations (above). Just as various bit designs are available for drilling and conventional coring, engineers can select from a variety of rotary sidewall coring bits based on expected formation and rock type. The bit cuts a round plug of formation material directly from the borehole wall. The tool then snaps off the core and pulls it into a core holding area inside the tool body. Depending on the tool design, this process is repeated until the core catching apparatus is full.

2. Bowman I: “Well-Drilling Methods,” Washington, DC: US Government Printing Office, Water-Supply Paper 257, 1911.

3. Collom RE: “Prospecting and Testing for Oil and Gas,” Washington, DC: US Government Printing Office, August 1922.

4. Leonardon EG and McCann DC: “Exploring Drill Holes by Sample-Taking Bullets,” Transactions of the AIME 132, no. 1 (December 1939): 85–99.

5. Webster GM and Dawsongrove GE: “The Alteration of Rock Properties by Percussion Sidewall Coring,” Journal of Petroleum Technology 11, no. 4 (April 1959): 59–62.

6. Fiedler FJ: “Toward Integrated Formation Evaluation,” Transactions of the SPWLA 29th Annual Logging Symposium, San Antonio, Texas, USA, June 5–8, 1988, paper Q.

7. Fertl WH, Cavanaugh RJ and Hammack GW: “Comparison of Conventional Core Data, Well Logging Analyses, and Sidewall Samples,” Journal of Petroleum Technology 23, no. 12 (December 1971): 1409–1414.

8. Amer A, Chinellato F, Collins S, Denichou J-M, Dubourg I, Griffiths R, Koepsell R, Lyngra S, Marza P, Murray D and Roberts I: “Structural Steering—A Path to Productivity,” Oilfield Review 25, no. 1 (Spring 2013): 14–31.

> Rotary coring bit. A diamond-tipped, circular bit is used for cutting rotary cores from the borehole wall. When the bit reaches its maximum depth, the assembly is canted upward, which breaks the core from the formation. After the core is pulled back into the tool, the operator repositions the tool to cut the next core.

34 Oilfield Review

Rotary sidewall coring offers several advan-tages over percussion sidewall coring. The mechanical distortion of the rock sample caused by the impact of percussion sidewall cores is eliminated when drilling rotary SWCs, and rotary coring preserves the rock pore structure. Unlike for percussion cores, laboratories can perform accurate porosity, permeability and capillary pressure measurements on rotary cores.9 Routine core analysis (RCAL) measurements are signifi-cantly better for rotary cores compared with those performed on percussion cores.10

A significant limitation of older generation rotary coring tools is the diameter and length of the core plugs taken from downhole formations. Insufficient core volume may cause substandard core analysis results. For conventional cores, rou-tine core analysis is performed on plugs or slices selectively taken from a slabbed portion of the core. Geologists routinely take plugs from a whole core at 0.6-m [2-ft] intervals, although variations in lithology as well as formation heterogeneity may require more frequent sampling. Laboratory-cut plugs typically measure about 6.4 cm [2.5 in.] in length by 2.3 to 3.8 cm [0.9. to 1.5 in.] in diam-eter. Although some rotary coring tools can deliver samples comparable in size to plugs cut from conventional cores in the laboratory, cores from older generation rotary coring tools are less than 2.5 cm [1 in.] in diameter (above left).

In addition, early rotary coring tools lacked control over the coring process. One crucial cor-ing parameter, weight on bit (WOB)—the pres-sure applied to the coring bit as the core is being cut—is set at the surface prior to running into the well. If the WOB is set too low, coring time is unnecessarily long; if it is set too high, the bit may stall and stop coring prematurely.

Building on many years of experience with the MSCT mechanical sidewall coring tool, design engineers at Schlumberger began developing a next generation tool. The first limitation they addressed was core size. The traditional 15/16-in. diameter core bit was replaced with a 11/2-in. diameter version, which delivers cores that are similar to industry-standard plugs taken from whole cores in the laboratory. Compared with the smaller diameter cores from the MSCT tool, the larger cores provide more than three times the volume for the same core length. The XL-Rock large-volume rotary sidewall coring service offers three options for core length: standard 7.6-cm [3.0-in.], and optional 8.9-cm [3.5-in.] or 6.3-cm [2.5-in.] lengths (left). The field engineer run-ning the tool can optimize the coring process by adjusting parameters such as WOB in real time.

> Rotary coring tools offered by major service companies.

Core Capacity

50, 44

20, 50, 75

30

36

60

Core Diameter,in.

1.5

0.92

15/16

1.5

1.5

1.0

Core Length,in.

2.5, 3.0, 3.5

2.0

1.75

2.3

2.5

1.8 60

Tool,Service Company

XL-Rock tool,Schlumberger

MSCT tool,Schlumberger

RSCT tool,Halliburton

HRSCT tool,Halliburton

MaxCOR tool,Baker Hughes

PowerCOR tool,Baker Hughes

> XL-Rock rotary coring tool. The XL-Rock tool (left) is lowered on wireline and positioned across from the zone to be sampled. A hydraulic arm anchors the tool at the desired depth; the bit pivots into a horizontal position and then begins coring. Up to 50 cores can be taken during each tool descent. The cores shown (inset) are samples taken from test blocks at the surface.

3 in.

1.5 in.

Winter 2013/2014 35

9. Fiedler, reference 6.10. For more on conventional coring and RCAL: Andersen

et al, reference 1.11. For more on geomechanical measurements on cores:

Cook J, Frederiksen RA, Hasbo K, Green S, Judzis A, Martin JW, Suarez-Rivera R, Herwanger J, Hooyman P, Lee D, Noeth S, Sayers C, Koutsabeloulis N, Marsden R, Stage MG and Tan CP: “Rocks Matter: Ground Truth in Geomechanics,” Oilfield Review 19, no. 3 (Autumn 2007): 36–55.

Because of the larger size of the XL-Rock drill bit, this control contributes to successful cor-ing. Stalling can be reduced, and coring time can be minimized.

Design engineers next looked at upgrading the mechanical components and ruggedizing the electronics of the XL-Rock tool. The redesigned tool offers improved reliability and features greater operator control than did earlier genera-tion tools. Lessons learned developing LWD tools, high-pressure, high-temperature wireline tools and the MDT modular dynamics tester, as well as years working with the MSCT tool, helped engi-neers design a tool that is more robust than previ-ous generation tools.

The standard tool is rated to 177°C [350°F] and 173 MPa [25,000 psi]. An optional version is rated to 200°C [400°F] and 207 MPa [30,000 psi.] At 11.3 m [37 ft] in length, the XL-Rock tool is the shortest sidewall coring tool in the industry. The shorter length greatly reduces the operational risk of sticking the tool while coring. The surface-controlled bit release, which can be actuated in the event the bit becomes stuck, also reduces operational risk.

At each core point, a hydraulic arm anchors the tool in place, and the core bit assembly pivots from its recessed transport position in the tool to a position perpendicular to the tool body. The hydraulically powered rotary bit provides high torque at low rotary speed and is effective in a wide range of rock types. After the full length of the core has been drilled, a core catcher ring holds the core within the bit assembly; the assembly is then canted upward to break the end of the core. For additional depth correlation, a core marker is inserted between samples inside the tool after each core is retracted. Up to 50 samples can be collected per descent (above right).

The larger cores from the XL-Rock tool also enable core laboratories to perform more-advanced geomechanical studies. In the past, because of core-size limitations, these studies were usually reserved for plugs taken from con-ventional cores.11 After the XL-Rock cores have been retrieved, they can be subsampled into miniplugs at angles parallel, perpendicular and at 45° to observed bedding planes to characterize anisotropic mechanical properties (right). The

>Rotary core recovery. After the cores are cut, they are pulled into the XL-Rock tool body and stored in a handling tube. At the surface, the tool operator removes the tube from the tool, separates the cores and seals them inside individual sample bottles (top left). The bottles are labeled and packed for shipping (top right) to the laboratory for analysis. (Photographs courtesy of Chris Tevis.)

> Subsampled XL-Rock cores. Because the 11/2-in. diameter cores from the XL-Rock tool (left) are larger than conventional sidewall cores, they can be subsampled into miniplugs (right) at angles parallel, perpendicular and at 45° to observed bedding planes. Technicians then use these smaller samples to make measurements that characterize anisotropic mechanical properties. The ruler is scaled in inches and decimal inches.

36 Oilfield Review

TerraTek Geomechanical Laboratory Center of Excellence in Salt Lake City, Utah, USA, offers a fast track geomechanics full anisotropic charac-

terization, which is available exclusively for XL-Rock cores. This service provides mechanical properties, including static and dynamic Young’s

modulus, static and dynamic Poisson’s ratio, com-pressional and shear velocities and horizontal stress profiles. Completion quality assessment of shale samples is included in the analysis.

Laboratory RCAL analysis is most often per-formed on traditional cores. The laboratory test-ing of cores retrieved by the XL-Rock tool may offer additional benefits over those for whole cores because the samples can be transported and tested at the laboratory within hours or days of having been cut. Analyses of conventional cores are often conducted many weeks after acquisition, and some rock properties may be affected by the delay. The TerraTek TRA tight rock analysis technique, which is used for shale characterization, can be performed more quickly and efficiently with the larger volume cores. Smaller core volume often requires that multiple samples be acquired at the same depth and com-bined for analysis.

The larger size of the XL-Rock cores also enables laboratory technicians to perform spe-cial core analysis (SCAL) measurements (left).12 The small total pore volume of a 1-in. by 1-in. plug after trimming, typical of older generation rotary coring tools, results in a high uncertainty on saturation measurements. The significantly larger volume of an XL-Rock core reduces the uncertainty by a factor of four. For this reason, 1.5-in. plugs are the industry standard for most SCAL, and the majority of SCAL laboratory equipment is designed to accept this size plug but may not be able to accommodate 1-in. and smaller plugs.

Typical measurements and techniques per-formed on XL-Rock cores include the following:• absolute and relative permeability• mineralogy (X-ray diffraction, scanning elec-

tron microscope image and isotopic analysis)• petrographic description• source rock and oil characterization• log calibration [grain density, porosity at reser-

voir confining pressure, resistivity, m and n exponents, dielectric permittivity, nuclear mag-netic resonanace analysis, T2 cutoffs, total organic carbon (TOC) and acoustic properties]

• reservoir storage and flow capacity• capillary pressure characteristics• formation damage and sensitivity tests.Operators rely on many of these measurements to make decisions regarding reservoir development.

The TerraTek HRA heterogeneous rock anal-ysis service, another core analysis option, is a rock typing workflow that can identify optimal sampling depths using openhole log data.13

> Sample cores. An XL-Rock tool took these core samples from a variety of rock types in the same well. Formations included soft sandstone (top) and dense limestone (bottom). The cores ranged in length from about 1.6 to 3 in. [4 to 7.6 cm] and are of excellent quality for laboratory analysis. (Photographs courtesy of William Murphy.)

Borehole End Rock EndSide View

1.5 in. 3 in.

Winter 2013/2014 37

12. For more on SCAL: Andersen et al, reference 1.13. For more on the TerraTek HRA technique: Suarez-

Rivera R, Deenadayalu C, Chertov M, Hartanto RN, Gathogo P and Kunjir R: “Improving Horizontal Completions on Heterogeneous Tight Shales,” paper CSUG/SPE 146998, presented at the Canadian Unconventional Resources Conference, Calgary, November 15–17, 2011.

14. For more on the Gas Technology Institute (formerly Gas Research Institute) methodology for crushed shale sample evaluation: Guidry FK, Luffel DL and Curtis JB: “Development of Laboratory and Petrophysical Techniques for Evaluating Shale Reservoirs,” Chicago: Gas Research Institute Report GRI-95/0496, April 1996.

Empirical data derived from various tests, and from the TerraTek TRA service, may then be used to optimize the modeled output from the TerraTek HRA service, especially for unconven-tional resource plays.

Unconventional Resource ApplicationOperators have taken various approaches to ana-lyzing samples from shale resource plays. Acquiring conventional cores using a drilling rig is expensive and time consuming, but during the ini-tial evaluation phase, conventional coring may be necessary. Geologists and engineers view rotary cores taken on wireline as a cost-effective alterna-tive, but the small sample size may inhibit exten-sive petrophysical analysis and full anisotropic mechanical properties studies. Also, limited sam-pling in heterogeneous formations may produce samples that are not representative of the majority of the rock. These were a few of the situations faced by Shell Appalachia Exploration in north-central Pennsylvania, USA, while developing a Marcellus Shale play (right).

To evaluate shale cores, Shell geologists fol-low the Gas Technology Institute protocols for crushed shale analysis.14 Typical core evaluation includes source rock evaluation, porosity, per-meability and saturation, in addition to X-ray diffraction. To perform all of these measure-ments requires a minimum sample mass of 200 g [0.25 lbm] (right). Previous attempts to per-form these techniques on rotary cores had required consolidating multiple cores from around the same point or nearby points to have a sufficient sample volume. Reservoir heteroge-neity observed in the Marcellus Shale had resulted in some of these combined samples introducing uncertainties in measured proper-ties, even if the cores were separated by as little as 0.15 m [0.5 ft]. Because of the need for a sample size of large volume, Shell turned to the XL-Rock rotary coring tool to acquire sufficient core material to permit single-sample recovery and analysis.

>Marcellus Shale exploration. Several operators are currently exploring the Marcellus Shale (blue) in the eastern US. Production from this prolific formation has exceeded experts’ estimates, and as of October 2013, the US Energy Information Administration estimated production at 12 Bcf/d [340 million m3/d].

km0 300

0 mi 300

Pennsylvania

Marcellus Shale

UNITED STATES

> Sample size masses. Conventional cores are typically taken in the exploration phase of field development; however, conventional coring can be expensive. Because core data are important for calibrating openhole logging measurements and determining rock properties, many operators rely on SWCs. Shell Appalachia Exploration geologists determined that a minimum of 200 g was required to ensure sufficient quantity of rock for proper analyses. Assuming a 2-in. [5-cm] core length cut with a 0.92-in. [2.34-cm] diameter bit, typical of older generation rotary coring tools, the operator would have needed four cores per depth to obtain 200 g (left, pink). A single 1.5-in. by 3-in. [3.8-cm by 7.6-cm] core, which can be cut with the XL-Rock tool, would provide at least 200 g of sample (green). The photograph (right) illustrates the size difference between a single XL-Rock core and four cores cut with an older generation 0.92-in. bit.

0.92-in. DiameterCore Mass, g

Core Length, in. 1.5-in. DiameterCore Mass, g

261.0 69

311.2 82

361.4 96

411.6 110

461.8 123

522.0 137

2.2 151

2.4 165

2.6 178

2.8 192

3.0 206

38 Oilfield Review

In one recent exploration well, the Shell geologist chose 100 core points, which required two coring trips. From the 100 sample points, 96 high-quality samples from the shale reservoir section were brought to the surface (above). The average time for drilling each core was 5.3 min.

Another potential challenge with sidewall sam-pling is determining whether the sample is repre-sentative of the surrounding resource rock. For whole cores, geologists in the laboratory have the luxury of visually identifying the best rock for tak-ing plugs. The points for taking rotary cores cannot be visually selected in advance unless an image log is run prior to cutting cores; however, the rock

section from which the rotary cores have been acquired can be verified after core acquisition.

After the cores have been cut and retrieved from the well, the operator may elect to run an FMI fullbore formation microimager to deter-mine the exact depth of core acquisition.15 The diameter of a retrieved XL-Rock core is 3.8 cm, but the bit has a larger diameter and leaves a 6.35-cm [2.5-in.] hole in the borehole wall. These holes can be clearly seen in an FMI image log (right). Geologists are able to correlate the core depths with openhole log data. As a standard practice, many operators who employ the XL-Rock service use the FMI tool after rotary cor-ing runs. The XL-Rock tool is combinable with

directional tools such as the GPIT general pur-pose inclinometry tool, which measures tool ori-entation at the core point.

In addition to petrophysical evaluation, Shell Appalachia is currently conducting geomechani-cal studies of the cores to determine rock elastic properties. Shell engineers have included multi-ple in situ stress, or minifrac, well tests in well evaluation programs using the MDT tool.16 Data from these tests and mechanical properties from core evaluations are used in fracture stimulation designs and drilling optimization.17

>Marcellus Shale cores. These four cores, acquired in the Marcellus Shale using the XL-Rock tool, are examples of the high-quality rock samples taken during rotary coring operations. (Photographs courtesy of Larissa Walker.)

1.5 in.

> Core depth confirmation. This FMI image log run after the rotary coring operation clearly indicates where cores were taken. Geologists can use this information to confirm that samples are representative of the formation being examined.

FMI Image

Core point

Winter 2013/2014 39

15. Adams J, Bourke L and Buck S: “Integrating Formation MicroScanner Images and Cores,” Oilfield Review 2, no. 1 (January 1990): 52−65.

16. For more on in situ stress testing and minifracs: Desroches J and Kurkjian AL: “Applications of Wireline Stress Measurements,” SPE Reservoir Evaluation & Engineering 2, no. 5 (October 1999): 451–461.

Carnegie A, Thomas M, Efnik MS, Hamawi M, Akbar M and Burton M: “An Advanced Method of Determining Insitu Reservoir Stresses: Wireline Conveyed Micro-Fracturing,” SPE paper 78486, presented at the Abu Dhabi International Petroleum Exhibition and Conference, Abu Dhabi, UAE, October 13–16, 2002.

17. Laronga R, Tevis C, Kaiser B, Lake P and Fargo D: “Field Test Results of a New-Generation Large-Bore Rotary Coring Tool,” Transactions of the SPWLA 52nd Annual Logging Symposium, Colorado Springs, Colorado, USA, May 14–18, 2011, paper TTT.

18. For more on geomechanics and UCS: Cook et al, reference 11.

Deepwater CoringDeepwater exploration and development projects introduce considerations for coring operations that may not be applicable to other drilling envi-ronments. The high hourly costs for rigs capable of drilling in deep water necessitate efficient and reliable operations. Conventional coring may not be economically viable in these situations. Percussion cores can be acquired quickly and provide selectivity, but the quality of cores offers operators little more than basic rock property analysis. If coring objectives can be met, selec-tively sampling with large-diameter rotary coring tools offers operators an alternative to conven-tional coring.

The ability to acquire plugs as large as those taken from whole core affords the opportunity to collect high-quality rock samples and perform a greater variety of petrophysical and rock analy-ses. By using rotary coring tools to acquire sam-ples, petrophysicists can have confidence that mechanical damage to the sample that might affect measured properties is minimized.

The reliable acquisition of representative rock samples ultimately determines the practicality of replacing conventional coring with rotary cores. Reliability includes both properly functioning equipment and the ability to cut, recover and bring cores to the surface. In some rock types, recovering samples can be difficult regardless of the method used. Many deepwater reservoirs are in sandstone formations of varying strength. Acquiring rotary cores in rocks with unconfined compressive strength (UCS) less than 1,000 psi [6.9 MPa] is especially challenging.18 This was one of the concerns BHP Billiton Petroleum faced when developing a coring strategy for evaluating a deepwater well offshore Australia.

BHP had successfully tested the XL-Rock cor-ing tool in a deepwater Gulf of Mexico well and achieved good recovery. The objectives of this

earlier coring program, which were achieved with the large cores, involved several types of studies, including sedimentology, biostratigraphy, geome-chanical properties analysis and validation of interpretations of petrophysical logs.

For the deepwater well offshore Australia, the coring objectives were similar to those of the Gulf of Mexico wells, and traditional rotary cores were deemed too small to provide representative sam-ples for analysis. Because of the successful acqui-sition of large cores in the Gulf of Mexico, BHP geologists and engineers elected to use the XL-Rock tool and acquire 1.5-in. [3.8-cm] by 3.0-in. [7.6-cm] cores.

Geologists interpreted the petrophysical log data and selected core points based on the coring objectives, which included routine core analysis, geomechanical studies, geochemistry testing and biostratigraphic studies. They then used acoustic data from the well to estimate UCS. These data were incorporated in the decision process for determining rotary core depths. The sampling program covered several formation types, and samples were taken in rocks with a UCS range of 500 to 5,000 psi [3.4 to 34.4 MPa].

Even in the challenging low-UCS intervals, the coring operation achieved 100% recovery (above). The ability to acquire large-diameter cores led to a revision of the coring program; BHP eliminated some originally planned conventional coring. Large-diameter core recovery and opera-tional efficiencies compared to those in conven-tional coring resulted in significant cost savings without sacrificing coring objectives.

> Core recovery in a challenging deepwater environment. During coring operations in a deepwater well offshore Australia, rotary cores were taken with the XL-Rock tool from rocks with UCS values ranging from 500 to 5,000 psi and from several formation types. The operation resulted in 100% recovery of the cores attempted; these examples are representative of the quality of the samples taken.

1 in.

Routine Coring OperationsRotary coring tools were introduced to overcome some of the limitations of percussion cores, but geologists wanted still more core volume. Early reports indicate that large-diameter rotary cores provide a viable alternative for geologists who have had to accept smaller cores provided by con-ventional sidewall coring tools or for operators that found conventional coring operations too inefficient and expensive.

The option of taking representative rock sam-ples without the expense and loss of efficiency of conventional coring operations has arrived at a crucial time in the industry. Geologists and engi-neers have the option to use large cores to evalu-ate unconventional resource plays and also determine mechanical properties in consolidated formations. With more core material for analysis, operators’ chances for successful development of these plays are greatly improved. —TS

40 Oilfield Review

A New Platform for Offshore Exploration and Production

Accurate data are essential for developing climate models and weather forecasts

used in planning offshore E&P operations. A new remotely controlled, autonomous

marine vehicle has been developed to carry a variety of sensors for conducting

detailed meteorological and oceanographic surveys across vast distances and under

extreme conditions. The role of this new sensor platform is expanding to support an

even broader range of missions.

Peter CarragherRose & Associates, LLPHouston, Texas, USA

Graham HineLiquid Robotics, Inc.Sunnyvale, California, USA

Patrick Legh-Smith Gatwick, England

Jeffrey MayvilleRod NelsonHouston, Texas

Sudhir PaiLiquid Robotics Oil & GasHouston, Texas

Iain ParnumCurtin UniversityPerth, Western Australia, Australia

Paul ShoneChevron Energy Technology CompanyLondon, England

Jonathan SmithShell Exploration and Production CompanyHouston, Texas

Christian TichatschkeTotal E&P Uruguay BVMontevideo, Uruguay

Oilfield Review Winter 2013/2014: 25, no. 4. Copyright © 2014 Schlumberger.For help in preparation of this article, thanks to Joanne Masters and Jona Steenbrink, Liquid Robotics Inc., Sunnyvale, California.DART is a mark of the US National Oceanographic and Atmospheric Administration.Wave Glider is a registered trademark of Liquid Robotics, Inc.

The oceans cover more than 70% of the Earth’s surface and have played a dominant role in its geologic history. Although the oceans contain a substantial portion of our planet’s natural resources, their depths remain largely unex-plored. Long-term monitoring over vast expanses of ocean may lead to better understanding of processes that continue to shape the planet while helping scientists discover new resources and predict the impact of ocean forces that could disrupt commerce or alter the course of everyday life.

Forces of nature, such as hurricanes and typhoons, pose a recurring threat to thousands of communities along the coast; earthquakes and tsunamis occur less frequently but often cause more damage.1 Sweeping events and weather pat-terns influenced by the ocean not only menace coastal dwellers but also impact industry and commerce around the world. The oil and gas industry feels the effects of weather in seasonal demand fluctuations. In the offshore environ-ment, the effects of weather translate into con-cessions that operators must make: Is it prudent to mobilize a drilling rig, are the waves too high to offload equipment, or are the winds too strong for helicopter operations? Meteorological and oceanographic, or metocean, data—especially height of waves and period of swells, speed and direction of wind and surface or subsurface cur-rents—provide crucial input for planning rig moves and placement. Geophysical survey crews must assess the effects of tides and currents on

the feathering of seismic streamers as they are towed through the water.2 Wave height is a key parameter used in designing production plat-forms, and pipelines must be installed to with-stand subsea currents. Ocean monitoring plays an integral role in risk assessment and manage-ment by providing information that helps fore-casters, planners and field personnel assess the degree to which they must accommodate forces of nature.

But monitoring is often a costly proposition. Conventional sensor platforms such as buoys, ships, aircraft and satellites are expensive and require extensive lead time for planning, procure-ment and construction. Personnel to support these platforms and their missions must also be trained and managed. Satellite-mounted sensors and storm-chaser aircraft evaluate the air column and ocean surface but are limited in their on-scene endurance, real-time sampling data rates and capability to measure conditions at or beneath the sea/air interface. Oceanographic ships can range over great distances while taking a variety of mea-surements, but vessel and crew are not meant to withstand extreme conditions and also must return to port for replenishment after a limited time. Ocean-observation buoys can also be outfit-ted with sensors but are anchored in place, so they measure conditions within only a relatively fixed location.3 The cost to build, deploy or crew a met-ocean survey platform often starts in the millions of dollars and increases with the intricacies, risks or ambitions of the mission.

Winter 2013/2014 4141

One complement, and in some cases alterna-tive, to satellites, planes and ships is an unmanned mobile sensor platform for monitoring ocean con-ditions. This concept is part of a progression that led to development of remotely operated vehicles (ROVs), which have become essential inspection and intervention devices for deepwater oilfield operations.4 With one or two skilled pilots at the surface, the ROV can wield the tools and power to carry out complex tasks in a forbiddingly dark, cold and high-pressure environment. Some ROVs even-tually dropped their command and control umbili-cals to take commands through subsea telemetry; now autonomous underwater vehicles (AUVs) are routinely used in subsea surveys. These unmanned vehicles have helped expand the envelope of deep-water operations and have been instrumental in

increasing productivity and safety in one of the most hostile environments on Earth. These vehi-cles, however, require support from the surface.5

The Wave Glider autonomous marine vehicle (AMV), developed by Liquid Robotics, Inc., is a hybrid sea-surface and underwater vehicle that has taken the concept of autonomy beyond that of the AUV.6 This wave-powered sensor platform enables collection and transmission of data gathered at sea on missions lasting up to a year. It is capable of crossing thousands of kilometers of ocean to gather oceanographic data, taking meteorological read-ings while maintaining a stationary position, or cir-cling a rig at a preset distance to provide early warning of security or environmental threats.

Once deployed, it uses no crew, requires no fuel and produces no emissions, thus eliminating

both risk to personnel and impact on the environ-ment. For much less than the cost of a moored buoy or a vessel and crew, the Wave Glider vehicle provides mobility and long-range endurance for extended ocean monitoring missions. It has already carried out hundreds of missions ranging from the Arctic region to Australia and from the Canary Islands to Loch Ness in Scotland.

This article discusses the development of this multimission, autonomous sensor platform and describes its applications—from measuring met-ocean parameters to detecting oil seeps. Examples from the Gulf of Mexico and other areas demon-strate how persistent, unmanned mobile monitor-ing platforms have proved beneficial to offshore exploration and production efforts.

1. Bunting T, Chapman C, Christie P, Singh SC and Sledzik J: “The Science of Tsunamis,” Oilfield Review 19, no. 3 (Autumn 2007): 4–19.

2. Feathering is the lateral deviation of a seismic streamer away from its intended towing direction as marine currents push the streamer off course.

3. The exact diameter of that fixed location is defined by the watch circle of the buoy’s anchor system, which is a function of the length of chain attaching the anchor to the

buoy. To withstand extremes in tides and wave height, the buoy is anchored with steel chain whose length is typically three to five times the water depth. Although this extra chain serves to reduce shock loading on the ground tackle used to anchor the buoy, it also means that the exact position of a buoy will vary with the tides, winds and currents.

4. For more on ROVs in deepwater applications: Downton G, Gomez S, Haci M, Maidla E and Royce C: “Robots to the Rescue,” Oilfield Review 22, no. 3 (Autumn 2010): 16–27.

5. Manley JE and Hine G: “Persistent Unmanned Surface Vehicles for Subsea Support,” paper OTC 21453, presented at the Offshore Technology Conference, Houston, May 2–5, 2011.

6. In 2012, Liquid Robotics, Inc. and Schlumberger created a joint venture known as Liquid Robotics Oil & Gas to extend autonomous marine vehicle services to the oil and gas industry.

42 Oilfield Review

Vehicle DesignThe Wave Glider AMV uses wave energy for thrust, while solar energy powers its rudder motor, navi-gation system and payload electronics. This AMV consists of a surface float and a submerged glider connected by an electromechanical umbilical (above). Each of these parts can support an array of sensors to create a custom payload for each mission. The float weighs about 68 kg [150 lbm] including a typical payload.

The float measures 208 by 60 cm [82 by 24 in.]. Its deck supports antennae for GPS, satellite com-

munications and collision avoidance systems, as well as a mast to support a position marker light and flag for increased visibility. Its surface also holds two photovoltaic panels that continually replenish the lithium-ion batteries used to power the vehicle’s navigation, communication systems and sensor pay-loads. Seven smart battery packs housed within the float are each electrically isolated with separate discharging and monitoring circuitry that permits only two batteries to be in use at a time.7 Two pay-load bays support a total of 18 kg [40 lbm] of sensors and equipment.

The umbilical, about 5.8 m [19 ft] long, pro-vides a flexible connection between the surface float and submerged glider. This line also serves as a conduit for transmitting power and steering commands to the glider.

The submerged glider, or sub, is 2 m [6.5 ft] long. The sub glides on six pairs of underwater wings that propel the entire Wave Glider system forward. The sub frame supports a rudder and its control package. The frame weighs about 68 kg and can support a variety of sensors.

The low-profile surface float, high-strength umbilical and sturdy sub allow the vehicle to carry on through high winds and waves of the open ocean. The sub is sheltered from surface weather conditions and acts as a drift anchor to counter the effects of wind and wave on the sur-face float. The current model, the Wave Glider SV2 platform, has survived five hurricanes and three tropical cyclones and has logged more than 560,000 km [300,000 nautical mi] since 2009.

Ocean LocomotionThe Wave Glider propulsion system is passive and mechanical; it converts energy from wave motion into thrust.8 This propulsion system exploits the natural difference in wave motion between the surface float and the submerged glider. Articulating fins, or wings, attached to the sub convert wave energy to generate more than 1.3 kN [300 lbf] of thrust as they pivot vertically. The vehicle produces forward thrust indepen-dent of wave direction as its float moves up and down with each wave and the sub tows the float forward (next page, top).

Forward speed is dependent on the overall buoyancy force provided by the float when teth-ered to the weight of the sub. The vehicle’s mass and buoyancy vary with payload, so the float, umbilical and sub must be balanced and tuned to provide optimal propulsion performance. The AMV is designed to operate in variable condi-tions, ranging from sea state 0 to state 6 (left).

The vehicle can achieve speeds up to 1 m/s [2 knots] and in typical wave conditions of 0.3 to 1 m [1 to 3 ft] reaches 0.5 to 0.75 m/s [1 to 1.5 knots].9 At this rate, it is able to travel about 1,000 km [620 mi, 540 nautical mi] in a month. It can also harvest energy from high-frequency, low-ampli-tude waves—such as wind ripples—so that even under calm conditions, its speed rarely drops below 0.25 m/s [0.5 knots].10

This AMV has demonstrated its capability to perform in extreme sea states. One Wave Glider vehicle, designated G2, experienced a close brush with Hurricane Isaac in August 2012. The storm

>Wave Glider system design. This autonomous marine vehicle is divided into three primary subsystems: surface float, umbilical and submerged glider. Each subsystem can be configured to meet client needs.

Aft payload bay

Payload electronics

Power andcommunications

Solar panel

Rudder

Payload electronics

Command and controlelectronics

Location marker light

Weather station

Lift point

Forward payload bay

Surface Float

Umbilical

Submerged Glider Subsea payload attachment points

Wings

> Table of sea states. The World Meteorological Organization categorizes the force of progressively higher seas according to wave height. The Wave Glider AMV can operate in conditions up to sea state 6.

0123456789

00 to 0.10.1 to 0.50.5 to 1.251.25 to 2.52.5 to 44 to 66 to 99 to 14More than 14

Glassy calmRippledSmooth or with waveletsSlightModerateRoughVery roughHighVery highPhenomenally high

Sea State Wave Height, m Ocean Surface Characteristics

Winter 2013/2014 43

passed within 100 km [60 mi] of G2’s location in the Gulf of Mexico. When the hurricane veered toward the vehicle, its pilot—who monitored the situation from the operations support center (OSC) in Sunnyvale, California, USA—issued a course change that took the vehicle out of danger. Outfitted with sensors to measure water speed, air and water temperature, wind speed and baro-metric pressure, G2 transmitted data despite its proximity to the storm (right). More recently, in October 2012, a different Wave Glider AMV suc-cessfully piloted through 130-km/hr [70-knot]

winds to transmit weather data in real time as Hurricane Sandy traveled northward along the US eastern seaboard.11 In stormy conditions, the

vehicle’s performance is boosted by increased wave energy, which allows it to maintain its intended course.

7. Pai S: “Wave Glider—Introduction to an Innovative Autonomous Remotely Piloted Ocean Data Collection Platform,” paper SPE 166626, presented at the SPE Offshore Europe Oil and Gas Conference and Exhibition, Aberdeen, September 3–6, 2013.

8. Leroy F and Hine G: “Persistent Unmanned Surface Vehicles for Well and Field Support,” paper OTC 22545, presented at the Offshore Technology Conference Brazil, Rio de Janeiro, October 4–6, 2011.

9. A knot, or nautical mile per hour, is equivalent to 1.151 statute mile/h [1.852 km/h].

10. Dalgleish FR, Ouyang B, Vuorenkoski AK, Thomas JC and Carragher PD: “Towards Persistent Real-Time Autonomous Surveillance and Mapping of Surface Hydrocarbons,” paper OTC 24241, presented at the Offshore Technology Conference, Houston, May 6–9, 2013.

11. Pai, reference 7.

>Wave propulsion. The Wave Glider system converts a portion of its vertical movement into forward thrust. As the surface float rises on the crest of a wave, it pulls the submerged glider upward by the umbilical. The glider’s six pairs of articulated wings are pressed downward as the glider rises, translating the glider’s upward rise into an upward and forward motion, which pulls the float forward (middle). As the float moves off the wave crest, the glider wings tilt upward, which again translates into forward motion (right). Wave motion is greatest at the water’s surface and decreases with depth. The magnitude of forward propulsive force is proportional to the difference between the ocean wave amplitudes at the surface float and at the submerged glider wings.

Waveamplitude

> Sensor readings from a storm. As Hurricane Isaac veered toward the Wave Glider G2, the AMV’s sensors recorded a dramatic drop in water temperature with sustained winds of 40 knots [74 km/h], gusting to 74 knots [137 km/h] as barometric pressure fell to 988.3 mbar [14.3 psi].

10

8

6

4

2

0

Sign

ifica

nt w

ave

heig

ht, m

1,020

1,010

1,000

990

980

970

Baro

met

ric p

ress

ure,

mba

r

34

32

30

28

26

24

Sea

surfa

ce te

mpe

ratu

re, °

C

100

80

60

40

20

0

Win

d sp

eed,

kno

ts

Time Time

Time Time

44 Oilfield Review

The Wave Glider propulsion system also allows it to hold station on one location, even in tidal or eddy currents. It simply steers to a desig-nated waypoint—a programmed set of geo-graphic coordinates. When it approaches the limits of a predefined watch circle, it turns around and heads back to the same point repeat-edly. The AMV can maintain station for long dura-tions with a watch circle diameter down to 50 m [164 ft], depending on currents and sea state. By contrast, the mooring systems of deep-ocean buoys employ ground tackle that produces a much wider watch circle (above).

The Sensor Platform The Wave Glider AMV can accommodate a broad range of off-the-shelf or custom sensors to fit the needs of the mission. A GPS receiver not only determines vehicle position, it provides a precise time stamp for all data recorded on the mission. Photovoltaic panels keep lithium-ion batteries charged to support WiFi, cellular or satellite com-munications systems, onboard data processing and various payloads.

Additional sensor payloads can be configured according to client specifications:• meteorological sensors to record barometric

pressure, air temperature and wind direction, speed and gusts

• wave sensors to record wave height, period and direction

• acoustic modems to harvest data from sensors mounted on subsea structures or the seafloor

• bathymetry sensors to map water depth• current sensors to record direction and speed • water salinity and temperature sensors• fluorometry systems to detect the presence of

oil, turbidity and chlorophyll in the water• magnetometers to measure the magnitude and

direction of magnetic fields• cameras to provide realtime imaging; also used

to monitor ice proximity or to verify the pres-ence of surface oil sheens

• passive acoustic recorders to detect and ana-lyze marine mammal vocalizations.

Clients can monitor vehicle status and data in real time. An account-based credentialing scheme provides security for communicating with the vehi-cle using the Internet. Updates are generally car-ried out at client-specified intervals ranging from 1 to 15 min. An onboard hard drive records higher resolution sampling rates.

Piloting by Remote ControlThe Wave Glider AMV can be programmed to travel directly from one location to another or to follow a specific route defined by multiple sets of geographic coordinates, or waypoints. The onboard GPS guides the vehicle from one way-point to the next. The vehicle uses a 12-channel

GPS receiver as its primary navigation sensor, along with a tilt-compensated compass with three-axis accelerometers and a water speed sen-sor. This system typically provides navigation accuracy of better than 3 m [10 ft].12

Alternatively, Wave Glider pilots can steer their charges remotely (next page, top). Command and control information is relayed via satellite link with a secure, web-based user interface for direct-ing the units.13 The Wave Glider Management System allows pilots to issue course commands using any Internet-enabled computer or cellular telephone that supports web browsing.14

Collision avoidance is crucial to the success of autonomous vehicle programs. A key strategy for the AMV is to see and be seen so that appropriate steering commands may be executed in time to avoid accidents. A mast, flag and light are typi-cally installed to visually mark the AMV float position. More importantly, the float carries an integrated package of electronics to highlight its position. A radar enhancer produces a distinctive target on the radar screens of approaching ves-sels. A satellite communications system, azi-muthal heading sensor and GPS are linked to an automatic identification system (AIS) for track-ing vessel movement.

Commercial vessels are required to carry radar and AIS (next page, bottom). Automatic interrogation and exchange of position, course and speed data are provided by the AIS, whose data are displayed on the radar screen to help navigators on an approaching vessel track the course of the autonomous vehicle. Reciprocal AIS data are automatically relayed from the AMV to Wave Glider pilots onshore, who also monitor ves-sel traffic and issue AMV steering commands to prevent collisions.

E&P ApplicationsWave Glider sensor platforms are suited to a vari-ety of scientific missions and applications. Its persistence and range allow this AMV to gather time-series data across wide geographic areas, enabling scientific research that was not practi-cal or economical using data gathered from buoys, ships or satellites.

Detection of naturally occurring hydrocarbon seeps is probably the oldest method of oilfield exploration. From a geologist’s perspective, ocean surface oil is a good indicator of more reserves beneath the seabed. Ecologists and oceanogra-phers are also interested in learning how organic carbon from these seeps might affect neighboring benthic and bentho-pelagic environments and the chemosynthetic communities they support.15

> Station keeping capability. An open ocean observation buoy (right) was moored next to a bottom pressure recorder (BPR) on the seafloor to relay data from the BPR to scientists on shore. Although it was moored beside the BPR, winds and currents tended to push the buoy to the southeast quadrant of its 3,400-m [11,000-ft] watch circle. A Wave Glider AMV (left) was tested for its feasibility as a relay station for the BPR data.

Winter 2013/2014 45

12. Pai, reference 7.13. Anderson BS and Beatman L: “Autonomous Surface

Vehicle Operations in the Arctic: Regional Baseline Data Acquisition,” paper OTC 23737, presented at the Arctic Technology Conference, Houston, December 3–5, 2012.

14. Pai, reference 7.15. Dalgleish et al, reference 10.

Biological interactions, mixing and dissolu-tion consume or disperse a portion of the hydro-carbons as they rise through the water column, but some hydrocarbon bubbles or droplets even-tually reach the surface. There, they spread out to form a thin oil patch, or sheen, whose depth and breadth depend on sea surface conditions—particularly wave agitation, temperature and evaporation, which affect the rate of dispersion. These patches occur regularly but are often short-lived. They can be observed visually or detected by satellite-mounted synthetic aperture radar (SAR). However, orbits of SAR satellites typically permit no more than two passes per day over a particular site. Unmanned sensor plat-forms that measure hydrocarbons and other envi-ronmental parameters and transmit the data to shore-based researchers are an effective alterna-tive to satellites or ship-borne measurements.

> Pilot control station. At an onshore operations support center, pilots monitor vessel traffic, sea conditions and AMV operating parameters around the clock.

> Typical AIS display. Vessel position, speed, heading, projected closest point of approach (CPA) in nautical miles (nm) and estimated time to CPA are displayed on an electronic chart overlay. The AIS updates this critical vessel information several times a minute. The vessel on which this display appears (orange, circled) will be passing close by three other vessels (red) if it maintains its present course and speed.

MV Richard Etheridge11.4 knots, 139°CPA 0.20 nm, 8 min

MV Nathan Bruckenthal13.4 knots, 156°CPA 0.14 nm, 2 min

MV Douglas Munro8.7 knots, 318°CPA 0.18 nm, 1 min

Ownvessel

46 Oilfield Review

Wave Glider sensor platforms have been used during a two-month mission in the Mississippi Canyon area of the Gulf of Mexico to evaluate natural oil seeps in the vicinity of salt domes and mud volcanoes. The AMV science payload con-sisted of a float-mounted water speed sensor, a mast-mounted weather station, a fluorometer that measured low concentrations of semivolatile hydrocarbons and two optical sensors that mea-sured concentrations of dissolved and suspended organic material via fluorescence.16 Prior to the AMV deployment, optical sensor response was calibrated to known concentrations of crude oil at various stages of weathering in a wave tank testing facility. The resulting Wave Glider sensor data helped scientists map the location and extent of the natural oil sheens (left).

On the other side of the world, Chevron’s Environmental Technology Unit, in collaboration with the Centre for Marine Science & Technology at Curtin University in Perth, Western Australia, deployed a unique sensor configuration on two Wave Glider sensor platforms.17 The AMVs obtained baseline turbidity data prior to the initiation of dredging operations for a pipeline offshore Australia. Deployed in three sorties, the AMV sen-sor platforms carried out metocean surveys and obtained measurements to assess turbidity through areas affected by the dredging.

During the first sortie, the system obtained a variety of metocean measurements, including the direction and magnitude of ocean currents, air temperature, wind speed and direction, atmo-spheric pressure and water temperature and salinity. These data provided valuable environ-mental baseline information that helped scien-tists plan for subsequent sorties (next page).

The next sortie, conducted to obtain detailed particle suspension data, also demonstrated the towing capability of the Wave Glider sub. An AMV trailed a towfish sensor module behind the sub-merged glider to measure turbidity (left). The towfish measured optical transmission to deter-mine light attenuation and measured backscat-ter at three wavelengths for calculating suspended sediment and mean particle size. Having established a predredge baseline, the AMVs were deployed again to measure suspended sediment during the dredging operation.

The third sortie allowed scientists to com-pare data obtained by towfish sensors during the second sortie with data obtained from a dif-ferent optical sensor to track suspended sedi-ment and particle size distribution. This comparison of results from state-of-the-art sen-sors helped the operator determine the best

> Hexagonal search pattern. Satellite-mounted synthetic aperture radar sensors detected a sheen (green) resulting from a seep in the Gulf of Mexico. During a Wave Glider sortie, the sensor vehicle encountered increased hydrocarbon concentrations. The Wave Glider trajectory is color coded to correspond to hydrocarbon concentration. Detected events, (large dots) in which spikes or sharp transitions are registered on multiple sensors, show where the AMV encountered higher concentrations of semivolatile hydrocarbons, thus indicating fresh accumulations.

Latit

ude

28.132°

28.130°

28.128°

28.126°

28.124°

28.122°

28.120°

28.118°

–89.148° –89.146° –89.144° –89.142° –89.140°

Longitude–89.138° –89.136° –89.134° –89.132°

Equi

vale

nt h

ydro

carb

on c

once

ntra

tion,

μg/

L

0.249

0.258

0.268

0.220

0.230

0.239

> AMV with towfish. This sensor platform obtained baseline measurements of particle suspension in the water column over a proposed pipeline dredging route. The AMV was configured to obtain surface weather measurements and current speed and direction, along with dissolved oxygen and water conductivity, temperature and pressure. Sensors on the towfish obtained turbidity measurements.

Water speedsensor

Current profiler

Weather station

Backscattermeter Water conductivity, temperature,

pressure and dissolvedoxygen sensorsTransmissometer

Towfish

Winter 2013/2014 47

16. Dalgleish et al, reference 10.17. Pai S and Shone P: “Remotely Piloted Ocean Vehicles to

Conduct METOC and Turbidity Pre-Site Survey,” paper presented at the 75th EAGE Conference and Exhibition, London, June 10–13, 2013.

sensor system for future deployments. A final survey will be conducted after dredging is com-plete. These time-lapse surveys will enable sci-entists to compare profiles before, during and after dredging to evaluate any short- or long-term impacts on the environment.

The AMV has also helped geophysicists design seismic surveys. Seismic vessels employ several acoustic streamers, towed in parallel, to acquire geophysical data. These streamers, thousands of meters long, do not always follow directly in line

behind the seismic vessel; instead, they drift lat-erally in response to the tides and currents they encounter. Although the streamers are steerable, this feathering can produce gaps in data coverage over an area and force the seismic vessel to steam back over that area to reacquire and infill miss-ing data. To counter the effects of tide and cur-rent, survey planners often orient surveys in line with the direction of the predominant current.

Streamer feathering becomes a bigger problem when surveying close to fixed objects such as buoys,

> Current and depth. Offshore Western Australia, a Wave Glider AMV recorded seafloor soundings down to 60 m [200 ft], along with current speed and direction. Tidal influence on current direction is pronounced over the shallower depths, with direction changing at about six-hour increments (red and blue, top). Currents showed irregular variations in speed along the survey path (bottom). All measurements are tied to GPS time and coordinates.

Dept

h, m

Elapsed time, h

–20

0 12 24 36 48 60

–40

Dept

h, m

Elapsed time, h

–20

0 12 24 36 48 60

–40

Current azimuth

180° 360°0°

Current velocity, m/s

.25 .500

drilling rigs or production platforms. In support of a WesternGeco seismic vessel operating in the Gulf of Mexico, three Wave Glider sensor vehicles were deployed to report real-time weather and cur-rent data in the vicinity of rigs and platforms in the survey area. Each AMV used an acoustic Doppler

48 Oilfield Review

current profiler (ADCP) to measure current speed and direction. The data were sent via secure Internet service to the party chief aboard the seis-mic vessel WG Columbus (left). This information helped the seismic survey party chief determine how closely the vessel could pass obstructions while avoiding streamer entanglement.18

In a similar case, Total used ADCPs to aid in designing seismic surveys offshore Uruguay. There, geophysicists sought to survey an area near the confluence of two ocean currents. To adapt the acquisition to the prevailing currents on a day-to-day basis and hence to increase oper-ational security, Total deployed a Wave Glider AMV to measure current strength. The data were transmitted in real time via satellite as the survey was underway.

Wave Glider AMVs can also provide a persis-tent platform to facilitate communication with subsea sensors and equipment via acoustic modem, either for operational control or to assess subsea assets (below left). Shell has used Wave Glider acoustic modems in benchmark tests to harvest data from subsea pressure monitoring transponders in the Gulf of Mexico. In most cases, such data can be recorded, transferred via satellite and analyzed anywhere in the world.

Beyond the Oil FieldEvents of the past decade highlight the devasta-tion visited upon coastal communities as a result of offshore earthquakes or major storms. To warn communities of impending danger, scientists need relevant data on a real-time basis. In the case of tsunamis, sensors deployed on buoys can help locate the epicenter of an earthquake and measure the magnitude of seafloor displace-ment. An array of ocean data buoys has been set up to monitor such data. The US National Oceanic and Atmospheric Administration (NOAA) monitors data from the DART (Deep-ocean Assessment and Reporting of Tsunami) network, established to detect tsunamis and acquire data for real-time forecasts. NOAA cur-rently has 39 DART monitoring stations in its network, and stations from other nations con-tribute data as well. Each DART station consists of a seafloor bottom pressure recorder (BPR) with a surface buoy anchored next to it. An acoustic link transmits data and commands between the buoy and the BPR, which collects pressure and temperature readings at 15-s inter-vals. The data are relayed from BPR to buoy, then transmitted by communications satellite to tsu-nami warning centers around the world.19

> Current data. Current velocity across a survey area was transmitted to the WG Columbus (inset) to aid in predicting seismic streamer positions as the vessel passed close to a production platform and other potential obstructions.

0.0 1.0 2.0 3.0 4.0 5.05.0 4.0 3.0 2.0 1.0Cross track current velocity relative to ship’s heading, knots

Port Starboard

06:00, Mar 24

00:00, Mar 24

18:00, Mar 23

12:00, Mar 23

> Communications gateway. In support of oilfield operations, the Wave Glider AMV will provide a useful link for relaying data and commands between the seabed and the operator’s facilities. In this example, the AMV can handle communications with a submerged AUV, ROV and subsea manifold as well as platforms, a satellite and surface support craft (from Manley and Hine, reference 5). (Copyright 2008, Offshore Technology Conference. Reproduced with permission of OTC. Further reproduction prohibited without permission.)

Wave Glider AMV

Satellite

Manifold

ROV

AUV

Winter 2013/2014 49

18. Pai, reference 7.19. Manley and Hine, reference 5. For more on the DART system: “Deep-Ocean

Assessment and Reporting of Tsunamis (DART) Description,” NOAA National Data Buoy Center, http://www.ndbc.noaa.gov/dart/dart.shtml (accessed November 20, 2013).

Green DS: “Transitioning NOAA Moored Buoy Systems From Research to Operations,” in Proceedings of the MTS/IEEE OCEANS 2006 Conference, Boston, Massachusetts, USA, September 15–21, 2006.

20. Dropsondes obtain a vertical profile of conditions as they fall by parachute through a storm. Temperature, humidity, pressure, wind speed, wind direction and GPS coordinates are transmitted to the launch aircraft every 0.5 s. A drop from 7,000 m [20,000 ft] can take 7 min.

For more on dropsonde capabilities: “GPS Dropsonde,” University Corporation for Atmospheric Research and National Center for Atmospheric Research Earth Observing Laboratory, https://www.eol.ucar.edu/node/3145 (accessed December 6, 2013).

NOAA scientists recognized that there would be operational challenges in maintaining some of the DART stations following their deployment. When a station experiences a failure, the cost of mobilizing a vessel to effect repairs can strain the program’s budget. To augment the network, NOAA has deployed a Wave Glider AMV with a low-frequency acoustic modem to obtain real-time tsunami observations. This unmanned mobile tsunameter serves as a communications gateway for transmitting live seismic data from the seafloor to the ocean surface then relaying the data to shore via satellite (above). The AMV, which also collects real-time meteorological information, can be programmed to travel to select locations or return to shore on command.

Accurate storm forecasting is also critical to protecting lives and assets of coastal communi-ties. Having developed tools to predict the gen-eral track a storm might take, NOAA now seeks to improve predictions of storm intensity. Along the US gulf and east coasts, the greatest threats come from hurricanes. In an effort to better under-stand how hurricanes gain or lose strength, NOAA is targeting the sea/air interface, where warm ocean waters transfer heat energy to the overlying storm system. Weather experts believe

that temperatures beneath the ocean’s surface may contribute significantly to this energy exchange as storm winds and waves churn the waters below the surface.

However, extracting data from the center of a hurricane can be difficult. Storm-chaser air-craft fly into these violent weather systems at several thousand feet above the ocean. They probe the storm using radar to gauge conditions at the ocean surface or they drop sensors from the plane to obtain a detailed vertical profile of atmospheric conditions inside the storm.20 Satellites observe surface temperatures from hundreds or thousands of miles above the ocean, but these measurements may be obscured by cloud cover, and they provide no information on the heat exchanged in storm-churned waters beneath the surface. Furthermore, they need to be compared with ground truth measurements obtained in the actual storm environment. Such measurements can be gathered only by ventur-ing into the storm itself.

These environments are too turbulent for manned weather vessels or research aircraft, so NOAA scientists are testing unmanned mobile sensor platforms to observe this energy transfer. NOAA has used the Wave Glider AMV in the

Atlantic Ocean to collect critical data in areas that would be too difficult or too dangerous to access by other means. Monitoring an area north of Puerto Rico, the sensor platform is equipped with a standard weather station to measure tem-perature, humidity, barometric pressure and wind speed, direction and gusts. It also has a direc-tional wave sensor and a thermistor chain to mea-sure the water temperature from the surface down to 7 m [23 ft] in depth. This sensor platform demonstrated that high-quality temperature mea-surements from the upper ocean can be collected

> Tsunami monitoring. NOAA uses bottom pressure readings to detect earthquake activity that could produce a tsunami. A Wave Glider sensor platform has been employed to relay real-time data from a BPR to an onshore tsunami early warning center.

Bottom pressure recorder

Wave Glider AMV

Satellite

Tsunami earlywarning center

50 Oilfield Review

by autonomous vehicles in a harsh environment and telemetered in real time. An AMV also allows scientists to gather data from various locations as the vehicle roves through the storm.

Wave of the FutureHaving traveled 14 months and 14,800 km [9,200 mi] across the Pacific Ocean, the Wave Glider AMV has gained a record of reliabil-ity. This autonomous sensor platform has proved capable of carrying out a variety of important ocean monitoring functions formerly assigned to manned vessels—but over a longer time frame and at lower cost than traditional meth-ods. Furthermore, its station keeping ability lets the AMV duplicate the persistence and mea-

21. Leroy and Hine, reference 8.22. Leroy and Hine, reference 8.

surement functions of a moored ocean-monitor-ing buoy. In this mode, it significantly reduces the expense, time and risk incurred by ships and crews to deploy, recover and maintain a tradi-tional network of buoys. Its compact size also permits great flexibility and adaptability for rapid deployment to unforeseen or fast-chang-ing events to monitor conditions on, above or below the ocean’s surface.21

As offshore exploration and production move into deeper and more remote areas of the world’s oceans, the drilling rigs, production platforms, vessels and pipelines used in these operations will increasingly rely on autonomous vehicles for support. The capability to operate at or below the sea/air interface will serve the AMV customer in providing communication relays between surface and subsurface installations.

A new generation of larger, purpose-built Wave Glider vehicles will support subsurface well installations and field operations. The SV3 model will be more than 35% longer than previous mod-els and will carry a larger payload. Future sys-tems may be able to generate electricity from wave motion and may also include auxiliary elec-tric propulsion to enhance maneuvering and col-lision avoidance capabilities (above).22 The next generation of Wave Glider autonomous vehicles will be instrumental in extending the frontiers of exploration and production. — MV

>Wave of the future. The current Wave Glider SV3 prototype features an electric propulsion system with a low-drag propeller (black cone beneath the vertical fin). This larger model will accommodate a payload of 45 kg [100 lbm].

Contributors

Winter 2013/2014 51

Abhishek Agarwal is Sampling and Coring Services Product Champion for Schlumberger Wireline. Based in Sugar Land, Texas, USA, he manages new product development and marketing for these services. Prior to this assignment, which he began in 2013, he has held a number of positions within Schlumberger in India, Saudi Arabia and West Africa, primarily related to wireline operations. He began his career with Schlumberger in 2004 as a field engineer in Rajahmundry, Andhra Pradesh, India. Abhishek holds a BTech degree in chemical engineering from the Indian Institute of Technology Bombay in Powai, Mumbai.

Mary R. Albert has been a Professor of Engineering at the Thayer School of Engineering at Dartmouth College in Hanover, New Hampshire, USA, and Executive Director of the US Ice Drilling Program Office since 2009. Previously, she was a mathematician and research mechanical engineer at the US Army Cold Regions Research and Engineering Laboratory in Hanover, New Hampshire. Mary earned a BS degree in mathematics from The Pennsylvania State University, State College, USA, and a BS degree in engineering and an MS degree in engineering sciences from the Thayer School of Engineering at Dartmouth College. She received a PhD degree in applied mechanics and engineering sciences with an emphasis on computa-tional fluid dynamics from the University of California, San Diego, USA.

Peter Carragher is a Partner at Rose & Associates, LLP in Houston. He has more than 39 years of experi-ence as an exploration geologist with Amoco and BP in locations worldwide. He served in a scientific role dur-ing the Deepwater Horizon incident and now consults with the BP Gulf Coast Restoration Organization and other clients. Peter was involved in two Wave Glider† surveys in the Gulf of Mexico.

Karen Sullivan Glaser is a Geological Advisor and the Schlumberger PetroTechnical Services (PTS) Director of Curriculum for geology and geophysics. Based in Houston, she oversees the fixed-step and advanced training programs for geology and geophysics within PTS and also designs and teaches advanced geology courses. She is developing and teaching courses on unconventional resources and works closely with experts in technologies and techniques applied to the exploration and production of unconventional resources. She joined Schlumberger GeoQuest in 1995 and subsequently worked for WesternGeco, Integrated Project Management and Data & Consulting Services in various technical, marketing and management roles. Before joining Schlumberger, she worked for Exxon Production Research as a research geologist focusing on sequence stratigraphy. She has also worked for Amoco Production Company in the Permian basin. Karen obtained a BA degree in geology from Colgate University, Hamilton, New York, USA, an MS degree in petroleum geochemistry from the University of Oklahoma, Norman, USA, and a PhD degree in geology from Rice University in Houston.

Geoffrey Hargreaves is the Curator of the US Geological Survey National Ice Core Laboratory in Denver. He oversees daily operations, the laboratory database and cutting and shipment of core samples. He began his career as a research technician with Lockheed Ocean Systems in 1981. Since then he has held engineering, technical and research positions in the marine industry, including as resident marine technician for Scripps Institution of Oceanography in San Diego, California. Geoffrey holds a BS degree in biological oceanography from Humboldt State University, Arcata, California.

Graham Hine serves as Senior Vice President of Mission Services and Oil and Gas Technical Director for Liquid Robotics, Inc. (LRI) in Sunnyvale, California. His experience with the Wave Glider system began in 2006, when he joined the Jupiter Foundation Wave Glider program to assist in the development of prototypes of the oceangoing autonomous surface vehi-cle. When LRI was formed in 2007, he became CIO and program manager for its first major contract and sub-sequently served in a variety of leadership roles; he is the primary liaison between LRI and its joint venture with Schlumberger, Liquid Robotics Oil & Gas. Graham received degrees in management and software engi-neering from Claremont McKenna College, California, and Columbia University, New York City.

Greg M. Johnson is Principal Area Geophysicist, Depth Imaging for Schlumberger in Denver and has produced land and marine seismic imaging solutions worldwide. He has 31 years of industry experience and has held various positions within WesternGeco in Europe, Canada, the US and India, including as man-ager of depth imaging, earth models and data process-ing and as senior geophysicist with the Geophysical Support group. Greg earned a BA degree in environ-mental science from the University of Colorado Boulder, USA, and is an active member of the SEG, the European Association of Geoscientists and Engineers and the Denver Geophysical Society.

Robert L. Kleinberg is a Schlumberger Fellow at Schlumberger-Doll Research in Cambridge, Massachusetts, USA. His area of interest is unconven-tional fossil fuel resources—oil shale, gas shale, heavy oil and gas hydrates. His other projects at Schlumberger have focused on ultrasonics, resistivity, nuclear magnetic resonance and gravimetry and include several tool inventions. Bob obtained a BS degree in chemistry at the University of California, Berkeley, and a PhD degree in physics from the University of California, San Diego. Before joining Schlumberger, he spent two years at the Corporate Strategic Research Laboratory of Exxon Research and Engineering Company in Clinton, New Jersey, USA.

Robert Laronga heads the Geology and Rock Sampling Domain for Schlumberger Wireline in Clamart, France. He advises Schlumberger engineer-ing teams on the development of new borehole imag-ing and coring technology and related interpretation software and provides support to Schlumberger geol-ogists and customers who introduce these technolo-gies in the field. He has held several positions during

his 20-year career with Schlumberger, starting as a wireline field engineer and serving in various posi-tions in west Texas, the Gulf of Mexico and aboard the Ocean Drilling Program drillship. He was the field test engineer for the experimental prototype of the OBMI* oil-base microimager tool and has worked in sales and marketing positions in Scandinavia and continental Europe. Rob holds a BA degree in archae-ology and geology from Cornell University, Ithaca, New York.

Patrick Legh-Smith is the Marine Manager, Eastern Hemisphere for WesternGeco in Gatwick, England. He has worked in the seismic industry since 1988 and in marine seismic surveys with WesternGeco since 1992. He began his career as a navigator before becoming a party manager, operations manager, fleet technical support manager and WesternGeco HSE manager. Patrick earned a bachelor’s degree in geophysical oceanography and electronics engineering from the University of Wales, Bangor and a master’s degree in marine resource management from Heriot-Watt University, Edinburgh, Scotland.

Jeffrey Mayville, who is based in Houston, is the Marine Manager, Western Hemisphere for WesternGeco. He began his marine seismic career with WesternGeco in 1977 as a navigator, then became an observer and party manager. Starting in 1988, his subsequent positions onshore ranged from global HSE advisor to marine support and manage-ment positions in North America and the Asia Pacific region, as well as marine manager Schlumberger electromagnetics and for the Eastern Hemisphere. Jeffrey has a degree in electronics from DeVry University, Phoenix, Arizona, USA.

Camron K. Miller is a Principal Geologist for Schlumberger Production Management and PetroTechnical Services in Houston; he has global experience in shale reservoirs, although his current focus is on the Eagle Ford Shale in the US. He has more than 10 years of experience with the company working as a wireline field engineer, borehole geolo-gist, Data & Consulting Services manager and senior geologist focused on unconventional resources. Camron obtained a BA degree in geology from The College of Wooster, Ohio, USA, and an MS degree in petroleum geology from The University of Akron, Ohio.

Paul Miller is a Senior Reservoir Geophysicist with Schlumberger PetroTechnical Services in Kuala Lumpur. He is responsible for project management and is a technical lead for the Asia region. He has 15 years of industry experience and is an expert in seismic inver-sion and seismically derived fault and fracture map-ping. His career has included a wide range of international postings in Libya, Egypt, India, the UAE, Nigeria, Argentina, Brazil and the US. Paul received a BSc degree (Hons) in physics with computing from the University of Sussex, Brighton, England, and an MSc degree in exploration geophysics from the University of Leeds, West Yorkshire, England.

52 Oilfield Review

Dave Monk, who holds a PhD degree in physics from Nottingham University in England, is the Director of Geophysics and one of only two Distinguished Advisors at Apache Corporation. Based in Houston, he is responsible for seismic activity, including acqui-sition and processing in Argentina, Australia, Canada, Egypt, the North Sea and the US. He started his career on seismic crews in Nigeria and has subse-quently been involved in seismic processing and acquisition in locations worldwide. Author of more than 100 technical papers or articles and a number of patents, he received best paper awards from the SEG in 1992 and 2005 as well as one from the Canadian SEG in 2002 and was recipient of the Hagedoorn Award from the European Association of Exploration Geophysicists in 1994. Dave received honorary mem-bership in the Geophysical Society of Houston and life membership in the SEG and is the immediate Past President of the SEG.

Rod Nelson is a Senior Advisor at Schlumberger and Managing Director of the Liquid Robotics Oil & Gas joint venture between Schlumberger and Liquid Robotics, Inc. He is also Chairman of Schlumberger Technology Corporation. He was previously vice presi-dent of government and community relations for Schlumberger Limited. He serves on several joint industry steering committees and boards including the corporate advisory boards of the AAPG, the Global Climate and Energy Project at Stanford and the Advanced Energy Consortium. Active in the commu-nity, he is on the board of directors of the Greater Houston Partnership, Fort Bend Education Foundation and University of Houston Energy Advisory Board. He led the technology task group for the National Petroleum Council’s 2007 global oil and gas study “Facing the Hard Truths about Energy.” Rod holds a bachelor’s degree in engineering from the University of Wisconsin and an executive MBA degree from Massachusetts Institute of Technology (MIT) Sloan School of Management, Cambridge, USA. He joined Schlumberger as a wireline field engineer in 1980.

Sudhir Pai serves as Vice President of Operations and Technology with Liquid Robotics Oil & Gas in Houston. He sets the technology vision and is responsible for worldwide operations for this joint venture between Schlumberger and Liquid Robotics, Inc. Previously, he held a number of management positions with Schlumberger, including vice president and general manager of completions and general manager of Schlumberger operations in Mumbai. He has more than 30 years of experience in oilfield services, includ-ing roles in corporate planning, supply chain manage-ment, human resources, service quality and HSE in the Middle East, Asia, Africa, India, the UK and the US. Sudhir earned a bachelor’s degree in electrical engi-neering from the University of Bombay, Mumbai. He is a member of the SPE and the DeepStar technology development project and was a technical steering com-mittee member and session chair for the 2013 Clean Gulf conference and exhibition.

Iain Parnum is a Research Scientist at Curtin University in Perth, Western Australia, Australia, where he carries out research and fieldwork and teaches marine acoustics and technology. He has used Wave Glider technology for marine environmental assessment since 2012. Iain has an MSc degree in marine science from the University of Wales, Bangor, and a PhD degree in marine acoustics from Curtin University.

Wayne D. Pennington is the Interim Dean of the College of Engineering at Michigan Technological University in Houghton, USA. He specializes in seismic petrophysics and provides courses on the subject through NExT, a Schlumberger company. Before joining the faculty at Michigan Technological University 20 years ago, Wayne was on the faculty at The University of Texas at Austin and spent nine years with Marathon Oil Company at their research facility in Littleton, Colorado. He served as the first vice president of the SEG and as president of the American Geosciences Institute. He received BA degrees in geology and geo-physics from Princeton University, New Jersey, an MS degree from Cornell University, Ithaca, New York, and a PhD degree from the University of Wisconsin-Madison, USA. An asterisk (*) denotes a mark of Schlumberger.

A dagger (†) denotes a mark of Liquid Robotics, Inc.

Whipstock Systems. Sidetracking is a common strategy for bypassing downhole obstructions, drilling new wellbores in search of more productive zones or drilling lateral wells to maximize wellbore exposure. Whipstocks have long been used for sidetracking in cased holes. Advances in whipstock technology now give operators more flexibility in sidetracking both cased and openhole wellbores.

Coming in Oilfield Review

Formation Testing. Well test results weigh heavily in operator decisions for asset development. A new compact formation test system with bidirectional, real-time communication and control is helping opera-tors make choices based on more high-quality data than was previously available using traditional forma-tion test technology.

Global Climate and Energy Project. Universities play a vital role in energy research and development. Schlumberger is a founding sponsor of the Global Climate and Energy Project, a unique industry and academia partnership housed at Stanford University and formed to support step-out research into new technologies to provide affordable energy and reduce greenhouse gas emissions.

High-Definition Spectroscopy Logging. Neutron capture spectroscopy logging measures relative elemental yields, which can then be converted to concentrations of minerals commonly found in hydro-carbon reservoirs. Dry-weight concentration data are further processed to determine in situ mineralogy, lithology and geochemical stratigraphy. A new tool measures, with higher resolution and greater accu-racy, more elements than previous generation tools were able to.

Paul Shone is based in London, where he serves as HSE Team Lead for Chevron Energy Technology Company. He has worked for the company for 17 years.

Jonathan Smith, who has worked for Shell since 1985, is Team Lead, Geomatics Operations for Shell Exploration and Production Company in Houston. He supervises the Shell Upstream Americas Geomatics Operations teams in Houston and Calgary. The teams provide surveying expertise along with project engi-neering and management for a wide range of geomat-ics projects.

Christian Tichatschke is an Exploration Manager for Total E&P Uruguay BV. He has worked for 10 years with the Total Group, where he has been assigned to various positions within headquarters and operational units. Christian, who is based in Montevideo, Uruguay, holds a master’s degree in geology from the Ludwig Maximilian University of Munich, Germany.

Brian Toelle, who joined Schlumberger in 1997, is a Schlumberger PetroTechnical Services Advisor on exploration and geophysics in Denver. He advises on projects worldwide and teaches classes on shale reser-voir exploration and development, geophysics and structural geology. He has 32 years of industry experi-ence, specializing in exploration and fully integrated projects for Texaco and Saudi Aramco. Brian, who is a 2014–2015 SPE Distinguished Lecturer, earned a PhD degree in applied geophysics from West Virginia University, Morgantown, USA.

Larissa Walker is a Senior Petrophysical Engineer for Shell Appalachia Exploration in Sewickley, Pennsylvania. She began her career in 2005 as a petro-physical engineer for Shell in Canada, gaining experi-ence in both development and exploration activities for unconventional reservoirs such as structured car-bonate plays, basin-centered gas reservoirs, coalbed methane and shale gas. As of 2011, her work with Shell Appalachia has focused on developing Marcellus Shale resources. Larissa has a BASc degree (Hons) in geolog-ical engineering from the University of Waterloo, Ontario, Canada.

Winter 2013/2014 53

NEW BOOKS

Antarctica: A BiographyDavid DayOxford University Press 198 Madison AvenueNew York, New York 10016 USA2013. 624 pages. US$ 34.95ISBN: 978-0-199-86145-3

David Day discusses the history of Antarctica by focusing on explorer biographies, whaling and sealing expeditions, the rivalries between nations to claim the land and the eventual adoption of the Antarctic Treaty, which established Antarctica as a cooperative base for scientific discovery.

Contents:

• 1770s

• 1780–1820

• 1821–1838

• 1839–1843

• 1843–1895

• 1895–1906

• 1907–1912

• 1912–1918

• 1919–1926

• 1926–1928

• 1929–1930

• 1931–1933

• 1934–1936

• 1937–1938

• 1939–1941

• 1941–1945

• 1945–1947

• 1948–1951

• 1952–1956

• 1957–1960

• 1961–2012

• Epilogue, Endnotes, Bibliography, Index

Solid as a block of Antarctic ice itself . . . [Day’s] book draws on five years of meticulous research to tell the story of human endeavour in Antarctica. . . . It paints a poignant biographical picture of the characters

involved, the gruelling expeditions undertaken, and the rivalries between nations as they raced to chart the continent and claim possession of it. . . . [E]xcellent account.

“Opening Up an Empty Quarter,” The Economist

(June 15, 2013), http://www.economist.com/news/

books-and-arts/21579428-opening-up-empty-

quarter-south-park (accessed July 16, 2013).

Day has done a remarkable job of collating information from rich and varied international sources. He draws from original accounts, news-paper articles, [and] the recently released papers of US naval officer and polar explorer Richard Byrd. . . . [T]hanks to Day, the intrigues and posturing that saturate the history of this distant land have now been exposed.

The dramas, played out in secret memoirs and in published statements in newspapers, give the book a slow, even glacial, pace at times.

Stump E: “Frozen Assets,” Nature 493, no. 7433

(January 24, 2013): 478–479.

Day weaves a masterly tale of expeditions and their leaders in this hugely detailed and well-researched tome. There are some absolute gems with new insights for even the most avid readers on the subject.

Turney C: “Book Review,” Times Higher

Education (February 21, 2013), http://www.

timeshighereducation.co.uk/books/antarctica-a-

biography-by-david-day/2001714.article (accessed

July 18, 2013).

Brilliant Blunders: From Darwin to Einstein— Colossal Mistakes by Great Scientists That Changed Our Understanding of Life and the UniverseMario Livio Simon & Schuster 1230 Avenue of the AmericasNew York, New York 10020 USA2013. 352 pages. US$ 26.00 ISBN: 978-1-4391-9236-8

The author looks at the lives of five scientists—Charles Darwin, William Thomson (Lord Kelvin), Linus Pauling, Fred Hoyle and Albert Einstein—to show that even great minds make major mistakes. The book goes on to explore how their errors are a part of the fabric of scientific discovery.

Contents:

• Mistakes and Blunders

• The Origin

• Yea, All Which It Inherit, Shall Dissolve

• How Old Is the Earth?

• Certainty Generally Is Illusion

• Interpreter of Life

• Whose DNA Is It Anyway?

• B for Big Bang

• The Same Throughout Eternity?

• The “Biggest Blunder”

• Out of Empty Space

• Coda

Notes, Bibliography, Index

Livio brings the care of a historian to his nimble narratives, avoiding heroic clichés.

He’s less adept at explaining why these great scientists made their mistakes, too often trotting out pop psychology to demonstrate why people stubbornly cling to ideas even when they see evidence to the con-trary. The psychology of bad science is a fascinating topic, but it requires a broader look at how the entire scien-tific community operates. Five scien-tists—no matter how great—cannot shoulder that load.

Zimmer C: “The Genius of Getting It Wrong,”

The New York Times (June 7, 2013), http://www.

nytimes.com/2013/06/09/books/review/brilliant-

blunders-by-mario-livio.html (accessed

June 14, 2013).

Readers with dim memories of high school physics may sometimes grapple for comprehension, despite Livio’s heuristic talents. But stick with him. . . . Thanks to Livio, such brilliant blunders as Einstein’s will help guide our way through the intellectual as well as the physical cosmos.

Dunlap J: “Book Review,” Washington

Independent Review of Books (June 12, 2013),

http://www.washingtonindependentreviewofbooks.

com/bookreview/brilliant-blunders-from-darwin-

to-einsteincolossal-mistakes-by-great-scient

(accessed June 18, 2013).

The Golden Ticket: P, NP, and the Search for the ImpossibleLance FortnowPrinceton University Press41 William StreetPrinceton, New Jersey 08540 USA2013. 192 pages. US$ 26.95ISBN: 978-0-691-15649-1

An unsolved problem in computer science, P versus NP, or P-NP, asks whether every problem that can be verified by a computer can also be solved by a computer. The author explores the history of the P-NP prob-lem, gives examples of the problem in a variety of disciplines, including econom-ics, physics and biology, and describes the benefits and challenges of the problem.

Contents:

• The Golden Ticket

• The Beautiful World

• P and NP

• The Hardest Problems in NP

• The Prehistory of P versus NP

• Dealing with Hardness

• Proving P ≠ NP

• Secrets

• Quantum

• The Future

• Chapter Notes and Sources, Index

Fortnow’s book . . . bills itself as a primer for the general reader, though you will likely regret not having paid slightly more attention during calculus class.

Nazaryan A: “A Most Profound Math Problem,”

The New Yorker (May 2, 2013), http://www.

newyorker.com/online/blogs/elements/2013/

05/a-most-profound-math-problem.html

(accessed July 1, 2013).

Fortnow effectively initiates readers into the seductive mystery and importance of P and NP problems.

“Book Review,” Publishers Weekly (January 28,

2013), http://publishersweekly.com/978-0-691-

15649-1 (accessed June 18, 2013).

Oilfield Review54

DEFINING DIRECTIONAL DRILLING

> Directional drilling applications. Reservoirs that are not readily accessible from available surface locations can be exploited through directional drilling.

The practice of directional drilling traces its roots to the 1920s, whenbasic wellbore surveying methods were introduced. These methods alerted drillers to the fact that supposedly vertical wells were actually deflecting in unwanted directions. To combat this deviation, drillers devised techniques to keep the well path as vertical as possible. The same techniques were later employed to deliberately deflect the well path to intersect hard-to-access reserves.

The first intentionally drilled directional wells provided remedial solu-tions to drilling problems: straightening crooked wellbores, sidetracking around stuck pipe and drilling relief wells to kill blowouts (below). Directional drillers used rudimentary survey instruments to orient the wellbore. By the 1930s, a controlled directional well was drilled in Huntington Beach, California, USA, from an onshore location to target offshore oil sands.

Today, operators use sophisticated drilling assemblies to drill complex geologic structures identified from 3D seismic data. Previously unreachable reserves have become accessible and economical to produce.

Directional drilling includes three main specialized applications: extended-reach drilling (ERD), multilateral drilling and short-radius drilling. Operators have used ERD to access offshore reservoirs from land locations, sometimes eliminating the need for a platform. As of 2013, the world’s longest ERD well is the 12,345-m [40,502-ft] well drilled from Sakhalin Island, Russia, to the offshore Odoptu field. Multilateral drilling helps increase wellbore con-tact with hydrocarbon-producing zones by branching multiple extensions off a single borehole. The first multilateral well was drilled in 1953 at Bashkiria field, Bashkortostan Republic, Russia. The main borehole had nine lateral branches that increased penetration of the pay zone by 5.5 times and produc-tion by 17-fold, and cost only 1.5 times that of a conventional well. Short-radius drilling produces wells with a curve of 44-m [144-ft] radius or smaller.

Principles of Directional DrillingMost directional wells begin as vertical wellbores. At a designated depth, known as the kickoff point (KOP), the directional driller deflects the well path by increasing well inclination to begin the build section. Surveys taken during the drilling process indicate the direction of the bit and the toolface, or orientation of the measurement sensors in the well. The directional driller constantly monitors these measurements and adjusts the trajectory of the wellbore as needed to intercept the next target along the well path.

Initially, directional drilling involved a simple rotary bottomhole assem-bly (BHA) and the manipulation of parameters such as weight on bit (WOB), rotary speed and BHA geometry to achieve a desired trajectory. Changes in BHA stiffness, stabilizer placement and gauge, rotary speed, WOB, hole diameter, hole angle and formation characteristics all affect the directional capability and drilling efficiency of a BHA.

By varying stabilizer placement in the drillstring, directional drillers can alter side forces acting on the bit and the BHA, causing it to increase, main-tain or decrease inclination, commonly referred to as building, holding or dropping angle, respectively (below).• To build angle, the directional driller uses a BHA with a full gauge near-bit

stabilizer, another stabilizer between 15 to 27 m [50 to 90 ft] above the first and a third stabilizer about 9 m [30 ft] above the second. This BHA acts as a fulcrum, exerting a positive side force at the bit.

• To hold angle, the directional driller uses a BHA with 3 to 5 stabilizers, placed about 9 m apart. This packed BHA is designed to exert no net side force.

• To drop angle, the directional driller uses a BHA with the first stabilizer 9 to 27 m behind the bit. This BHA acts as a pendulum, exerting a negative side force at the bit.

During well planning, the directional driller must consider several fac-tors to determine the required trajectory, particularly dogleg severity (DLS)—the rate of change in wellbore trajectory, measured in degrees per 30 m [100 ft]—as well as the capabilities of the BHA, drillstring, logging tools and casing to pass through the doglegs. Drilling limitations include rig specifications such as maximum torque and pressure available from surface systems. Geologic features such as faults or formation changes need to be

The Art of Controlling Wellbore Trajectory

Oilfield Review Winter 2013/2014: 25, no. 4.

Copyright © 2014 Schlumberger.

For help in preparation of this article, thanks to Steven Hough, Stonehouse, England; and Richard Hawkins, Midland, Texas, USA.

Kate MantleDirectional Drilling Advisor

N

270

0

90

180

> Using the BHA to change angle. Bending of the pipe above a bit influences borehole deviation. Through strategic placement of drill collars and stabilizers, the directional driller can increase or decrease flexibility and bowing of the BHA to build or drop angle.

Fulcrum Assembly Pendulum Assembly

Stabilizer

Near-bit stabilizer Buildin

g

angle

Droppin

g

angle

Multiple wellbores Offshore targets Relief well Horizontal well

Sidetracking

carefully considered; for example, very soft formations may limit build rates,and formation dip may cause a bit to walk, or drift laterally. Local knowl-edge of drilling behavior enables the directional driller to derive the correctlead angle needed to intercept the target.

The skill of the directional driller lies in projecting ahead, estimatingthe spatial position of the bit and, based on the specific circumstances,deciding what course to take to intercept the target or targets. In the earlydays of directional drilling, a manual slide rule device was used to calculatethe toolface angle required to drill from the last survey station to a target.Today, computer programs perform the same function.

Directional Drilling OperationsTo steer a well to its target, directional drillers employ the following techniques:

Jetting—A jetting assembly provides directional capability while drill-ing through loose or unconsolidated formations. Jetting bits are roller conebits with either a large extended nozzle in place of one of the cones, or withone large nozzle and two small nozzles. The large nozzle provides the “highside” reference, and the well path is deflected by alternately sliding or rotat-ing the drillstring.

Nudging—This technique is often used in tophole sections, where sev-eral wellbores in close proximity to one another can pose magnetic interfer-ence issues and increase the risk of collision with other wellbores. The wellpath is nudged, or deflected, from vertical to pass the hazard—then steeredback to vertical when the hazard has been passed.

Kicking off—Diverting a well path from one trajectory to another iscalled kicking off. The number of KOPs in a single well path depends on thecomplexity of the planned trajectory.

Sidetracking—Deflecting a well path from an existing wellbore, or side-tracking, is performed for a variety of reasons such as avoiding a well col-lapse, a zone of instability or a section of previously drilled wellborecontaining unretrieved fish (junk or tools left in the well). This technique isalso used to initiate multilateral drilling operations. Operators also drill ver-tical pilot holes to confirm reservoir true vertical depth (TVD), then side-track horizontally to maximize reservoir exposure. They sometimessidetrack wells when expected targets are not encountered.

Whipstock operations—A whipstock is a wedge-shaped steel tooldeployed downhole to mechanically alter the well path. The whipstock isoriented to deflect the bit from the original borehole at a slight angle and inthe direction of the desired azimuth for the sidetrack. It can be used incased or open holes.

Geosteering—Geosteering uses formation evaluation data obtainedwhile drilling—primarily through measurements-while-drilling (MWD) orlogging-while-drilling (LWD) sensors—to provide real-time input for steer-ing decisions in horizontal and high-angle wells. Recent improvements intelemetry allow MWD and LWD data to be transmitted faster and withgreater data density than in the past, greatly increasing the accuracy withwhich the well trajectory can be controlled.

Advances in Directional DrillingThe development of reliable mud motors provided an important advance indirectional drilling technology. Wellbore direction could then be controlledusing a bent motor housing, which was oriented to point the drill bit in thedesired direction. Mud motors use the mud pumped through a rotor and statorassembly to turn the bit without rotating the drillstring from the surface. Byalternating intervals of rotating mode and sliding mode, the directional driller

can control the wellbore trajectory and steer it in the desired direction. Inrotating mode, the drilling rig’s rotary table or its topdrive rotates the entiredrillstring to transmit power to the bit. By contrast, in sliding mode, the bendand bit are first oriented in the desired direction, then the downhole mudmotor alone powers the bit, with no rotation of the drillstring above the bit.

Drilling motors and rotary steerable systems (RSSs) presented direc-tional drillers with an efficient way to steer wells—and to do so with greateraccuracy. The RSS enables wells to be drilled directionally while the drill-string continuously rotates. The advantages of this method are improvedwell cleaning through rotation, a smoother wellbore and more accuratedirectional control. To steer the RSS, the directional driller transmits com-mands from the surface using pressure fluctuations in the mud column.

Today, hybrid RSSs use pads inside the tool to push against an internalsleeve that pivots and points the bit in the desired direction. These tools candeliver DLS of up to 18°/30 m. Hybrid RSS tools enable directional drillersto kick off from vertical at greater depths and land, or transition the well tohorizontal, more quickly than was previously possible. This techniqueincreases wellbore exposure to the reservoir (above).

Advanced steering systems use a system-matched mud motor in combi-nation with an RSS tool below it. This BHA design enables higher revolu-tions per minute at the bit, enhanced steering control and increased rateof penetration.

Future DevelopmentsThe introduction of fully automated downhole control systems is likelywithin the near future. Such advances, however, do not herald the removalof directional drillers from the process; their experience will always beneeded to oversee the full scope of directional drilling operations. Thefuture promises to be fast moving and technologically groundbreaking forthis highly specialized niche within the oil and gas industry.

Hybrid rotary steerable systemConventional rotary steerable systemKickoff point

Landing point

Landing point

TDTD

Deeperkickoff point

> Directional capabilities. A hybrid system lets the operator kick off from greater depths, yet land sooner in the reservoir zone than was possible with conventional rotary steerable systems.

Winter 2013/2014 55

Oilfield Review Annual Index—Volume 25

ARTICLES

Blowing in the Solar Wind: Sun Spots, Solar Cycles and Life on EarthArsentiev A, Hathaway DH and Lessard RW.Vol. 25, no. 3 (Autumn 2013): 48–60.

Bridging the Talent GapAl-Abdulbaqi S, Alobaydan A, Chhibber R, Jamaluddin A, Murphy L, Venugopal K and Johnson JD.Vol. 25, no. 1 (Spring 2013): 4–13.

Casing Corrosion Measurement to Extend Asset LifeAbdallah D, Fahim M, Al-Hendi K, Al-Muhailan M, Jawale R, Al-Khalaf AA, Al-Kindi Z, Al-Kuait AS, Al-Qahtani HB, Al-Yateem KS, Asrar N, Aziz SA, Kohring JJ, Benslimani A, Fituri MA and Sengul M.Vol. 25, no. 3 (Autumn 2013): 18–31.

Core Truth in Formation EvaluationAndersen MA, Duncan B and McLin R.Vol. 25, no. 2 (Summer 2013): 16–25.

Developments in Full Azimuth Marine Seismic ImagingBrice T, Buia M, Cooke A, Hill D, Palmer E, Khaled N, Tchikanha S, Zamboni E, Kotochigov E and Moldoveanu N.Vol. 25, no. 1 (Spring 2013): 42–55.

Drilling Through Ice and into the PastAlbert MR and Hargreaves G.Vol. 25, no. 4 (Winter 2013/2014): 4–15.

Formation Density from a Cloud, While DrillingAllioli F, Cretoiu V, Mauborgne M-L, Evans M, Griffiths R, Haranger F, Stoller C, Murray D and Reichel N.Vol. 25, no. 2 (Summer 2013): 4–15.

Geomagnetic Referencing— The Real-Time Compass for Directional DrillersBuchanan A, Finn CA, Love JJ, Worthington EW, Lawson F, Maus S, Okewunmi S and Poedjono B.Vol. 25, no. 3 (Autumn 2013): 32–47.

Multistage Stimulation in Liquid-Rich Unconventional FormationsAviles I, Baihly J and Liu GH.Vol. 25, no. 2 (Summer 2013): 26–33.

New Dimensions in Wireline Formation TestingAyan C, Corre P-Y, Firinu M, Garcia G, Kristensen MR, O’Keefe M, Pfeiffer T, Tevis C, Zappalorto L and Zeybek M.Vol. 25, no. 1 (Spring 2013): 32–41.

A New Platform for Offshore Exploration and Production Carragher P, Hine G, Legh-Smith P, Mayville J, Pai S, Parnum I, Shone P, Smith J and Tichatschke C.Vol. 25, no. 4 (Winter 2013/2014): 40–50.

Rotary Sidewall Coring— Size MattersAgarwal A, Laronga R and Walker L.Vol. 25, no. 4 (Winter 2013/2014): 30–39.

Seeking the Sweet Spot: Reservoir and Completion Quality in Organic ShalesGlaser KS, Miller CK, Johnson GM, Toelle B, Kleinberg RL, Miller P and Pennington WD.Vol. 25, no. 4 (Winter 2013/2014): 16–29.

Stimulating Naturally Fractured Carbonate ReservoirsAsiri KS, Atwi MA, Jiménez Bueno O, Lecerf B, Peña A, Lesko T, Mueller F, Pereira AZI and Tellez Cisneros F.Vol. 25, no. 3 (Autumn 2013): 4–17.

Stimulation Design for Unconventional ResourcesAjayi B, Aso II, Terry IJ Jr, Walker K, Wutherich K, Caplan J, Gerdom DW, Clark BD, Ganguly U, Li X, Xu Y, Yang H, Liu H, Luo Y and Waters G.Vol. 25, no. 2 (Summer 2013): 34–46.

Structural Steering— A Path to ProductivityAmer A, Chinellato F, Collins S, Denichou J-M, Dubourg I, Griffiths R, Koepsell R, Lyngra S, Marza P, Murray D and Roberts I.Vol. 25, no. 1 (Spring 2013): 14–31.

EDITORIALS

Core Analysis: Combining Expertise for Insight into the ReservoirAndersen MA.Vol. 25, no. 2 (Summer 2013): 1.

Evolving RevolutionMacGregor C.Vol. 25, no. 1 (Spring 2013): 1.

Geomagnetic Referencing for Well PlacementBuchanan A.Vol. 25, no. 3 (Autumn 2013): 1.

Unconventional Reservoir Sweet Spots from GeophysicsMonk D.Vol. 25, no. 4 (Winter 2013/2014): 1.

DEFINING…INTRODUCING BASIC CONCEPTS OF THE E&P INDUSTRY

Defining Directional Drilling: The Art of Controlling Wellbore Trajectory Mantle K.Vol. 25, no. 4 (Winter 2013/2014): 54–55.

Defining Drilling Fluids: Drilling Fluid Basics Williamson D.Vol. 25, no. 1 (Spring 2013): 63–64.

Defining Hydraulic Fracturing: Elements of Hydraulic Fracturing Nolen-Hoeksema R.Vol. 25, no. 2 (Summer 2013): 51–52.

Defining Production Logging: Principles of Production Logging Mukerji P.Vol. 25, no. 3 (Autumn 2013): 63–64.

NEW BOOKS

Antarctica: A BiographyDay D.Vol. 25, no. 4 (Winter 2013/2014): 53.

Atmosphere, Clouds, and ClimateRandall D.Vol. 25, no. 1 (Spring 2013): 62.

Brilliant Blunders: From Darwin to Einstein—Colossal Mistakes by Great Scientists That Changed Our Understanding of Life and the UniverseLivio M.Vol. 25, no. 4 (Winter 2013/2014): 53.

Diamondoid Molecules: With Applications in Biomedicine, Materials Science, Nanotechnology & Petroleum ScienceMansoori GA, de Araujo PLB and de Araujo ES.Vol. 25, no. 2 (Summer 2013): 49.

Digital Wars: Apple, Google, Microsoft and the Battle for the InternetArthur C.Vol. 25, no. 1 (Spring 2013): 61.

Earthmasters: The Dawn of the Age of Climate EngineeringHamilton C.Vol. 25, no. 1 (Spring 2013): 62.

The Earthquake Observers: Disaster Science from Lisbon to RichterCoen DR.Vol. 25, no. 2 (Summer 2013): 50.

Fundamentals of Condensed Matter and Crystalline PhysicsSidebottom DL.Vol. 25, no. 2 (Summer 2013): 50.

Global Environment: Water, Air, and Geochemical Cycles, Second EditionBerner EK and Berner RA.Vol. 25, no. 1 (Spring 2013): 60.

The Golden Ticket: P, NP, and the Search for the ImpossibleFortnow L.Vol. 25, no. 4 (Winter 2013/2014): 53.

The Great Fossil Enigma: The Search for the Conodont AnimalKnell SJ.Vol. 25, no. 2 (Summer 2013): 50.

Great Inventions That Changed the WorldWei J.Vol. 25, no. 1 (Spring 2013): 61.

Henri Poincaré: A Scientific BiographyGray J.Vol. 25, no. 1 (Spring 2013): 61.

Interop: The Promise and Perils of Highly Interconnected SystemsPalfrey J and Gasser U.Vol. 25, no. 1 (Spring 2013): 62.

Lynn Margulis: The Life and Legacy of a Scientific RebelSagan D (ed).Vol. 25, no. 1 (Spring 2013): 59.

Maverick Genius: The Pioneering Odyssey of Freeman DysonSchewe PF.Vol. 25, no. 2 (Summer 2013): 49.

The Million Death Quake: The Science of Predicting Earth’s Deadliest Natural DisasterMusson R.Vol. 25, no. 1 (Spring 2013): 59.

Reverse Innovation: Create Far from Home, Win EverywhereGovindarajan V and Trimble C.Vol. 25, no. 1 (Spring 2013): 60.

Seismic Imaging and Inversion: Application of Linear Inverse TheoryStolt RH and Weglein AB.Vol. 25, no. 1 (Spring 2013): 60.

Solved Problems in GeophysicsBuforn E, Pro C and Udías A.Vol. 25, no. 1 (Spring 2013): 60.

56 Oilfield Review

Oilfield GlossaryAvailable in English and Spanish, the Oilfield Glossary is a rich accumulation of more than 5,800 definitions from18 industry disciplines. Technical experts have reviewed each definition; photographs, videos and illustrationsenhance many entries. See the Oilfield Glossary at http://www.glossary.oilfield.slb.com/.

The Last Word