Summer 2014

Steam-Assisted Gravity Drainage

Scientific Drilling

Land Seismic Surveys

Redefining PDC Bits

Oilfield Review

14-OR-0003

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.

†Android is a trademark of Google Inc.‡Apple, iPad, and iTunes are marks of Apple Inc., registered in the U.S. and other countries.

Coming soon—Oilfield Review

app for the Android

† platform

www.slb.com/oilfieldreview

Schlumberger

Oilfield Review

4 Warming to Heavy Oil Prospects

Steam-assisted gravity drainage methods are allowing operators to produce bitumen and heavy oil profitably. Researchers are using this evolving secondary recovery tech-nique to help operators exploit massive reserves while mini-mizing impact on the environment.

16 Ultradeep Scientific Ocean Drilling— Probing the Seismogenic Zone

Improvements in deepwater drilling and measurement technologies enable scientists to extend the limits of marine scientific drilling to previously inaccessible domains. Ultradeep scientific drilling is helping to advance scientists’ understanding of earthquakes and other fundamental Earth processes.

Executive EditorLisa Stewart

Senior EditorsTony SmithsonMatt VarhaugRick von Flatern

EditorsIrene FærgestadRichard Nolen-Hoeksema

Contributing EditorsH. David LeslieGinger OppenheimerRana Rottenberg

Design/ProductionHerring DesignMike Messinger

Illustration Chris LockwoodTom McNeffMike 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 bit is readied for its trip into the well. This polycrystalline diamond compact (PDC) bit is equipped with a conical dia-mond element, mounted in the center of the bit. Whereas conventional cutters on a PDC bit shear the rock, the conical dia-mond element instead crushes the rock.

2

Summer 2014Volume 26Number 2

ISSN 0923-1730

3

32 Land Seismic Surveys for Challenging Reservoirs

Land seismic point-receiver technology enables cost-effective acquisition of finely sampled 3D surveys over large areas. Detailed images extracted from the processed seismic data may be calibrated and used by operators to plan drilling, completion and stimulation operations in tight reservoirs.

48 PDC Bit Technology for the 21st Century

New developments in polycrystalline diamond compact bit technology are helping drillers achieve higher rates of penetration and increase footage drilled per bit run.

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

58 Contributors

60 Coming in Oilfield Review

61 Books of Note

63 Defining Coiled Tubing: Big Reels at the Wellsite

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

4 Oilfield Review

Warming to Heavy Oil Prospects

In recent years, because of innovation aimed at exploiting unconventional resources,

oil and gas industry economists have substantially increased estimates of the world’s

remaining recoverable oil reserves. Now, operators are using those new technologies

and others to exploit heavy oil sands and push peak oil even further into the future.

Farrukh Akram Terry StoneAbingdon, England

William J. BaileyCambridge, Massachusetts, USA

Euan ForbesCalgary, Alberta, Canada

Michael A. Freeman Houston, Texas, USA

David H.-S. LawEdmonton, Alberta

Glenn WoiceshynAbsolute Completion TechnologiesCalgary, Alberta

K.C. Yeung Brion Energy Calgary, Alberta

Oilfield Review Summer 2014: 26, no. 2. Copyright © 2014 Schlumberger.For help in preparation of this article, thanks to Marty Chisholm, Calgary; Adrian Francis and Basim Abd Hameed Moustafa, Houston; Joseph Hayes, Rosharon, Texas, USA; and Herb Illfelder, Katy, Texas. ECLIPSE, HotlineSA3, Merak, Petrel, RADAR, ThermaSTONE, VISAGE and Vx are marks of Schlumberger.FluxRite, MeshFlux and MeshRite are marks of Absolute Completion Technologies.SAGDRIL is a mark of M-I, L.L.C.

The world’s reserves of heavy oil are on par with those of the largest conventional oil fields in the Middle East and are located in more than 30 coun-tries around the globe. Heavy oil reservoirs are expensive to drill and difficult to complete and require unique techniques to produce. Shallow, unconsolidated oil sands present drillers with wellbore stability and steering challenges. Completions must be designed to withstand high-temperature environments because many heavy oil production strategies require thermal recovery methods. At ambient temperatures, heavy oil and

bitumen are resistant to flow through reservoir rock because of their high viscosities. Conse-quently, the energy expended to produce and upgrade a barrel of oil can be as high as 40% of the total energy available from the heavy oil resource.1

To overcome these challenges, engineers have developed many technologies and recovery meth-ods, including combinations of horizontal drill-ing, chemical and water injection, artificial lift and in situ heating. Operators in the oil sands of Western Canada are finding commercial success producing extraheavy oil and bitumen through

> The steam chamber. To create a steam chamber in SAGD operations, the operator injects steam into a formation through a horizontal well. The steam chamber grows around and above the injection well. At the edge of the steam chamber, heated bitumen and steam condensate flow under the force of gravity to the production well. Ideally, the production well is located parallel to and below the injection well and a few meters above the formation bottom. (Adapted from Gates et al, reference 17.)

Steam chamber

Steam

Shale

Shale

Rese

rvoi

r

Toe ofinjection

well

Bitumenand steam

condensateflow zone

Nativebitumen

Toe ofproduction

well

1. Heavy oil is defined as having 22.3 degree API or less. Oils that are denser than water—those of 10 degree API or less—are known as “extraheavy” when viscosity is less than 10,000 cP [10,000 mPa.s] at reservoir conditions and as bitumen when viscosity is greater than 10,000 cP.

For more on heavy oil: Alboudwarej H, Felix J, Taylor S, Badry R, Bremner C, Brough B, Skeates C, Baker A, Palmer D, Pattison K, Beshry M, Krawchuk P, Brown G, Calvo R, Cañas Triana JA, Hathcock R, Koerner K, Hughes T, Kundu D, López de Cárdenas J and West C: “Highlighting Heavy Oil,” Oilfield Review 18, no. 2 (Summer 2006): 34–53.

Viscosity is a measure of a fluid’s resistance to flow and is defined as the ratio of shear stress to shear rate. Density is the ratio of mass per unit volume. Although density may vary slightly with temperature, viscosity decreases rapidly in response to increasing temperature.

Summer 2014 55

the process of steam-assisted gravity drainage (SAGD). The SAGD method employs pairs of par-allel, horizontal wellbores drilled one above the other in the same vertical plane. During SAGD operations, steam is pumped into the upper well-bore and forced out into the formation to form a steam-affected volume called a steam chamber. As the steam chamber expands upward and laterally, the oil viscosity at the steam/oil front decreases, and the oil becomes more mobile. The mobile oil and condensed steam mixture flows by gravity downward along the steam/oil boundary to the lower, horizontal wellbore from which it may be pumped to the surface (previous page).

Heat reduces fluid viscosity (right). However, dispersing steam evenly throughout a formation is difficult, and such uneven dispersal often results in viscous fingering effects from oils of low viscosities flowing faster in the formation than

> Heavy oil viscosity versus temperature. For two heavy oil samples (blue and red) that were obtained from fields located in different parts of the world, viscosity decreases as temperature increases.

Visc

osity

, cP

Temperature, °C

10,000,000

0 50 100 150 200 250 300 350 400

1,000,000

100,000

10,000

1,000

100

10

0.1

1

6 Oilfield Review

oils of higher viscosities; a significant volume of oil may be left behind because of uneven steam chamber development along the lengths of a SAGD well pair.2 Therefore, production engineers must manage the flow of formation fluids to the production well, primarily through control of steam injection. To do so, they must understand the geologic and permeability heterogeneity of the formation.

This article looks at some of the tools and methods employed by SAGD operators to opti-mize production of heavy oil. The implementa-tion of these innovations and their impact on production of bitumen and extraheavy oil are illustrated through case histories from Canada, currently the only country in the world with com-mercially successful SAGD projects.

Where to Drill The economic success of most enhanced oil recovery (EOR) projects depends on efficient dis-placement of oil from the formation by another injected fluid. In the case of SAGD, displacement occurs at the expanding front of the steam cham-ber, where steam heats the bitumen, thereby increasing its mobility. The mobile oil and con-densed steam then flow under the force of gravity to the production well.3 A uniform steam cham-ber can be maintained only when the oil in the reservoir is initially relatively immobile, which provides resistance to vertical steam fingering.

Reservoirs favorable for exploitation by SAGD methods must meet certain minimum require-ments (left). Ideally, SAGD candidate reservoirs should be free of laterally extensive shale barri-ers, which may prevent steam chamber growth or uniformity. A SAGD reservoir should also have minimal thief zones and have a pay thickness greater than 15 m [50 ft] to provide sufficient height for steam chamber growth. Additionally, the formation must be sealed by an impermeable top layer, or caprock. These criteria may be estab-lished via typical oil and gas exploration tools such as vertical pilot wells, logs, formation test-ing, seismic data and cores.

Thief zones, in the form of a water leg below the oil zone or gas above it, impact the effective-ness of the steam chamber. The thermal effi-ciency of the steam chamber may be compromised by the gas leg thief zone, and heated mobile oil may flow more readily to a water thief zone below the formation than to the production wellbore.

An indispensable element of most gas and oil zones is the presence of impermeable upper boundaries that isolate hydrocarbon-bearing intervals from surrounding formations. These bar-riers trap hydrocarbons in place to create reser-voirs. During production, the barriers ensure that oil or gas flows or is swept to the production well instead of migrating to neighboring formations.4

However, in SAGD wells, the caprock is exposed to continuous steam injection that may trigger complex thermal and hydraulic processes. It is, therefore, imperative that engineers plan-ning SAGD wells analyze the caprock to ascertain if and how these processes might alter critical

rock parameters of in situ stresses, rock strength or fracture systems. Engineers can then establish maximum safe operating pressures to ensure any effects on the caprock do not result in a contain-ment breach.5

How to Drill After an operator has deemed an oil sand forma-tion to be a candidate for exploitation through SAGD methods, engineers typically drill numer-ous pairs—a producer and an injector—of hori-zontal wells from a single pad. Each well has a length of 1,400 to 1,600 m [4,600 to 5,200 ft] mea-sured depth that includes about 800 to 1,200 m [2,600 to 3,900 ft] of horizontal section in the pay zone. Subject to operator specifications, produc-tion wells are placed above and as close to the base of the formation as possible, and the injec-tion wells are placed parallel to and about 5 to 6 m [16 to 20 ft] above the producers with no more than 2-m [6-ft] offset from the vertical plane con-taining the producer. Proper separation between the horizontal sections of the two wells is critical to ensure maximum recovery and efficiency. If the two are too close together, the steam will, in most cases, reach only the heel of the producer, resulting in inefficient recovery, lost production and poor asset economics. If the wells are too far apart, production could be delayed by months while a very large steam chamber is created.

A production well is drilled first using conven-tional directional drilling and MWD tools. An injection well is then drilled using conventional directional tools until the two well paths begin to converge. This typically occurs when the injector and producer are about 10 m [33 ft] apart and the injector is within 120 to 150 m [390 to 490 ft] of landing in the pay zone. This proximity of the injection well to the casing of the production well causes magnetic interference that renders con-ventional, magnetic-based MWD tools inaccurate.

Determining the position of one well relative to another well using magnetic measurements is called magnetic ranging; this method is commonly used for drilling planned well intersections such as those used for relief wells (next page, top right).6 At the point of magnetic interference, drillers may turn to active ranging, in which a magnetic source is conveyed in the producer by coiled tubing or a wireline tractor. When the MWD tool sensor pack-age is nearly perpendicular to the magnetic source, the latter is activated, and the resulting measurements taken by the MWD sensors allow technicians to calculate the spatial relationship between the two wellbores. Once the injection well position has been determined, the source is conveyed down the production well to the next

> Poor SAGD reservoir candidates. For an oil sand to be successfully exploited using SAGD methods, it must be free of shale barriers, or lenses (top), which may impede steam chamber growth or uniformity. The oil sand must also be free of thief zones (center) that may impact thermal efficiency or channel the steam chamber away from the production well. And the oil sand must meet minimum requirements of pay zone thickness (bottom) to provide room for development of an effective steam chamber.

Gas-bearing zone

Water-bearing zone

Steam injection well

Production well

Thin Pay Zone

Presence of Thief Zones

Presence of Shale Lenses

Caprock

Steam

Summer 2014 7

predetermined depth, the injection well is drilled ahead and the scenario repeated.

As an alternative to the active magnetic source method, engineers may use premagne-tized casing in the first well as a passive magnetic source (below right). Drillers then do not require access to both wells simultaneously and do not need a tractor or coiled tubing to move the source. Additionally, engineers are able to use standard directional drilling methods while obtaining a nearly definitive, real-time survey during drilling.7

Schlumberger has developed the RADAR real-time analysis of drilling and advanced ranging service to help operators accurately determine the relative position of two wells. The RADAR ser-vice is a suite of software programs that may be used to drill a second well parallel to and 5 to 6 m above an existing horizontal wellbore with a pre-cision of about 1 m [3 ft] over a length of 1 km [0.6 mi]. Among other applications, the RADAR service allows drillers to determine azimuth changes in magnetically challenging regions using gravity MWD tools, which are designed for use when magnetic interference prevents the use of a conventional MWD tool.

The nature of heavy oil sands causes other drilling problems. The bitumen and sand of the formation stick to the bottomhole assembly, gen-erating increased drillstring torque. Additionally, when the bitumen reaches the surface, it often

> Relative wellbore separation measurements. The proximity of the injection and production wellbores is critical to SAGD success and is measured as a relative separation between the two along their horizontal sections. This relationship is typically presented as a bull’s-eye with a target box (red). The production well, already drilled, lies at the center of the bull’s-eye, and the relative position of the injection well being drilled is displayed as a series of dots (blue) in the box, which represent survey points. In this display, the most recent survey point is represented by a green dot. Measurements include the following: toolface to target—the angle from injector to producer measured clockwise from the injector; distance—radial distance between wells; right side—the lateral displacement of the injection well relative to the production well measured from the vertical plane of the production well; and high side—vertical displacement of the injection well relative to the production well measured from the horizontal plane of the production well. The sensor measurement is taken at the measured depth (MD), and TVD is the true vertical depth of the injection well path at the measurement point. Inclination and azimuth of the injection well path are also taken at the measurement point.

Toolfaceto target168.85°

5.11 m

–0.99 m

5.01 m

1,221.22 m

477.04 m

90.00°

211.45°

Distance

Right side

High side

MD

TVD

Inclination

AzimuthDi

stan

ce, m

Production well

10

10

0

0°

180°

270° 90°

> Premagnetized casing pattern. Manufacturers premagnetize production well casing in a specific pattern to maximize the extruded magnetic field. A series of opposing poles direct the magnetization away from the casing and increase the distance over which accurate ranging is possible. The magnetic gaussing effect, or pattern, indicates flux direction (black lines), and flux intensity is indicated by color, ranging from most intense (magenta) to least (aqua). The amount of magnetization that can be imparted to the casing is a function of the amount of metal in the casing. The amount of magnetization imparted to the casing and the design of the magnetic pattern control the distance over which ranging can be reliably performed. (Adapted from Rennie et al, reference 7.)

Casing surfaceWellbore

Dist

ance

from

the

wel

lbor

e

2. For more on viscous fingering: Homsy GM: “Viscous Fingering in Porous Media,” Annual Review of Fluid Mechanics 19 (January 1987): 271–311.

3. Mobility is the ratio of permeability to dynamic viscosity and a measure of how easily a fluid can move through the formation. Because mobility is inversely proportional to viscosity, it improves as viscosity decreases in response to increasing temperature.

4. For more on faults and sealing: Cerveny K, Davies R, Dudley G, Fox R, Kaufman P, Knipe R and Krantz B: “Reducing Uncertainty with Fault-Seal Analysis,” Oilfield Review 16, no. 4 (Winter 2004/2005): 38–51.

5. Khan S, Han H, Ansari S and Khosravi N: “Geomechanical Modeling to Assess Caprock Integrity in Oil Sands,” presented at the Canadian Society of Petroleum Geologists, the Canadian Society of Exploration Geologists and the Canadian Well Logging Society Joint Annual Convention, Calgary, May 9–12, 2011.

6. Grills TL: “Magnetic Ranging Technologies for Drilling Steam Assisted Gravity Drainage Well Pairs and Unique Well Geometries—A Comparison of Technologies,” paper SPE/Petroleum Society of CIM/CHOA 79005, presented at the SPE International Thermal Operations and Heavy Oil Symposium and International Horizontal Well Technology Conference, Calgary, November 4–7, 2002.

Illfelder H, Forbes E, McElhinney G, Rennie A, Schaepsmeyer H and Krawchuk A: “A Systematic Approach for Wellbore Drilling and Placement of SAGD Well Pairs and Infill Wells,” WHOC paper 11-503, presented at the World Heavy Oil Congress, Edmonton, Alberta, Canada, March 14–17, 2011.

7. Rennie A, McElhinney G, Illfelder H, Ceh L, Schaepsmeyer H and Krawchuk A: “A Case Study of a New Technique for Drilling SAGD Twin Wells in Heavy Oil Reservoirs,” WHOC paper 2008-395, presented at the World Heavy Oil Congress, Edmonton, Alberta, March 10–12, 2008.

8 Oilfield Review

clogs the shaker screens of the mud treatment equipment, and if the sand and bitumen sepa-rate, the sand may build beds that block flow in the return line. Solvents in the mud system can dissolve the bitumen, but they may also cause unacceptable wellbore washouts.8

In response to these problems, researchers at M-I SWACO, a Schlumberger company, developed the SAGDRIL water-base drilling fluid. The sys-tem contains a strong water-wetting agent that minimizes sand accretion on the bit and tool-string and that encapsulates cuttings so that they are more easily removed by shakers and solids control equipment.

Thermal recovery methods also present zonal isolation challenges. During SAGD operations, downhole temperatures typically reach 275°C [530°F]. These elevated temperatures cause the well casing to expand, which imposes stresses on the surrounding cement sheath. To reduce these stresses and maintain well integrity, the cement used for isolation must have a thermal expansion coefficient similar to that of the

in volume, because of the increased pore volume of steam and thermal expansion of the steam chamber contents.

As the steam chamber is confined along its sides, most of the dilation manifests itself as uplift of the overburden. Uplifting the overbur-den stretches, or extends, the caprock laterally. Above the steam injector, lateral extension works against the horizontal principal compressive stresses. If, as a result, the minimum horizontal principal stress becomes tensile, the caprock will fracture in tension. Toward the sides of the steam chamber, lateral extension pushes outward and induces shearing stresses, which, if they exceed the shear strength, will cause shear fractures. These fractures become avenues of enhanced permeability that carry pressure and mobile fluid away from the steam chamber.10

Of overarching concern to SAGD operations is preservation of the caprock, which is exposed to many steam cycles throughout the life of the proj-ect. To establish the integrity of the caprock and estimate its response to cyclic heating in the Athabasca oil sands area in Alberta, Canada, engi-neers constructed geomechanical models from sonic log data, image logs, minifrac tests, forma-tion pressure sensor measurements and core analyses. These models allow analysts to estimate the induced stresses and changes in rock strength resulting from steam injection and to predict shear and tensile failure of the rock (left).

Researchers analyzed various injection sce-narios and used the ECLIPSE reservoir simulator to model changes in temperature (ΔT) and pres-sure (ΔP). The corresponding changes in stress, strain, porosity (Δφ) and permeability (Δk) were computed using the VISAGE 3D finite-element geomechanics simulation software. The values of Δφ and Δk were then fed back to the reservoir simulation model, which computed new ΔT and ΔP values. The new in situ stresses and stress paths—the ratio of the change in horizontal stress to the change in pore pressure—obtained from these models were checked against various failure criteria to predict possible occurrence and location of mechanical failure.11

Thermal Reservoir SimulationsWhile the SAGD method has been commercially successful for more than a decade, in the early days of its use, operators sometimes experienced disappointing recovery rates. These rates occurred partly because industry planners calculated reservoir response to steam based on simulation studies that assume oil sands are homogeneous. These assumptions, which have served reason-ably well for many years in traditional EOR proj-

casing. Additionally, the cement must not degrade when exposed to these high tempera-tures for extended lengths of time.9

ThermaSTONE thermally responsive cement is specifically designed for heavy oil and geother-mal applications. It sets at low temperatures, withstands high temperatures and features high flexibility, thermal stability and a high coefficient of thermal expansion. The cement can expand up to 2%, has a low Young’s modulus at steam condi-tions and has undergone laboratory testing to 344°C [651°F] for six months.

Geomechanics and Steam Injection Injecting high-pressure steam into oil sands has implications beyond testing the limits of steel and cement. It also challenges reservoir model-ing techniques. High-pressure steam injection into the steam chamber causes pore pressure and temperature to increase. Increasing pore pressure reduces the effective stresses—total stresses minus pore pressure—on the rock matrix. The steam chamber dilates, or increases

>Modeling the potential for caprock failure. Researchers employed a coupled reservoir simulator–geomechanical model to predict the effects of steam pressure on caprock integrity after three years of continuous steam injection at a rate of 200 m3/d [7 Mcf/d] and pressure of 3 MPa [435 psi]. The steam chamber was constrained to dilate primarily upward; adding heat induced horizontal tension (blue) in the reservoir above the chamber (top left ) and vertical tension (blue) near the sides of the chamber (bottom left ). Inside the steam chamber, the edges experienced added compression (yellow to red). This stress contrast can induce shearing stresses; however, in both cases, the caprock remained intact. To determine the maximum safe operating pressure, researchers increased the injection pressure to 6 MPa [870 psi], which is below the 7.35-MPa [1,070-psi] fracture pressure. After three years, the effective minimum horizontal compressive stress (top right ) had not reached zero (red); red would indicate caprock failure in tension. On the other hand, the shear failure index (bottom right ) indicated the caprock was close to failing (red) in shear.

Caprock

3 MPa 6 MPa

3 MPa 6 MPa

Steam chamber

Caprock failure

– 0 +

Change in minimum horizontal stress

– 0 +

Effective minimum horizontal stress

No failure 0 Failure

Shear failure index

– 0 +

Change in vertical stress

Summer 2014 9

ects, often caused engineers to inaccurately predict steam and pressure requirements and overestimate the volume of recoverable reserves within a bitumen reservoir.

That practice changed as SAGD experts real-ized that oil sands exhibit vast variations in geo-logic and reservoir properties. Taking advantage of recent improvements in simulation methods and computing technology, analysts today employ a fine-scale grid to capture the details of reser-voir heterogeneity and are able to run full-field models. Additionally, with greater computing power in hand, engineers are able to make simu-lations of SAGD pads with multiple wells and to account for the interplay of steam chambers for adjacent well pairs.12

Simulation models may be used to gauge the impact of SAGD completion options on produc-

8. Freeman MA, Stoian A, Potapinski JW, Elias LC and Tetreault R: “Novel Drilling Fluid Eliminates Tar Problems Associated with Drilling SAGD Wells,” paper SPE 90986, presented at the SPE Annual Technical Conference and Exhibition, Houston, September 26–29, 2004.

9. Tomilina EM, Chougnet-Sirapian A and Aboutourkia W: “New Thermally Responsive Cement for Heavy Oil Wells,” paper SPE 157892, presented at the SPE Heavy Oil Conference Canada, Calgary, June 12–14, 2012.

10. Collins PM, Carlson MR, Walters DA and Settari A: “Geomechanical and Thermal Reservoir Simulation Demonstrates SAGD Enhancement Due to Shear Dilation,” paper SPE/ISRM 78237, presented at the SPE and International Society of Rock Mechanics Conference, Irving, Texas, USA, October 20–23, 2002.

11. Khan et al, reference 5. For more on caprock integrity: Khan S, Han H, Ansari S,

Vishteh M and Khosravi N: “Caprock Integrity Analysis in Thermal Operations: An Integrated Geomechanics

tion, the steam/oil ratio (SOR) and project eco-nomics.13 Targeting a SAGD operation in the Athabasca oil sands of Alberta, Canada, one study used the Schlumberger Petrel E&P software

platform for static modeling and the ECLIPSE thermal reservoir simulator to test the impact of a completion strategy known as smart, or green, completions (above).

> Horizontal sections of three SAGD completion options. For conventional SAGD completions (left), both production and injection wells are cased; tubing is run to the toe of the producer, and the injector is completed with tubing halfway through the horizontal section. The last 610 m [1,970 ft] of both wells, below about 1,500 m [5,100 ft], were perforated. In smart SAGD completions (center ), both wells are cased, and tubing is run to the toe of both wells. Inflow control devices (ICDs) and packers are used to create individual sections in the injection well annulus. The horizontal sections of both wells are perforated only where there is a minimum of 5 m [16 ft] of continuous sand (blue and green). Sections that have less than 5 m of continuous sand (purple) are not perforated. Simple completions (right ) are cased and perforated along the entire horizontal section, and tubing is run only to the heel of both wells. (Adapted from Akram, reference 14.)

1,600

1,550

1,650

1,700

1,800

1,850

1,950

2,000

2,100

2,050

2,150

2,200

1,900

1,750

1,600

1,650

1,700

1,800

1,850

1,950

2,000

2,100

2,050

2,150

2,200

1,900

1,600

1,500

1,400

1,700

1,800

2,000

2,100

2,200

1,900

1,600

1,500

1,400

1,700

1,800

2,000

2,100

2,200

1,900

1,600

1,500

1,400

1,700

1,800

2,000

2,100

2,200

1,900

1,600

1,500

1,400

1,700

1,800

2,000

2,100

2,200

1,900

1,750

Rese

rvoi

r

Rese

rvoi

rRe

serv

oir

Rese

rvoi

rRe

serv

oir

Rese

rvoi

r

Rese

rvoi

rN

onre

serv

oir

Non

rese

rvoi

r

Non

rese

rvoi

rN

onre

serv

oir

Non

rese

rvoi

rN

onre

serv

oir

Non

rese

rvoi

rN

onre

serv

oir

Non

-re

serv

oir

Non

-re

serv

oir

Non-reservoir

Non-reservoir

Non-reservoir

Reservoir

Reservoir

Reservoir

Injection well Production well

MD, m MD, m MD, m MD, m MD, m MD, m

Conventional SAGD Completion Smart SAGD Completion Simple SAGD Completion

Injection well

Packer

ICD

Production well Injection well Production well

Tubing

TubingTubing

Tubing

Perforations

Perforations

Approach,” WHOC paper 11-609, presented at the World Heavy Oil Congress, Edmonton, Alberta, March 14–17, 2011.

12. Akram F: “Multimillion-Cell SAGD Models—Opportunity for Detailed Field Analysis,” WHOC paper 11-534, presented at the World Heavy Oil Congress, Edmonton, Alberta, March 14–17, 2011.

For more on optimization of SAGD well pairs through full-field and thermal simulations: Akram F: “Multi-Million Cell SAGD Models—Opportunity for Detailed Field Analysis,” paper SPE 11RCSC–SPE 145679, presented at the SPE Reservoir Characterisation and Simulation Conference and Exhibition, Abu Dhabi, UAE, October 9–11, 2011.

13. SOR, or steam/oil ratio, is a measure of the volume of steam required to produce a volume of oil. The ratio is commonly used to gauge the efficiency of a SAGD operation based on the assumption that the lower the SOR, the more efficiently the steam is used and the lower the fuel costs.

10 Oilfield Review

Engineers used the coupled model to deter-mine how the location of baffles and barriers within the reservoir would interfere with the desired steam flow path, allowing them to config-ure the completion so that steam would flow upward in the reservoir and avoid the obstruc-tions. Financial analysis was also performed using the Merak Peep planning, risk and reserves software to compare the economic outcomes of various technical options.

The study modeled and compared the conven-tional, smart and simple SAGD completion types over five years, reaching the following conclusions:• The conventional design achieved the best SOR

but because of high capital and operating expenditures (capex and opex) had the lowest rate of return on investment.

• The simple design achieved maximum recovery but required more steam and produced more water, increasing capex and opex not compen-sated for by incremental production increases.

• The smart design achieved optimized steam injection at slightly higher capex and lower opex, which resulted in the best net present value (NPV) of the three options.

Results of the study highlight the value of modeling thermal recovery operations and the potential pitfall of using a single indicator, such as SOR, to grade SAGD project success. Simulations showed that the conventional completion design produced the lowest SOR and that the simple completion design resulted in the highest cumula-tive oil production. However, when an economics model is included, the smart completion resulted in lower overall costs and yielded the best return on operator investment (left).14

Optimizing ProductionOptimal economic results using SAGD methods require uniform steam chamber growth, or uni-form conformance. However, the flow of bitumen and steam through the formation between SAGD well pairs is often irregular (next page, top right). Reservoir heterogeneities create uneven steam flow through the oil sands and varying oil phase mobility, which results in nonuniform oil flow. Additionally, steam is diverted by shale and mud layers. As a consequence, more than 80% of injected steam exits the well at the heel through the path of least resistance, and almost all the remaining steam exits at the toe.15 To improve conformance through injection control, opera-tors have used various strategies, including dual tubing strings inside slotted liners or other sand control screens for both the production and injection wells (left).

In the dual tubing configuration, one tubing string injects steam at the heel of the horizontal section of the injection well and a second tubing string carries steam to the toe. Because steam passes through the casing slotted liner into the formation along the entire horizontal length of the injection well, hydrocarbons enter the pro-duction tubing at both the toe and heel of the well. By placing injection and production points at both ends of the horizontal sections of both wells, flow is more evenly distributed between the well pair.

Dual tubing SAGD completions in Western Canada usually include gas lift rather than elec-tric submersible pumps (ESPs) to lift oil to the surface but do not have downhole control valves. Dual tubing completions may also contain an instrumented coiled tubing string with a distrib-uted temperature string or a thermocouple array. One study has proposed proportional integral derivative (PID) feedback controllers on each

> Incremental results from three completion strategies. Five-year forecasts for all three completion designs included casing, tubing, perforating, ICDs, packers, water treatment and recycling costs. Financial analysis of the smart completion yielded a higher net present value (NPV) over the same time period compared with the conventional and simple designs despite a higher production volume from the simple completion.

Production Volume,million bbl

Capex,Canadian

dollar, million

Opex,Canadian

dollar, million

NPV at 10%,Canadian

dollar, million

CompletionStrategy

6.46

7.47

7.89

8.753

7.778

7.385

265

304

333

63.3

76.2

74.4

Conventional completion

Smart completion

Simple completion

> Controlling steam injection and bitumen production in horizontal sections. When an operator completes a SAGD injection well (right ) with multiple tubing strings and a slotted casing liner, steam (red arrows) may then be injected into the casing-tubing annulus at both the toe and heel of the well to promote a more uniform injection profile along the length of the horizontal section. A production well completed with dual tubing strings and a slotted liner (left ) allows the gravity-driven bitumen and steam condensate (green arrows) to enter the tubing-casing annulus more evenly along the horizontal section. A proportional integral derivative feedback controller (not shown) monitors the temperature difference between the injected and produced fluids, or the subcool, through instrumented coiled tubing in the production well (red line) and regulates injection rates according to a target subcool.

Surface pipe

Injection Well

Production Well

Surface pipe

Intermediatecasing

Intermediatecasing

Heel string

Slotted linerSteam

Slotted linerHeel string

Toe string

Toe string

Gas lift string

Summer 2014 11

injector tubing string to control injection rates. The PID controller monitors the temperature dif-ference between the injected and produced flu-ids and maintains a specified difference between the two by regulating the rate of injection.16 The temperature difference between the injected steam and the produced fluids is a key control variable in SAGD operations and is called the subcool; it is typically maintained at between 15°C and 30°C [27°F and 54°F].17 Dual tubing completions with PID controllers have improved steam chamber conformance by controlling injection rates to maintain a specific subcool value as the reservoir conditions change.

A follow-up study aimed at optimizing produc-tion and NPV examined the use of PID controllers in SAGD well pairs. Researchers concluded that the controllers can adjust injection rates quickly and thus attain and maintain a targeted subcool and achieve efficient SORs. Additionally, because the same subcool target is used on both the heel and toe halves of the well pair, PIDs may be able to improve steam chamber conformance along the length of the well pair.18

Engineers may also attempt to create steam chamber conformance by installing inflow con-trol devices (ICDs) as part of a sand screen assembly in either the injection or production well or in both. ICDs are designed to cause the pressure distribution, or flux, along the length of the wellbore to vary. When installed as part of the injection well completion, ICDs serve to better equalize the toe-to-heel steam flux. When installed as part of the production well comple-tion, ICDs help equalize toe-to-heel influx of the steam-oil emulsion and thereby provide a more uniform toe-to-heel subcool (right).

14. Akram F: “Effects of Well Placement and Intelligent Completions on SAGD in a Full-Field Thermal-Numerical Model for Athabasca Oil Sands,” paper SPE/PS/CHOA 117704, presented at the SPE International Thermal Operations and Heavy Oil Symposium, Calgary, October 20–23, 2008.

15. Banerjee S, Abdelfattah T and Nguyen H: “Benefits of Passive Inflow Control Devices in a SAGD Completion,” paper SPE 165478, presented at the SPE Heavy Oil Conference Canada, Calgary, June 11–13, 2013.

16. Stone TW, Brown G, Guyaguler B, Bailey WJ and Law DH-S: “Practical Control of SAGD Wells with Dual Tubing Strings,” Journal of Canadian Petroleum Technology 53, no. 1 (January 2014): 32–47.

17. Gates ID, Kenny J, Hernandez-Hdez IL and Bunio GL: “Steam-Injection Strategy and Energetics of Steam-Assisted Gravity Drainage,” paper SPE/PS-CIM/CHOA 97742, presented at the SPE International Thermal Operations and Heavy Oil Symposium, Calgary, November 1–3, 2005.

18. Stone TW and Bailey WJ: “Optimization of Subcool in SAGD Bitumen Processes,” WHOC paper 14-271, presented at the World Heavy Oil Congress, New Orleans, March 5–7, 2014.

> Ideal and real steam chambers. An ideal steam chamber (left ) displays uniform steam distribution along the horizontal length of the injector well and permeates the formation evenly to efficiently drive bitumen to the production well below. In practice, without intervention, steam chambers are highly irregular and highly inefficient (right ).

Uniform shapeThree-quarterview

Side view

Even flow Varied flow

Uneven shape

Ideal Uniform Steam Chamber Irregular Steam Chamber

> Heel-toe effect. The steam-oil emulsion (blue) created by steam injection during SAGD operations tends to flow through higher permeability zones and arrive unevenly at the production well slotted liner, often at the well’s heel (top). Inflow control devices (ICDs) inside sand screen assemblies equalize the pressure drop along the entire length of the wellbore, promoting more evenly distributed emulsion flow through the formation and more uniform flow along the length of the horizontal production string (bottom).

Slotted liner

ICDs with sand screens

12 Oilfield Review

Nozzle-based ICDs are viscosity independent, and the pressure drop varies with the square of the velocity through the nozzles, providing high steam choking capacity. The nozzles therefore act as self-regulating valves in SAGD production well completions because as the liquid level comes into close proximity to the ICD sand screen, the liquids flash, or vaporize, inside the valve, causing additional flow restriction for the same pressure drop. This process works to dis-courage steam from entering the production wellbore; if steam does enter the wellbore, it is at a much reduced rate that will not cause localized erosion damage to the sand screen, known as “hot spots.” Consequently, SAGD completions with ICDs are able to improve conformance with-out the need for a second tubing string extending to the toe of the production well.19

Experts from Schlumberger ran wellbore simulations of a SAGD well pair that included a base case in which the producer was equipped with ICDs and the injector was completed as a dual string PID-controlled well. Steam was injected at a maximum rate of 250 m3/d [8,800 ft3/d]; the subcool target was 3°C [5.4°F]. For this study, researchers used FluxRite ICDs, now called MeshFlux ICDs, which are a combina-tion of MeshRite sand control technology and nozzle-type ICDs.

Installed with screens on 14 m [46 ft] long, 7-in. diameter basepipe, the production well ICD nozzle contained a 4.2-mm [0.17-in.] throat diam-eter; each well of the single SAGD well pair was 700 m [2,290 ft] long with 5-m [16-ft] vertical spacing. The simulated reservoir was based on available data for the McMurray formation in northeastern Alberta, Canada, which contains high viscosity bitumen at initial conditions and is highly heterogeneous.20

Four simulations were run:• In Case 1 (base case), the average tempera-

tures in the heel and toe halves of the producer were calculated using a temperature sort algorithm.

• In Case 2, the average temperatures in the heel and toe halves of the producers were calculated as an average of all inflowing temperatures.

• In Case 3, the target subcool changed from 3°C to 15°C.

• In Case 4, the producer was completed with dual tubing strings.

The study concluded that dual tubing string completions with PID controllers improved SOR and cumulative oil production. Use of a tempera-ture sort algorithm to screen out low tempera-tures improved calculation of the subcool; a lower subcool target resulted in improved production and economics.21 Use of ICDs in the production

completion resulted in a more stable pressure environment, more easily controlled production and more evenly distributed production along the entire horizontal length of the well than did pro-ducers completed with dual tubing strings.

Encouraged by reports of the impact of ICDs on production and efficiency in SAGD operations, engineers at Brion Energy performed a prelimi-nary study to quantify the potential benefits of liner-deployed ICDs. They used a reservoir model based on their Mackay River Commercial Project (MRCP), located about 30 km [18.7 mi] north-west of Fort McMurray, Alberta. Because the ini-tial model, which was based on ideal conditions and a perfectly homogeneous reservoir, did not show any benefit from the ICD, it was later replaced with one in which the absolute permea-bility of the reservoir cells on some of the planes perpendicular to the well trajectory were increased or reduced according to the maximum expected variation in the same reservoir area.

To accommodate the sand screens that are part of the ICD installation, the liner diameter was reduced from 85/8 in. to 7 in. Modeling indi-cated this size change had no impact on the well pair SOR and cumulative production. For eco-nomic and technical reasons, the team chose nozzle-type ICDs combined with a low-profile fil-ter media to allow the assembly to be run inside 95/8-in. casing.

With this configuration, simulation showed that the well pairs with the ICDs in the producers had a higher cumulative production and lower SOR than wells without ICDs; most of the produc-tion benefit occurred during the first two years. At the end of this period, cumulative production was 12.2% higher in liners with ICDs compared with the same wells without ICDs. After six years, that difference fell to only 2.5% higher. However, SOR was reduced by 9.84% at the end of Year 2 and 10.3% at Year 6. The company deemed these benefits sufficient to move ahead with field tests.

Prior to field installations, a more detailed dynamic simulation was performed using an actual well pair trajectory and an updated reser-voir geomodel in which the operator planned to run the first liner ICD completion. The simula-tion was run with a Petrel workflow using the ECLIPSE reservoir simulator in combination with a fully coupled multisegment well model. Also, based on the results of simulations using various nozzle sizes and downhole drawdown pressures, the operator chose to install two 2.5-mm nozzles per joint of liner in the producer, maintaining the wellbore subcool at 1°C [2°F].

> Bitumen production from standard SAGD wells with inflow control devices (ICDs). Simulations run by Brion Energy indicate that cumulative production (area under each flow rate curve) is higher from SAGD production wells that include two ICDs, each fitted with 2.5-mm nozzles per joint of tubing, than from base case, standard slotted liner production wells. Simulations were run using 2.5-mm nozzle ICDs at varying drawdown pressures (0, 25, 50, 75 and 100 kPa) below the standard well pair drawdown pressure. (Adapted from Becerra et al, reference 22.)

120

100

80

60

40

20

0 1 2 3 4 5 6 7 8 9 10 11 12 130

140Bi

tum

en fl

ow ra

te, m

3 /d

Years

Base case

2.5 mm, 0 kPa

2.5 mm, 25 kPa [4 psi]

2.5 mm, 50 kPa [7 psi]

2.5 mm, 75 kPa [11 psi]

2.5 mm, 100 kPa [15 psi]

Summer 2014 13

With the well drawdown pressure set at 70 kPa [10 psi] lower than that of a standard comple-tion, simulation results showed that the cumula-tive production could improve by 34% at Year 4 and 23% at Year 12 (previous page).

Based on the results of these simulations and concluding that ICDs have the potential to improve the performance of SAGD development, in October 2013, Brion Energy completed the first of two wells it planned to equip with ICDs. A second such well is planned for completion in 2014. Steam circulation is expected to begin dur-ing the second half of 2015, and production is expected to begin in the first half of 2016.22

Lightening the LoadAs for all oil and gas production operations, SAGD operators continually strive to improve production, reduce costs and minimize the envi-ronmental impact of their operations. In SAGD wells, production and costs are both driven by steam. Maintaining bitumen production from SAGD wells without mechanical intervention requires constant increases in the steam injec-tion rate and pressure to compensate for steam chamber leakoff and to help lift the oil-water emulsion to the surface. SAGD operators, know-ing such increases are unsustainable, have turned instead to artificial lift.

Operators investigated several artificial lift techniques and tools in the oil sands of Western Canada, including multiphase pumps, rudimen-tary gas lift and electric submersible pumps (ESPs). Because they had limited success with multiphase pumps and gas lift installations, oper-ators have opted to install ESPs. Engineers understood that for these pumps to be effective, they had to control the subcool at the pump intake. When the subcool becomes too low, steam is able to flow directly into the production string, and energy efficiency drops. Steam entering the slotted liner may also cause liner failures, sand production and pump cavitation if the intake pressure falls below the specified net positive suction head.23

ESPs have a history of solid performance in fairly shallow oil wells. However, service life is reduced significantly when ESPs are exposed to high bottomhole temperatures or when the con-ditions at the intake are such that water vapor or steam is present. To avoid this mode of failure, pumps must be manufactured of materials with higher tolerances for thermal expansion than those used in standard applications. The motor oil must be able to maintain its dielectric strength and lubricating properties in high tem-peratures, and the electric line to the motor must

be able to withstand constant submersion in high-temperature fluids.

To address these requirements, engineers from Schlumberger and ConocoPhillips designed and tested a high-temperature ESP in a flow loop at C-FER Technologies laboratories in Edmonton, Alberta. The facility made it possible for the team to use a variety of downhole instruments to moni-tor the new ESP performance in a high-tempera-ture environment (below). The REDA HotlineSA3 high-temperature ESP ran without failure for almost 42 days at fluid temperatures ranging from 150°C to 260°C [300°F to 500°F], which is the upper temperature design limit of the test loop.24

Real-Time Production NumbersWith time and experience, SAGD experts have significantly improved production and reduced costs of heavy oil recovery. Further fine-tuning of these operations requires timely and accurate flow rate data to optimize artificial lift efficien-cies, to adjust steam injection rates and pres-sures and to test and revise the reservoir models used to furnish production forecasts.

Capturing these data through traditional, grav-ity-based separation systems is a daunting task in SAGD wells because production fluids often have very small contrasts between water and oil densi-ties. Additionally, production from SAGD wells is usually marked by unstable flow regimes, high temperatures, emulsified foamy oil, hydrogen sul-fide [H2S] and abrasive sand particles.

These and other possible sources of error led engineers from Suncor Energy, in Calgary, and Schlumberger to conclude that flow rate mea-surements using traditional production monitor-ing methods were insufficient to enable SAGD well optimization. In 2007, engineers sought a way around these limitations by testing and qual-ifying a multiphase flow meter (MPFM) on a SAGD well.25

The MPFM was based on Vx multiphase well testing technology originally developed by Schlumberger engineers for deepwater applica-tions. The Vx system combines an instrumented venturi with a multienergy fraction meter and is able to measure total flow rate and fractions of

> Instrumenting an electric submersible pump (ESP) for high-temperature testing. By equipping an ESP with multiple sensors during laboratory testing, engineers were able to monitor surface and internal temperatures and vibrations at points where ESPs typically fail in high-temperature environments. (Adapted from Noonan et al, reference 24.)

Motor surfacetemperature sensor

Motor oiltemperature sensor

Motor windingtemperature sensor

Downhole horizontaland vertical vibrations sensor

near pump intake

Fluid intaketemperature sensor Produced fluids

19. Stone TW, Law DH-S and Bailey WJ: “Control of Reservoir Heterogeneity in SAGD Bitumen Processes,” paper SPE 165388, presented at the SPE Heavy Oil Conference Canada, Calgary, June 11–13, 2013.

For more on ICDs: Ellis T, Erkal A, Goh G, Jokela T, Kvernstuen S, Leung E, Moen T, Porturas F, Skillingstad T, Vorkinn PB and Raffn AG: “Inflow Control Devices—Raising Profiles,” Oilfield Review 21, no. 4 (Winter 2009/2010): 30–37.

20. Stone et al, reference 19.21. The temperature sort algorithm averages all

temperatures in the producing wells with the exception of the coolest temperatures in each half of the well if they were significantly lower than the hottest temperatures in each half of the well and affected permeability-height calculations.

22. Becerra O, Kearl B and Sanwoolu A: “A Systematic Approach for Inflow Control Devices Testing in Mackay River SAGD Wells,” paper SPE 170055, presented at the SPE Heavy Oil Conference Canada, Calgary, June 10–12, 2014.

23. Gaviria F, Santos R, Rivas O and Luy Y: “Pushing the Boundaries of Artificial Lift Applications: SAGD ESP Installations in Canada,” paper SPE 110103, presented at the SPE Annual Technical Conference and Exhibition, Anaheim, California, USA, November 11–14, 2007.

Pressure losses occur when liquids flow into a pump impeller. The net positive suction head is the minimum pressure required at the suction port of a pump to keep the pump from cavitating.

24. Noonan SG, Dowling M, D’Ambrosio L and Klaczek W: “Getting Smarter and Hotter with ESPs for SAGD,” paper SPE 134528, presented at the SPE Annual Technical Conference and Exhibition, Florence, Italy, September 19–22, 2010.

25. Pinguet B, Gaviria F, Kemp L, Graham J, Coulter C and Perez-Damas C: “SAGD Real-Time Well Production Measurements Using a Nucleonic Multiphase Flowmeter: Successful Field Trial at Suncor Firebag,” WHOC paper 11-514, presented at the World Heavy Oil Congress, Edmonton, Alberta, March 14–17, 2011.

14 Oilfield Review

26. For more on Vx technology: Atkinson I, Theuveny B, Berard M, Conort G, Groves J, Lowe T, McDiarmid A, Mehdizadeh P, Perciot P, Pinguet B, Smith G and Williamson KJ: “A New Horizon in Multiphase Flow Measurement,” Oilfield Review 16, no. 4 (Winter 2004/2005): 52–63.

27. Pinguet B, Gaviria F, Kemp L, Graham J, Coulder C, Damas C and Ben Relem K: “First Ever Complete Evaluation of a Multiphase Flow Meter in SAGD and Demonstration of the Performance Against Conventional Equipment,” presented at the 28th International North Sea Flow Measurement Workshop, St. Andrews, Scotland, October 26–29, 2010.

28. Gonzalez LE, Ficocelli P and Bostick T: “Real Time Optimization of SAGD Wells,” paper SPE 157923, presented at the SPE Heavy Oil Conference Canada, Calgary, June 12–14, 2012.

29. Mohajer M, Perez-Damas C, Berbin A and Al-kinani A: “An Integrated Framework for SAGD Real-Time Monitoring,” WHOC paper 2009-390, presented at the World Heavy Oil Congress, Margarita Island, Venezuela, November 3–5, 2009.

30. For more on DTS: Brown G: “Downhole Temperatures from Optical Fiber,” Oilfield Review 20, no. 4 (Winter 2008/2009): 34–39.

31. Mohajer et al, reference 29.32. Canadian Association of Petroleum Producers (CAPP):

“Crude Oil Forecast, Markets and Transportation,” Calgary: CAPP, June 2013.

gas, oil and water in multiphase production streams (above).26

In 2009, following numerous design changes based on results of the 2007 tests, the team pro-posed replacing a centralized test separator with a Vx MPFM at each of nine wellheads on a single pad at the Suncor Firebag project in northeast

Alberta (below). In addition to higher accuracy measurements from the MPFM, this arrangement would allow continuous flow measurements from each well. In the original arrangements, on the other hand, one separator per pad allowed engi-neers to test wells only intermittently for short time periods.

Flow measurements using the MPFM and the test separator for the same stable flow periods indicated consistent results between them. However, researchers found that the Vx meter consistently reported lower water/liquid ratio (WLR) measurements than did the test separa-tor. Investigation showed that the test separator was over-reporting water and under-reporting oil production. More significantly, the Suncor and Schlumberger team concluded from the results of the three-year project that the Vx technology had good dynamic response, repeatability and measured flow rates from SAGD wells with a high degree of accuracy, which made it well suited as an optimization tool.27

OptimizationApplying the SAGD method is capital intensive; steam generation costs account for the bulk of operating expenses. SAGD engineers continually strive to improve steam distribution along well pairs through the practice of real-time optimiza-tion (RTO).

SAGD operations, however, are complex and require that many parameters be monitored and controlled; the most important variables include steam injection rates, subcool and downhole tem-perature and pressure.28 The task of applying RTO practices to SAGD operations is further compli-cated by the fact that engineers derive each required parameter by combining data from numerous sources (next page).29 While these many variables make optimizing SAGD operations difficult, their complexity also means these opera-tions are good candidates for RTO solutions.

> Vx multiphase well testing technology. Vx meter measurements do not rely on fluids separation or flow calibration and are not impacted by foam or emulsions. The meter has no moving parts or sensors in direct contact with the fluid. Absolute and differential pressure measurements are made at the same location in the venturi throat. Nuclear-transparent windows in the venturi allow gamma rays to pass from source to detector with little loss caused by hardware. A flow computer provides sensor processing and flow rate data.

Nucleardetector

Flow computer

Nuclear source

Differentialpressuretransmitter

Flow

Venturi

Pressuretransmitter

> Firebag Project. The Suncor Firebag project, site of the Vx multiphase flow meter tests, is in northeast Alberta.

Alberta

Edmonton

Calgary

Alberta

CANADA

UNITED STATES

Suncor FirebagSAGD project

2000 mi

0 200kmArctic Ocean

Summer 2014 15

Two of the most important measurements for use in RTO—temperature and pressure profiles along the length of the horizontal sections—are available through optical fiber distributed tem-perature sensors (DTSs).30 And MPFMs supply a third critical piece of information—real-time surface flow rates for each phase.

For RTO, these critical data are subjected to basic quality checks using software to remove obvious errors such as negative pressures and extremely high or low temperatures. These results are then further refined by a more rigor-ous procedure to ensure all parameters obey the laws of thermodynamics and are physically realis-tic and resemble what the system has seen in the past. Missing or previously discarded data are replaced using estimates based on related

measurements. The measured data are quickly analyzed, and nonobvious relationships in a mul-tidimensional dataset are identified to expose hidden correlations or trends. Often, these correlations are strong enough to describe the behavior of the observed data as the result of only a few input parameters.31

Optimization may then proceed by comparing the subcool calculated from real-time DTS tem-perature measurements with a reservoir model and a target subcool range. When the system notifies the operator that the subcool value is out of range, engineers make changes to controls such as steam injection and multiphase pump rates. Ideally, these changes are made automati-cally in a closed loop system that constantly fine-tunes controls.

Heavy Oil FutureAccording to the Canadian Association of Petroleum Producers (CAPP), Canada produced 290,000 m3/d [1.8 million bbl/d] of oil from oil sands in 2012. Of that total, 130,000 m3/d [800,000 bbl/d] were from mining and the remainder from in situ methods, primarily SAGD. In that same report, CAPP predicted that by 2030, mining would account for 270,000 m3/d [1.7 million bbl/d] of production while in situ methods would increase to 560,000 m3/d [3.5 mil-lion bbl/d].32

The ratio of production volumes from SAGD methods to production volumes from mining is increasing in favor of SAGD operations because much of Western Canada’s bitumen is too deep to mine, and SAGD project capital and operating costs are significantly less than those for mining operations. Small SAGD projects can be profit-able and can be scaled up over time. Wells also have shorter lead times than mines; thus, compa-nies can react to changing markets. Additionally, whereas bitumen mining operations require removal of all top soil and overburden, SAGD wells impose a relatively small footprint, render-ing them much more environmentally attractive.

The oil sands of Canada offer exploration and production companies one other advantage: the reserves are known; thus, exploration costs and risks are virtually eliminated. Economic and environmental incentives, aided by the applica-tion of decades of upstream technology develop-ment, almost certainly will mean the oil sands of Canada will be a critical component of the global oil market for many years. —RvF

> Surface and downhole measurements. Engineers must use various techniques to measure all the required variables for monitoring, surveillance, diagnosis and optimization of SAGD well operations. (Adapted from Mohajer et al, reference 29.)

Downhole flow for ESP

Method

Flow Rate Measurement

ESP Lift Completion Measurement

Gas Lift Completion Measurement

Steam Injection Completion Measurement

Surface Measurement

Surface Measurement

Surface Measurement

Advantage

Downhole Measurement

Downhole Measurement

Downhole Measurement

Limitation

Episodic Measurement

Episodic Measurement

Episodic Measurement

True pump flow rate Single phase only; limitationon free gas

Well test using separators Readily available

Tubing pressure

Pump discharge pressure

Inconsistent, time-laggedresultsAffects system backpressure

Consistent and accurate

Ability to measure instability

Minimum interference with systempressures

Readings require adjustmentto stock tank conditions

Temporarily installed multiphasewell testing

Thermal profile survey withdistributed temperature system

Distributed temperature

Multiphase metering well testing

Pump intake pressure Flowing gradient survey of pressure, temperature and flow

Tubing pressure and temperature

Tubing pressure below the orifice Flowing gradient survey of pressure and temperature

Tubing pressure and temperature

Injection pressure and temperature

Thermal profile survey withdistributed temperature system

Distributed temperatureInjection rate

Injection rate

Multiphase flow data

Temporarily installed multiphasewell testing

Injection pressure and temperature Casing pressure below the orifice

Casing pressure

Pump flow rateTotal flow

Power Intake temperature

Vibration

Motor temperatureMultiphase flow rate

16 Oilfield Review

Ultradeep Scientific Ocean Drilling— Probing the Seismogenic Zone

High above the shoreline of Japan, stone tablets mark historical high water and

provide a centuries-old warning of the devastating effects of tsunamis. Off that coast,

deep within the Nankai Trough and the Japan Trench, a geologic record of seismo-

genesis and tsunamigenesis extends through the millennia. Scientists are probing the

depths of these subduction zones, and others worldwide, to better understand

geologic processes occurring at tectonic plate boundaries. With this information,

they hope to improve tsunami warning systems and mitigate seismic risk.

Nobuhisa EguchiKyaw MoeJapan Agency for Marine-Earth Science and TechnologyYokohama, Japan

Masafumi FukuharaSagamihara, Japan

Koji KusakaTokyo, Japan

Alberto MalinvernoLamont-Doherty Earth Observatory of Columbia UniversityPalisades, New York, USA

Harold Tobin University of Wisconsin–MadisonMadison, Wisconsin, USA

Oilfield Review Summer 2014: 26, no. 2.Copyright © 2014 Schlumberger.For help in preparation of this article, thanks to Lifeng Gan, Takashi Monden and Ushio Takahashi, Nagaoka, Japan; and Gokarna Khanal, Kuala Lumpur. adnVISION, arcVISION, FlexSTONE, FMI, Formation MicroScanner, geoVISION, MDT, PowerPulse, PowerV, RAB, seismicVISION, sonicVISION, TeleScope, UBI and VSI are marks of Schlumberger.

Subduction zone earthquakes are among the planet’s greatest natural hazards. The most dan-gerous ones are initiated within a depth range of approximately 5 to 40 km [3 to 25 mi], known as the seismogenic zone.1 At shallower and greater depths, faults may creep aseismically—without generating strong seismic waves. At shallow depths, stresses are usually too small to initiate major earthquakes. At great depths, rocks become ductile at high temperatures.

By drilling into seismogenic zones and study-ing core samples from them, scientists hope to clarify how material properties and stress fields affect fault slip, which can propagate to the sea-bed during earthquakes and generate tsunamis.2 The 2004 Sumatra-Andaman earthquake and subsequent tsunami and the 2011 Tohoku-Oki earthquake and tsunami in Japan demonstrated the potential for devastation associated with these natural phenomena.

In the aftermath of such destructive tectonic events, scientists have endeavored to understand earthquake-prone regions better. From 2003 to 2013, the Integrated Ocean Drilling Program (IODP) functioned as an international marine research collaboration dedicated to advancing scientific understanding of the Earth by monitor-ing and sampling subseafloor environments.3 The IODP initial science plan identified three princi-pal themes:• the deep biosphere and the subseafloor ocean• environmental change, processes and effects• solid earth cycles and geodynamics, including a

seismogenic zone initiative.The work of the 48 IODP expeditions, based

on these themes, built on that of its predecessors: Project Mohole, the Deep Sea Drilling Project

and the Ocean Drilling Program. Schlumberger has been involved in scientific deep-ocean drill-ing on many of these projects.

The IODP greatly extended the scientific com-munity’s capability to drill kilometers below the seafloor. Improvements in drilling technology, coring and logging techniques and interpretation techniques that link borehole measurements with core and seismic data facilitated drilling for scien-tific purposes. Many of these advances were devel-oped for oil and gas exploration. This article reviews the objectives of the recently renamed International Ocean Discovery Program (IODP). It then examines current and emerging technolo-gies that have enabled ultradeep scientific drill-ing, presents IODP seismogenic zone case studies and describes future directions and challenges.4

A Renewed Era in Scientific DrillingTo understand the history and structure of the Earth, the IODP conducts seagoing expeditions to study sediments and rocks beneath the sea-floor.5 The IODP science plan for 2013 to 2023 includes major themes related to past and future climate and ocean change, the deep biosphere, deep processes and their connection to and impact on the surface environment and Earth processes and hazards on human time scales. Geoscientists with the IODP study dynamic pro-cesses that lead to earthquakes, landslides and tsunamis; changes of in situ properties during an earthquake cycle related to fault rupture; and fluid flow in sediments and volcanic crust. They use long-term, subseafloor observatories in bore-holes for fluid and microbial sampling and to monitor stress and strain.

1. The seismogenic zone is the depth range of the Earth’s crust within which earthquakes are intitiated. Sections of some fault surfaces and tectonic plate interfaces are locked together and accumulate stress. Earthquakes occur when static friction is overcome, leading to fault slip and the radiation of seismic energy. Seismologists believe that this locking and release occur when dynamic friction is less than static friction and where fault friction exhibits velocity weakening. For more on the seismogenic zone: Dixon TH and Moore JC (eds): The Seismogenic Zone of Subduction Thrust Faults. New York City: Columbia University Press, 2007.

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

3. For more on the history of the IODP and its predecessors up to 2004: Brewer T, Endo T, Kamata M, Fox PJ, Goldberg D, Myers G, Kawamura Y, Kuramoto S, Kittredge S, Mrozewski S and Rack FR: “Scientific Deep-Ocean Drilling: Revealing the Earth’s Secrets,” Oilfield Review 16, no. 4 (Winter 2004): 24–37.

4. Many government agencies have defined deep water and ultradeep water as areas where water depths exceed 300 m [1,000 ft] and 1,500 m [5,000 ft], respectively.

5. For more on the new IODP: Bickle M, Arculus R, Barrett P, DeConto R, Camoin G, Edwards K, Fisher F, Inagaki F, Kodaira S, Ohkouchi N, Pälike H, Ravelo C, Saffer D and Teagle D: “Illuminating Earth’s Past, Present and Future: Science Plan for 2013–2023,” International Ocean Discovery Program: Exploring the Earth Under the Sea, http://www.iodp.org/program-documents (accessed April 20, 2014).

172500

3000

3500400

03500

20002000

000

2000

2500

3500

3500

2000

2000

2000

300000

2222002222000

00000

20000

2250000

2220000

220000

2002002020

33333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333335555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555555500000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000

17Summer 2014

JAMSTEC/IODP

18 Oilfield Review

The IODP seeks to advance this research by deploying state-of-the-art ocean drilling technolo-gies, by facilitating the dissemination of data and by providing scientific context to inform and increase global awareness of geohazards and envi-ronmental change. As of 2014, the IODP is finan-cially supported by 26 countries.6 The Center for Deep Earth Exploration (CDEX) in Japan is the riser drillship operator, and the US Implementing Organization (USIO) is the riserless drillship operator. The European Consortium for Ocean Research Drilling (ECORD) Science Operator (ESO) manages mission-specific operations.

The International Continental Scientific Drilling Program (ICDP) serves as an infrastruc-ture that facilitates scientific research drilling in onshore environments and now coordinates its activities with those of the IODP.7 Service contractors support MWD, LWD and wireline logging–related activities, offering technical expertise and providing drilling and completion services during expeditions.

Recent Technical AdvancesScientific drilling has three primary objectives: retrieval and analysis of core and fluid samples, acquisition of downhole measurements and installation of borehole observatories. Deepwater scientific drilling has historically faced a host of challenges such as compensating ship motion during drilling and measurement, maintaining borehole stability and balancing pore pressure while avoiding fracture initiation. Scientists require coring and logging equipment that can withstand the high temperatures and pressures in ultradeep boreholes, and cementing special-

> Riserless drilling vessel JOIDES Resolution. After nearly 25 years of service, the drilling vessel JOIDES Resolution (JR ) was retrofitted and updated for expanded capabilities. After sea trials, it returned to operations as the riserless drilling vessel for the Integrated Ocean Drilling Program (now the International Ocean Discovery Program, IODP) in February 2009. The ship is operating under a one-year extension of the previous 10-year award to the Consortium for Ocean Leadership. (Photograph courtesy of the IODP and Texas A&M University.)

> Riser drilling vessel Chikyu. The Center for Deep Earth Exploration, under the auspices of the Japan Agency for Marine-Earth Science and Technology (JAMSTEC), is responsible for the overall management and operation of the riser drilling vessel Chikyu. The Chikyu, which was the first riser-equipped drilling vessel for scientific research, supports both riser drilling and riserless drilling and is equipped with state-of-the-art drilling, core handling and laboratory facilities. The ship was damaged by the tsunami that struck Tohoku, Japan, in March 2011; it was returned to service later the same year and has been used to study the origin of the tsunami that caused its damage. (Photograph courtesy of IODP and JAMSTEC.)

6. Countries providing support to IODP are Australia, Austria, Belgium, Brazil, Canada, China, Denmark, Finland, France, Germany, Iceland, India, Ireland, Israel, Italy, Japan, Korea, the Netherlands, New Zealand, Norway, Poland, Portugal, Sweden, Switzerland, the UK and the US.

7. The International Continental Scientific Drilling Program (ICDP) and the International Ocean Discovery Program now jointly publish the journal Scientific Drilling. For more on the ICDP and its activities: http://www.icdp-online.org/ (accessed April 20, 2014).

8. For more on deepwater cementing: Cuvillier G, Edwards S, Johnson G, Plumb D, Sayers C, Denyer G, Mendonça JE, Theuveny B and Vise C: “Solving Deepwater Well-Construction Problems,” Oilfield Review 12, no. 1 (Spring 2000): 2–17.

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

10. For more on coring technology: Huey DP: “IODP Drilling and Coring Technology: Past and Present—Phase 2—Final Report,” Stress Engineering Services, Inc. report to IODP-MI (September 2009), http://www.iodp.org/doc_download/3464-iodp-drilling-coring-tech-final (accessed April 20, 2014).

11. For more on JOIDES Resolution : “Riserless Vessel,” International Ocean Discovery Program: Exploring the Earth Under the Sea, http://www.iodp.org/riserless-vessel (accessed April 20, 2014).

Summer 2014 19

ists must design cement systems that are effec-tive at low seabed temperatures. Time and cost constraints impose the need for efficient drilling and coring operations and timely core analysis.

To meet these challenges, dedicated research vessels have been commissioned and equipped with state-of-the-art drilling, measurement and well completion technologies.8 Services such as MWD and LWD are now essential in scientific drilling. Coring is at the heart of scientific drill-ing and a major activity aboard scientific drill-ships.9 Engineers have developed a variety of specialized core extraction methods.10 These technologies and methods are crucial for the suc-cess of the IODP, and with them, the research vessels serve as the vehicles for effectively and efficiently accessing the subsurface.

Research Vessel JOIDES ResolutionThe sole scientific drilling vessel for the Ocean Drilling Program was the research vessel JOIDES Resolution (JR); it also operated during the ini-tial riserless drilling phase of the Integrated Ocean Drilling Program (previous page, top). Since 1985, scientists working aboard the JR have completed more than 120 expeditions and have retrieved core samples in water depths ranging from 85 m [280 ft] to nearly 6,000 m [20,000 ft] and as deep as 2,100 m [6,900 ft] below seafloor (bsf).

Operators withdrew the JR from service in 2006 to conduct a major retrofit of the vessel; the modifications focused on enabling reliable, cost- and time-efficient drilling operations. Enhancing core quality and core recovery and increasing penetration rates in challenging lithologies and at extreme depths were the central objectives of this project. Improving the ability to deploy a wide variety of downhole logging instruments and sampling tools was an additional objective. The refitted JR provides new laboratory facilities, improved core handling capabilities, an expanded information and technology network and infra-structure, newly refurbished derrick and passive heave compensation systems, enhanced logging capabilities and larger and better-organized living and working spaces.11 The Texas A&M Foundation will manage the JR for a five-year period beginning in October 2014.

Using riserless drilling techniques, JR drillers pump seawater, the primary drilling fluid, down through the drillpipe and take drilling fluid returns to the seafloor (right). This process allows the operator to drill many shallow bore-holes in a relatively short period of time. Seawater is an inexpensive drilling fluid that cleans and cools the drill bit and flushes cuttings from the

hole. If better hole cleaning is necessary, seawa-ter is occasionally supplemented by “pills” of higher-viscosity mud. Hole cleaning and stability problems make it difficult to drill more than 1,000 m [3,300 ft] bsf or to exploit the hole for subsequent long-term observations. Drilling unconsolidated layers using seawater is difficult. The lack of traditional drilling mud additives and weighting materials may lead to hole collapse or fluid influx from higher-pressure formations should they be encountered. Geoscientists select drilling sites to avoid potential oil- and gas-bear-ing formations because of the risk of environmen-tal damage.

Riser drilling, which is standard practice in the oil and gas industry, became available to the IODP for scientific drilling in 2005. The tech-nique is more time consuming and costly than riserless drilling. Marine riser pipe connects the drillship to the blowout preventer (BOP) at the seafloor. The riser system includes an outer cas-ing that surrounds the drillpipe to provide an

annulus for return circulation of drilling fluid. The driller can control mud weight to counterbal-ance the formation fluid pressure and prevent collapse of the borehole.

Drillers use viscous drilling mud to displace drill cuttings, which scientists sample onboard. Cuttings provide a continuous record of subsur-face formations. Riser drilling also enables oper-ation crews to drill some unstable formations and sample active fault zones. Wireline logging can be conducted in both riserless- and riser-drilled holes. However, some tools that can be run through risers may be too large to run in riserless mode through narrower drillstrings. Among these are tools that measure downhole pore pressure and take fluid samples.

Drilling Vessel Chikyu The drilling vessel Chikyu is the first riser-equipped drilling vessel designed specifically for scientific research (previous page, bottom). Inaugurated in 2005, the ship, built as a national

> Riser and riserless drilling technologies. Riser drilling (left ) includes an outer casing that surrounds the drillpipe to provide an annulus for return circulation of drilling fluid to maintain pressure balance within the borehole. A blowout preventer protects the vessel from explosive overpressures of fluids. Riser drilling is required for deep boreholes, unstable or overpressured formations and where hydrocarbons may be encountered. Riserless drilling (right ) uses seawater as the primary drilling fluid, which is pumped down through the drillpipe. Riserless drilling can be employed in ultradeep water for boreholes that extend less than about 1,000 m [3,300 ft] below the ocean floor.

Drill bit

Drill bit

Seafloor

Cuttings

Seafloor

Uncased hole

Uncased hole

Second casing

Second casing

Surface casing

Surface casing

Blowout preventer

Drilling fluid ispumped down

through the drillpipe.

Drilling fluid ispumped down

through the drillpipe.

Riser

Drillpipe

Drillpipe

Drilling fluid and cuttings flow upbetween the drillpipe and the riser.

Drilling fluid and cuttings flow upbetween the drillpipe and theborehole or casing.

Drilling fluid and cuttings flow upbetween the drillpipe and theborehole or casing.

Drilling fluid and cuttings flow ontothe seafloor.

20 Oilfield Review

project funded by the Japanese government, is operated by the Japan Agency for Marine-Earth Science and Technology (JAMSTEC).12 The Chikyu provides access to the deep oceanic crust, targeting the underlying mantle, subduction zone environments, their associated seismogenic zones and geologic and biological systems in hydro carbon-prone regions.

The Chikyu supports both riserless and riser drilling. The ship is capable of riser drilling in 2,500-m [8,200-ft] water depth with borehole lengths up to 7,000 m [23,000 ft] bsf. Scientists at JAMSTEC consider 7,000-m water depth plus penetration of 1,000 mbsf the limit for riserless drilling—a limit affected by formation condition and sea state. Operators use GPS data from satel-lites, an acoustic underwater positioning system and azimuthal thrusters to control ship position dynamically. The ship can drill safely and contin-uously in surface currents up to approximately 2.1 m/s [4 knots], wind speeds up to 80 km/h [50 mi/h] and wave heights up to 4.5 m [15 ft].

Chikyu features include a 121-m [397-ft] der-rick, a sophisticated pipe handling system with a hydraulic roughneck and pipe racking system, a drilling mud circulation system, riser pipe and blowout preventers. State-of-the-art onboard core handling and analysis facilities include an X-ray computed tomography scanner laboratory, a microbiology laboratory, a sampling room enabling noncontaminated sampling in an anaer-obic environment, a core splitting room, a core laboratory for physical measurements, a magneti-cally shielded room for measuring rock paleo-magnetism and log-seismic data processing, integration and analyses facilities.

Benefitting from experience gained in the years since its launch, scientists at JAMSTEC continue to develop, foster and evaluate new technologies for the Chikyu. Engineers have introduced a system for real-time monitoring of riser motion and for enhanced riser drilling oper-ations in strong currents and ultradeep water. The drilling industry has developed a class of

high-strength drillpipe for ultradeep applications to 12,000 m [39,000 ft]; JAMSTEC engineers con-tinuously evaluate this drillpipe for fatigue life. Engineers are developing long-term borehole monitoring systems (LTBMSs) and advanced cor-ing technology such as a turbine-driven coring system, measurement-while-coring systems and high-temperature coring barrels.

Nankai Trough Seismogenic Zone ExperimentMegathrust earthquakes are large magnitude earthquakes that occur in subduction zones, where one of the Earth’s tectonic plates is thrust under another. One such zone is the Nankai Trough, south of Japan (above). This is an accre-tionary convergent margin along which sediment on the descending Philippine Sea plate is con-tinuously scraped off and added to the overriding Eurasian plate, forming an accretionary prism. The Nankai Trough is one of the world’s most active subduction zones, with a 1,300-year history of generating earthquakes that often cause

> Seismic survey and drilling area for the Nankai Trough Seismogenic Zone Experiment (NanTroSEIZE) Project. Stars mark the epicentral locations of two recent large magnitude earthquakes. The black outline defines the perimeter of the 3D seismic survey area. NanTroSEIZE drilling sites are marked with red dots. The Philippine Sea plate (PHS) subducts under the Eurasian plate (EP) in the Nankai Trough off the southwest coast of Japan (inset). The Pacific plate (PAC) subducts under the North American plate (NAP) in the north of Japan. Gray arrows indicate the plate convergence vector. (Adapted from Tobin et al, reference 15.)

Kii Peninsula

Kumano basin

Shikoku basin

35°

34°

33°

135° 136° 137° 138°

Tonankai, 1944

Nankaido, 1946

Site C0002Site C0001

Sites C0003, C0004, C0008

Sites C0006, C0007

500 mi

0 50km

EP

PAC

PHS

NAP

Site C0009

Site C0011

Site C0012

Site C0010

Latit

ude

Longitude

Summer 2014 21

tsunamis. It is also one of the most studied subduction zones. The 1944 Tonankai (M 8.1) and 1946 Nankaido (M 8.3) earthquakes are just two of the relatively recent major events associated with this subduction zone.13

The Nankai Trough Seismogenic Zone Experiment (NanTroSEIZE) is a multiyear ocean drilling project implemented by JAMSTEC on behalf of the IODP involving multiple expeditions led and staffed by multinational teams of scien-tists. The purpose of the experiment is to study the origins of subduction zone earthquakes; a primary objective is to understand why, during some earth-quakes, large displacement extends to the sea-floor, leading to tsunamis. Scientists hope to accomplish this by drilling into and acquiring data within the upper plate and down through the plate boundary approximately 7,000 mbsf. Seismologists are using these data to test hypotheses about the mechanisms that control the transition from aseis-

mic creep at shallow depth to intermittent locking and coseismic slipping at greater depth along the major plate interface, called the décollement. An additional major fault system, termed the mega-splay fault, branches from the décollement into the upper plate (above).

The partitioning of strain between the décol-lement and the megasplay system is not well understood, nor is the role of each in seismogen-esis and tsunamigenesis. To clarify this, scientists are studying the absolute mechanical strength of the plate boundary fault and the frictional and hydrologic processes thought to govern the mode of fault slip, strain accumulation and strain release. Establishing the state of stress in the region of the fault is critical to understanding the mechanism of seismic energy release.

The NanTroSEIZE project now comprises four stages. In Stage 1, the project team used riserless drilling and sampled at multiple sites to charac-terize the geology and provide geotechnical infor-mation for later stages. In Stage 2, the team performed riser drilling in the Kumano basin above the seismogenic zone and riserless drilling to sample subducting formations seaward of the Nankai Trough. The first observatory installations by the Chikyu targeted the shallow megasplay fault and the site of the future ultradeep bore-hole. In Stage 3, scientists are focused on drilling into the seismogenic zone, targeting the deep megasplay fault at approximately 5,200 m [17,000 ft] bsf and the deeper plate interface. In Stage 4, scientists will install a long-term observatory in the ultradeep borehole.

Investigators aboard the Chikyu are conduct-ing drilling and monitoring operations at sites offshore the Kii Peninsula along a line orthogonal to the Nankai Trough. Seismic crews have gath-

ered several generations of 2D survey data in the Kumano basin and farther offshore the Kii Peninsula. In 2006, Petroleum Geo-Services con-ducted a 3D seismic reflection survey in this area that provides high-resolution images of the accretionary complex. Analysts have used this seismic volume to refine the selection of drillsites and targets in the complex megasplay fault region, and IODP planned and completed 10 expeditions between 2007 and 2014 as part of NanTroSEIZE Stages 1, 2 and 3.

IODP Expeditions 314, 315, 316, part of Stage 1, were conducted from 2007 to 2008. The operations crew aboard the Chikyu completed drilling and coring in riserless mode at seven sites along the survey transect. These operations provided samples of the incoming sediment of the Shikoku basin and of the underlying oceanic crust seaward of the trench, samples of the fron-tal thrust system at the toe of the accretionary prism near the trench, samples of the midprism megasplay fault system landward of the trench and samples in the Kumano forearc basin. Early in these expeditions, scientists performed two feats: They drilled the deepest borehole— 1,401 m [4,596 ft] bsf—up to that point in the history of scientific ocean drilling while using LWD and for the first time penetrated the Nankai margin megasplay thrust system, which scientists believe to be involved in tsunami generation.14 Drillers also prepared 1,000 m deep boreholes at two sites planned for later deep penetrations of the seismogenic zone faults.

Researchers in the ocean drilling program have learned from experience that at convergent plate margins—in unstable formations such as those that occur in the accretionary prism—LWD is the best option to obtain high-quality logs.

12. For more on Chikyu and its specifications: “A New Frontier of Earth and Life Science: Deep Sea Drilling Vessel CHIKYU,” JAMSTEC: Center for Deep Earth Exploration, http://www.jamstec.go.jp/chikyu/eng/CHIKYU/index.html (accessed April 20, 2014).

13. Seismologists use the moment magnitude scale (abbreviated Mw or M) to classify earthquakes in terms of energy release. Introduced in the 1970s, the moment magnitude scale replaced the local magnitude Richter scale and corrected weaknesses associated with the older method while retaining a logarithmic scaling. An increase of one unit on the moment magnitude scale corresponds to an increase in the amount of energy released by a factor of approximately 32. For more on earthquake measurement techniques: “Measuring the Size of an Earthquake,” US Geological Service, Earthquake Hazards Program, http://earthquake.usgs.gov/learn/topics/measure.htm (accessed April 1, 2014).

14. For Expedition 314, Schlumberger provided geoVISION services for LWD resistivity and gamma ray; the sonicVISION tool for velocities and traveltimes; PowerPulse MWD for annular pressure and BHA direction and inclination; seismicVISION while drilling for interval velocities; and adnVISION density, porosity and ultrasonic caliper logs while drilling.

> Sites of borehole drilling during the NanTroSEIZE project along the Nankai Trough transect. Long-term borehole monitoring systems have been planned for installation at Sites C0002, C0009 and C0010. Holes at Sites C0006 and C0007 penetrate the frontal thrust portion of the accretionary prism and intersect the plate interface décollement near the trench axis. Engineers plan to extend Hole C0002F to penetrate the megasplay fault at the updip limits of the seismogenic zone. (Adapted from Expedition 332 Scientists, reference 22.)

Site C0009 Site C0002 Sites C0001, 03 Sites C0021, 18 Site C0006 Site C0007 Site C0011

Oceanic basement

Shikoku basinsediments

Accretionary prism

Subducting Philippine Sea plate

Kumano basin

Old accretionary sediments

Seismogenic zone

Trenchturbidites

Site C0012

Sites C0004,10 Site C0008

Site C0022De

pth

belo

w s

ea le

vel,

km

10

8

6

4

2

100 mi

0 10km

DécollementMegasplayfault

22 Oilfield Review

> Borehole stress indicators. A borehole subject to horizontal compressive stress may experience a breakout (top). (Adapted from Zoback et al, reference 15.) In RAB LWD resistivity-at-the-bit images at Sites C0002, C0001, C0004 and C0006 (bottom ), vertical paired dark bands indicate borehole breakouts. Geologists interpret breakout directions to generate the average σHmax azimuths. (Adapted from Tobin et al, reference 15.)

0

200

Dept

h, m

400

600

800

1,000

1,200

1,400

Conductive Resistive

N NSSite C0002 Site C0001 Site C0004 Site C0006

N NS N NS N NS

MPa0 150

Inducedfracture

Minimum horizontalcompressive stressdirection

σHminMaximum horizontalcompressive stressdirection

σHmax

Boreholebreakout

Using LWD measurements and core analysis, scientists succeeded in documenting the struc-ture, lithology, age, physical properties and state of stress of the accretionary complex and its fault system (above).15 Stress conditions differed at various positions above the seismogenic zone, in the accretionary prism and in the thrust sheet above the megasplay fault (next page).16 Stage 1 operations formed a necessary prelude to deep riser drilling to the seismogenic zone.

IODP Expeditions 319, 322, 332 and 333 were conducted in 2009 and 2010 as part of Stage 2. During these expeditions, Chikyu staff con-ducted the first riser operations of the scientific

ocean drilling program. Scientists acquired data for lithology identification, borehole imaging, environmental characterization, formation eval-uation and cement evaluation. Engineers con-ducted LWD, MWD and wireline logging to assess hole conditions and determine formation proper-ties of the cover sediments in the Kumano basin above the seismogenic zone and in the underly-ing accretionary prism.17 Because of the riser, log-ging engineers were able to acquire a variety of wireline measurements that were new to scien-tific drilling, including resistivity images from the FMI fullbore formation microimager and mea-surements with the MDT modular formation

dynamics tester.18 These tools are too large to be deployed through riserless drillpipe.

Engineers deployed the VSI versatile seismic imager tool as part of an extensive wireline seismic operation. The survey consisted of a zero-offset ver-tical seismic profile (VSP), a walkaway VSP with a 55-km [34-mi] line length and a circular VSP with a 3.5-km [2.2-mi] radius.19 The data acquisition used a combination of Schlumberger downhole seismic tools and acquisition software as well as source controlling and navigation systems. JAMSTEC per-sonnel onboard the source vessel Kairei deployed an extremely large gun array of 128,000 cm3 [7,800 in.3]. The data will help scientists analyze

Summer 2014 23

seismic velocities in the subduction zone forearc basin and identify the seismic attributes of the plate boundary in the region beneath the VSP bore-hole at depths of about 10 to 12 km [6 to 7.5 mi]. Analysts used zero-offset VSP results to verify or adjust depths of the 3D seismic reflection volume.

Logging engineers used the MDT tool to mea-sure pore pressure, permeability and stress.20 They conducted single probe tests to measure for-mation pore pressure and fluid mobility. They also conducted dual packer tests, including a draw-down test to measure formation hydraulic proper-ties and several hydraulic fracturing tests to determine the least principal stress magnitude. NanTroSEIZE scientists expect that by conduct-ing future tests deeper within the accretionary prism and in the vicinity of major fault zones, they will improve their understanding of in situ stress and subduction zone fault mechanics.

Riser drilling also enabled mud logging.21 During Expedition 319, analysts took cuttings samples for the first time in the history of IODP operations. A cuttings sample is assumed to be an averaged mixture of rock fragments and sedi-ments from a 5-m [16-ft] drilling interval. For this expedition, scientists estimated cuttings depth precision to be about 10 m [33 ft]. Interpreters used information from cuttings and cores as well as several of the LWD, MWD and wireline logging datasets to define lithologic units and to establish accurate ties to the 2006 Kumano basin 3D seismic reflection dataset.

15. Tobin H, Kinoshita M, Ashi J, Lallemant S, Kimura G, Screaton EJ, Moe KT, Masago H, Curewitz D and the Expedition 314/315/316 Scientists: “NanTroSEIZE Stage 1 Expeditions: Introduction and Synthesis of Key Results,” in Tobin H, Kinoshita M, Ashi J, Lallemant S, Kimura G, Screaton EJ, Moe KT, Masago H, Curewitz D and the Expedition 314/315/316 Scientists (eds): Proceedings of the Integrated Ocean Drilling Program 314/315/316, 2009, http://publications.iodp.org/proceedings/314_315_316/EXP_REPT/314315316_101.PDF (accessed May 15, 2014).

For more on stress determination from breakouts: Zoback MD, Barton CA, Brudy M, Castillo DA, Finkbeiner T, Grollimund BR, Moos DB, Peska P, Ward CD and Wiprut DJ: “Determination of Stress Orientation and Magnitude in Deep Wells,” International Journal of Rock Mechanics and Mining Sciences 40, no. 7–8 (October– December 2003): 1049–1076.

16. For stress direction and magnitude determined from LWD: Chang C, McNeill LC, Moore JC, Lin W, Conin M and Yamada Y: “In Situ Stress State in the Nankai Accretionary Wedge Estimated from Borehole Wall Failures,” Geochemistry, Geophysics, Geosystems 11, no. 12 (December 16, 2010), http://dx.doi.org/ 10.1029/2010GC003261 (accessed May 15, 2014).

For stress direction determined from LWD: Tobin H, Kinoshita M, Ashi J, Lallemant S, Kimura G, Screaton E, Moe TK, Masago H, Curewitz D and the IODP Expeditions 314/315/316 Scientific Party:

> Orientations of maximum horizontal compressive stress, σHmax, (red lines) inferred from borehole breakouts. Data at Sites C0001, C0002, C0004 and C0006 were acquired as LWD resistivity images during Stage 1. Data at Sites C0009 and C0010 were obtained from wireline FMI images during Stage 2. At Site C0002, the red and blue lines represent the orientation of σHmax in forearc basin and underlying accretionary prism sediments, respectively. At shallow sites in the accretionary prism and near the megasplay fault, σHmax trends NW–SE, roughly parallel to the plate convergence vector. At Site C0002 in the outer forearc basin, σHmax trends NE–SW, which is consistent with margin-normal extension. White arrows indicate a range of suggested convergence rates between the Philippine Sea plate and Japan on the Eurasian plate. (Adapted from McNeill et al, reference 16.)

Site C0002

Site C0001

Site C0010

Site C0006

Nankai Trough

Site C0009

Kumano basin

Site C0004

33°30’

33°20’

33°10’

33°00’

32°50’136°20’ 136°30’ 136°40’

Longitude

Latit

ude

136°50’ 137°00’

Philippine Sea plate–Eurasian plate

~4.1 to 6.5 cm/year

“NanTroSEIZE Stage 1 Expeditions 314, 315, and 316: First Drilling Program of the Nankai Trough Seismogenic Zone Experiment,” Scientific Drilling 8 (September 2009): 4–17.

For more on stress direction determined from wireline measurements: McNeill L, Saffer D, Byrne T, Araki E, Toczko S, Eguchi N, Takahashi K and IODP Expedition 319 Scientists: “IODP Expedition 319, NanTroSEIZE Stage 2: First IODP Riser Drilling Operations and Observatory Installation Towards Understanding Subduction Zone Seismogenesis,” Scientific Drilling 10 (September 2010): 4–13.

17. Measurements taken by LWD at Site C0009 included azimuthal resistivity images and laterolog resistivity at the bit. MWD measurements included rate of penetration, downhole torque, inclination and orientation of the hole, weight on bit, gamma ray emissions and downhole annular pressure while drilling.

Measurements from wireline logging at Site C0009 included bulk density, neutron porosity, photoelectric factor (PEF), laterolog resistivity, spontaneous potential (SP), natural and spectral gamma ray, sonic velocity (P- and S-waves), various types of calipers, mud resistivity and temperature.

18. For more on stress determination using FMI and MDT data: Lin W, Doan M-L, Moore JC, McNeill L, Byrne TB, Ito T, Saffer D, Conin M, Kinoshita M, Sanada Y, Moe KT,

Araki E, Tobin H, Boutt D, Kano Y, Hayman NW, Flemings P, Huftile GJ, Cukur D, Buret C, Schleicher AM, Efimenko N, Kawabata K, Buchs DM, Jiang S, Kameo K, Horiguchi K, Wiersberg T, Kopf A, Kitada K, Eguchi N, Toczko S, Takahashi K and Kido Y: “Present-Day Principal Horizontal Stress Orientations in the Kumano Forearc Basin of the Southwest Japan Subduction Zone Determined from IODP NanTroSEIZE Drilling Site C0009,” Geophysical Research Letters 37, no. 13 (July 2, 2010), http://dx.doi.org/10.1029/2010GL043158 (accessed May 15, 2014).

19. Modeling results produced by an industry contractor guided survey design of the Expedition 319 walkaway VSP experiment.

20. For more on stress determination in boreholes using the MDT tool: Saffer DM, Flemings PB, Boutt D, Doan M-L, Ito T, McNeill L, Byrne T, Conin M, Lin W, Kano Y, Araki E, Eguchi N and Toczko S: “In Situ Stress and Pore Pressure in the Kumano Forearc Basin, Offshore SW Honshu from Downhole Measurements during Riser Drilling,” Geochemistry, Geophysics, Geosystems 14, no. 5 (May 17, 2013), http://dx.doi.org/10.1002/ggge.20051 (accessed May 15, 2014).

21. For more on mud logging: Ablard P, Bell C, Cook D, Fornasier I, Poyet J-P, Sharma S, Fielding K, Lawton L, Haines G, Herkommer MA, McCarthy K, Radakovic M and Umar L: “The Expanding Role of Mud Logging,” Oilfield Review 24, no. 1 (Spring 2012): 24–41.

24 Oilfield Review

During Expedition 319, engineers prepared boreholes at Sites C0009 and C0010 for future installation of the LTBMS. Drillers used casing jetting and the PowerV vertical drilling rotary steerable system to drill an observatory well with vertical deviation of less than 0.2° and remain within the design specification of less than 1° required for effective downhole sensor installa-tion. Later, Expedition 332 engineers succeeded in installing an LTBMS observatory at Site C0002 for measurements of strain, tilt, seismicity, tem-perature and pore pressure (left).22 Long-term observation with the borehole sensors began in late 2011 after the installation of a recording sys-tem by a remotely operated vehicle deployed by the JAMSTEC research vessel Kaiyo. About a year later, engineers recovered the recording sys-tem for evaluation and established a connection to the Dense Oceanfloor Network System for Earthquakes and Tsunamis (DONET) submarine cable network.23

Expedition 319 scientists and crew drilled and cased the C0010A borehole and then instru-mented it with a temporary “smart plug” bore-hole monitoring system. This hole was located in the shallower portion of the megasplay fault branching from the seismogenic zone. The smart plug system is designed to document ambient conditions and provide proxies for strain and fluid flow related to tectonic and seismic activity within the megasplay fault; the smart plug achieved these tasks by recording formation pressures and temperatures. In November 2010 during Expedition 332, engineers recovered the smart plug and replaced it with a “genius plug.” The genius plug monitors pressure, temperature, microbial activity and fluid geochemical signa-tures of the screened interval of the shallow megasplay fault zone. Both instruments were mounted beneath a Baker Hughes retrievable casing packer, which was set just above a major out-of-sequence splay fault (next page, top left). Analysis of the smart plug data allowed scientists to identify prominent earthquake and tsunami events in the record from August 2009 to November 2010. The pressure sensor recorded many earthquake events and associated Rayleigh and tsunami waves, including those from the February 27, 2010, Mw 8.8 earthquake off the coast of Maule, Chile.

During Expedition 322, scientists used riser-less drilling and coring to investigate material destined to enter the seismogenic zone. They characterized the composition and state of off-shore sediments being conveyed toward the sub-duction system near the Kii Peninsula. The sedimentary structure consists of turbidite sedi-

> Long-term borehole monitoring system (LTBMS) observatory. At Site C0002, four lithofacies units were determined using LWD and MWD data and cores. Engineers installed an LTBMS at this site targeting the basal forearc basin (Unit 3) and the upper accretionary prism (Unit 4). Instrumentation for the LTBMS includes a deepsea borehole strain meter, a thermistor array for long-term temperature monitoring and a multilevel pore pressure monitoring unit with four ports (pink). In addition, a sensor carrier contains a broadband seismometer and tilt combo package housing a tiltmeter array, three-component geophones and accelerometers (none shown). A FlexSTONE cement slurry with nonshrinking characteristics was used to cement the strain meter and seismic instrument. The slurry was optimized and tested to be as compatible as possible to the formation material surrounding the installed strain meter; the Young’s modulus of the slurry needed to be a close match to that of the formation. A swellable packer at the target depth of 746 m [2,450 ft] bsf isolates the lower section of the borehole for pore pressure measurements. Sensors and cables were conveyed on tubing. Electrical lines and a flatpack containing hydraulic lines from pressure sensors connect to a circulation obviation retrofit kit (CORK) at the wellhead through which data are transferred to a recorder or submarine cable network. (Adapted from Expedition 332 Scientists, reference 22.)

Packer

CORK

Screens

Cement

Brine

Dept

h, m

bsf

Casing shoe

Thermistor array

Borehole

Unit 1

Seafloor

Unit 2

Unit 3

Unit 4

Pressure port

Flatpack

Sensor carrier

Strain meter

980

937

931

917

908

888

827

746

129

41

757

780

Summer 2014 25

ments overlying a smooth oceanic basement. Data characteristics will provide constraints on the initial conditions of the incoming sediments and ocean crust before they are subjected to higher pressure and temperature conditions as they are carried under the accretionary pile by the subduction conveyor. The condition of these sediments is an important factor affecting the onset of seismogenic fault behavior. Expedition 333 scientists conducted sediment coring and heat flow measurements of the sub-duction inputs and basement at the margin of the Philippine Sea plate.

Expeditions 326, 338 and 348 were conducted from 2010 to 2014 as part of NanTroSEIZE Stage 3. During this stage, scientists focused on deep drilling and coring into the seismogenic zone and across the plate interface into the sub-ducting crust using riser-based drilling and a carefully planned casing program. At Site C0002,

the oceanic crust subducts at depths of about 7,000 mbsf. The results of pilot hole operations, 3D seismic data and real-time LWD and MWD measurements guided the borehole design.

During Expedition 326, engineers prepared Hole C0002F for future ultradeep riser drilling by opening the borehole and installing the wellhead and casing to 860 m [2,800 ft] bsf. During Expedition 338, drilling continued in C0002F toward the megasplay fault, in the depth range to 3,600 m [12,000 ft] bsf, reach-ing 2,005 m [6,578 ft] bsf before operations were hindered by severe weather and by dam-age to drilling equipment.

Strong currents present another challenge in deepwater drilling. The Kuroshio current, a west-ern Pacific boundary current, flows northeast-ward with velocities up to 2 m/s [3.9 knots] along the south coast of Japan. Vortex induced vibration (VIV)—the transverse oscillation of a

pipe placed in high current—is caused by vortex shedding around the drillstring or riser and can lead to pipe fatigue damage. JAMSTEC engineers overcame this challenge and reduced VIV by installing fairings on riser joints located in the strong current zone (above). During

> A “smart plug” observatory. Scientists use a smart plug to monitor pressure and temperature changes within the megasplay fault zone at Site C0010. The smart plug was later replaced with a “genius plug,” which also measured microbial activity and fluid geochemical signatures. Casing screens allow fluids in the fault zone to enter the sensor chamber. (Adapted from Expedition 332 Scientists, reference 22.)

Corrosion cap

Bridge plug with instrumentpackage (smart plug) below

Megasplay fault

Casing screens

544 mbsf

555 mbsf

41 mbsf

20-in. casing

95/8-in. casing

Borehole

Suspension fluid

Cement

Float collar

> Battling the current. Without fairings that help reduce vortex induced vibrations (VIVs), riser joints move with the currents (right ) and are at risk of fatigue. The drilling crew on the Chikyu installed fairings on riser joints (left ) located in the strong current zone near the Nankai Trough. Fairings were designed to reduce VIVs and extend the life of the riser. Riser motion was monitored over a 2,000-m [6,600-ft] span. Multiship real-time current monitoring was employed during critical operations.

Seafloor

Fairing

Buoyancy material

Buoyant riser

Strong current

22. Expedition 332 Scientists: “Expedition 332 Summary,” in Kopf A, Araki E, Toczko S and the Expedition 332 Scientists (eds): Proceedings of the Integrated Ocean Drilling Program 332, 2011, http://publications.iodp.org/proceedings/332/EXP_REPT/CHAPTERS/332_101.PDF (accessed May 27, 2014).

23. DONET is a real-time submarine cable seafloor observation network in the Kumano-nada area designed to monitor earthquakes and tsunamis in the hypocentral region of the Tonankai earthquake. The Earthquake Research Committee in Japan predicts a 70% probability of another earthquake occuring there within the next 30 years. DONET consists of about 300 km [190 mi] of backbone cable, 5 science nodes and 20 observatories. Installation began in 2006 and was completed in 2011. JAMSTEC has also been developing DONET2, an expansion of the seafloor observatory network, to the west and off the Kii Channel.

26 Oilfield Review

Expedition 348, engineers aboard the Chikyu used riser drilling at Site C0002, obtaining cut-tings, mud-gas samples, direct core samples and LWD data. Drillers ultimately reached 3,059 m [10,036 ft] bsf on this expedition. The hole will be extended in future expeditions toward the mega-splay fault at about 5,200 m [17,060 ft] bsf (above).24

During NanTroSEIZE Stage 4, scientists plan to install a final LTBMS in the ultradeep bore-hole. The monitoring station will acquire data to help them ascertain the behavior and evolution of the plate interface fault system during a sig-nificant portion of the seismic cycle of strain rate variation.25 The sensor and sample data may

clarify how water and rock interact in the Nankai subduction zone to influence the occurrence and magnitude of earthquakes and tsunamis. The LTBMS will be connected to the DONET seafloor fiber-optic monitoring system.

Costa Rica Seismogenesis Project IODP scientists are also conducting studies on the eastern rim of the Pacific Ocean as part of the Costa Rica Seismogenesis Project (CRISP). In the Costa Rica margin subduction zone near the Osa Peninsula in Central America, the Cocos plate and the Cocos Ridge are currently subduct-ing beneath the Caribbean plate (next page, top).

This is a well-studied area, with a relatively shal-low seismogenic zone. The collision of the Cocos Ridge with the Middle America Trench (MAT) offshore the Osa Peninsula brings the trench close to the coastline and puts the seismogenic zone within reach of new IODP riser drilling capabilities. In June 2002, an M 6.4 thrust earth-quake occurred at a depth of about 9 km [5.6 mi] in the CRISP drilling area. One possible loca-tion chosen for deep riser drilling in this area targets the seismogenic zone at a depth of about 6 km [4 mi].

The primary objective of CRISP is to investi-gate seismogenic processes such as dynamic fric-tional weakening, thermal pressurization or melt lubrication, which are common to most faults. However, CRISP is different from but complemen-tary to other deep fault drilling projects such as NanTroSEIZE and the Japan Trench Fast Drilling Project, described below. Subduction in the CRISP region occurs at an erosional convergent margin where input material to the seismogenic zone is scraped off the bottom of the overriding plate. This material has unknown lithologic, phys-ical and frictional properties, except where revealed by deep drilling. CRISP provides the opportunity to learn about the mechanisms of earthquake nucleation and rupture propagation at an erosional subduction zone.

CRISP is a multiplatform, multisite, multi-phase drilling project administered by the USIO for IODP. The USIO completed Program A, the first phase of the project, using the JR for riser-less drilling during Expeditions 334 in 2011 and 344 in 2012 (next page, bottom). Scientists obtained borehole data by coring and logging while drilling.26 During Expedition 334, the JR visited four sites, acquiring LWD data at Sites U1378 and U1379 and coring at Sites U1378 to U1380 on the MAT slope and at Site U1381 on the Cocos plate. Site U1379 is located above the frontal prism at a water depth of 127 m [417 ft]. Site U1381 is in the subducting plate in water depth of about 2,000 m.27 During Expedition 344, the JR visited five sites, coring at Sites U1380, U1412 and U1413 on the MAT slope in the overrid-ing plate and at Sites U1381 and U1414 seaward of the MAT slope in the subducting Cocos plate. Engineers attempted riserless wireline logging at Site U1380, but tool descent was limited because of borehole obstructions. They successfully com-pleted shallow logging runs at Site U1413 and to a depth of 421 m [1,380 ft] bsf at Site U1414.28

The USIO succeeded in meeting its scientific objectives for CRISP Program A. Scientists used core and log data to characterize the lithology, stratigraphy and age of the slope and incoming

>Well plan for a deep riser hole. This seismic section is overlain with the initial well plan for Hole C0002F. During Expedition 348, drilling progressed only to 3,059 mbsf, where an 113/4-in. liner was set. Engineers plan to extend the hole through the megasplay fault (dashed green line) in future Expedition 3XX and eventually to the oceanic crust (dashed blue line). Hole C0002G contains the shallow LTBMS. Interpreters identified a shallow bottom-simulating reflector (BSR), possibly related to reflection at an interface containing gas hydrates. (Adapted from Hirose et al, reference 24.)

NW Hole C0002FHole C0002G LTBMS SE

Dept

h, k

m

2

3

4

5

6

7

8

9

1010 mi

0 1km

Accretionary prism

Megasplay fault

Oceanic crust

Forearc basin strata 856 m: 20-in. casing depth

2,300 m: 13 3/4-in. casing depth,Expedition 348

4,400 m: 9 5/8-in. casing depth,Expedition 348

3,600 m: 11 3/4-in. casing depth,Expedition 348100-m coring interval

5,200 m: target depth,Expedition 3XX300-m coring interval

BSR

Summer 2014 27

sediments; the petrology of the subducting Cocos Ridge; and the influence of the ridge subduction on the evolution of the Central American volca-nic arc. Geochemical and temperature data sug-gest fluid transport from greater depths. Log

24. Hirose T, Saffer DM, Tobin HJ, Toczko S, Maeda L, Kubo Y, Kimura G, Moore GF, Underwood MB and Kanagawa K: “NanTroSEIZE Stage 3: NanTroSEIZE Plate Boundary Deep Riser 3,” (2013), http://publications.iodp.org/scientific_prospectus/348/348SP.PDF (accessed May 27, 2014).

25. For more on the siting and use of a borehole observatory in the seismogenic zone at the San Andreas Fault in the US: Coates R, Haldorsen JBU, Miller D, Malin P, Shalev E, Taylor ST, Stolte C and Verliac M: “Oilfield Technologies for Earthquake Science,” Oilfield Review 18, no. 2 (Summer 2006): 24–33.

26. Scientists acquired LWD logs during CRISP Program A using adnVISION, arcVISION, geoVISION and TeleScope tools.

27. Expedition 334 Scientists: “Expedition 334 Summary,” in Vannucchi P, Ujiie K, Stroncik N, Malinverno A and the Expedition 334 Scientists (eds): Proceedings of the Integrated Ocean Drilling Program 334, 2012, http://publications.iodp.org/proceedings/334/EXP_ REPT/CHAPTERS/334_101.PDF (accessed July 15, 2014).

28. Harris RN, Sakaguchi A, Petronotis K, Baxter AT, Berg R, Burkett A, Charpentier D, Choi J, Diz Ferreiro P, Hamahashi M, Hashimoto Y, Heydolph K, Jovane L, Kastner M, Kurz W, Kutterolf SO, Li Y, Malinverno A, Martin KM, Millan C, Nascimento DB, Saito S, Sandoval Gutierrez MI, Screaton EJ, Smith-Duque CE, Solomon EA, Straub SM, Tanikawa W, Torres ME, Uchimura H, Vannucchi P, Yamamoto Y, Yan Q and Zhao X: “Expedition 344 Summary,” in Harris RN, Sakaguchi A, Petronotis K and the Expedition 344 Scientists (eds): Proceedings of the Integrated Ocean Drilling Program 344, 2013, http://publications.iodp.org/proceedings/344/EXP_REPT/CHAPTERS/344_101.PDF (accessed May 28, 2014).

> Topographic and bathymetric map near the Middle America Trench. The Costa Rica Seismogenesis Project (CRISP) drilling area is near the Osa Peninsula, Costa Rica, where the Cocos plate and Cocos Ridge subduct beneath the Caribbean plate. (Adapted from Expedition 334 Scientists, reference 27.)

20°

18°

16°

14°

12°

10°

8°

6°

4°

–4°

2°

–2°

0°

104° 102° 100° 98° 96° 94° 92° 90° 88° 86° 84° 82° 80° 78° 76°

Cocosplate

North American plate

Caribbeanplate

Cocos

Ridge

Plate convergence rate,88 mm/year

Guatemala basin

Galapagos hotspot

Carnegie Ridge

MiddleAmericaTrench

Coiba RidgePanamaFracture

Zone

Osa Peninsula

Proposedarea ofdrilling

2002 Mw 6.4

Malpelo Ridge

Cocos Ridge

MEXICO

COSTA RICA

–8,500

–7,000

–6,000

–5,000

–4,000

–3,000

–2,000

–1,000

1,0002,0003,0004,0005,640

0

Elev

atio

n, m

Latit

ude

Longitude

> Structure of the Costa Rica subduction zone with CRISP drilling sites. An interpreted wide-angle seismic section (top) shows borehole locations and reflections from the dipping plate boundary. Dots mark the top of the upper plate basement, which is overlain by thick slope sediment. The schematic cross section (bottom) through the Osa Peninsula margin shows Sites U1381, U1412, U1378, U1380 and U1379 and subsurface P-wave velocities (km/s) within the upper plate. (Adapted from Harris et al, reference 28.)

Dept

h, k

mDe

pth,

km

Distance, km

Frontal prism

Plate boundary Plate boundary

Distance, km

Site U1381,Reference

Site U1381 Site U1412

Vertical exaggeration = 2

Vertical exaggeration = 1.3

Plate boundary

V = 3.5V = 4.8

Sediment Frontalprism

Wedge

Site U1412,Toe

Sites U1378and U1380,Midslope

Site U1378 Site U1380Landward-dippingreflections

Site U1379

Site U1379,Upper slope

0

00

1

2

3

35 30 25 20 15 10 5

4

5

6

7

89

5

1050 40 30 20 10 0 –10 –20

Top basement

28 Oilfield Review

analysts used borehole breakout and core analy-sis to determine in situ stress orientation at the updip limit of the seismogenic zone. Using cali-per data from LWD nuclear and ultrasonic tools obtained during Expedition 334, they determined that the direction of the maximum horizontal stress varied as a function of position along the trench slope at Sites U1378 and U1379. Using wireline resistivity and ultrasonic image data obtained during Expedition 344, they found that the orientation of the maximum stress at Site U1413 on the upper slope is nearly orthogo-nal to relative plate motion and suggests an extensional stress regime (above).29 Similar vari-ations of stress orientation in the overriding plate have also been observed at the Nankai Trough.30 In a future CRISP Program B, scientists onboard the drilling vessel Chikyu will continue the study using riser drilling technology for penetration of the megathrust.

Japan Trench Fast Drilling ProjectOn March 11, 2011, the (Mw 9.0) Tohoku-Oki earthquake occurred on the megathrust fault west of the Japan Trench at an erosive conver-gent margin where the Pacific plate subducts below the North American plate (next page, top right). The ensuing tsunami devastated large sec-tions of the northeast coast of the island of Honshu, Japan. Scientists estimated fault move-ment to have exceeded 50 m [165 ft], which is

among the largest displacements ever recorded in an earthquake. Seismologists had not expected such a large slip on the shallow portion of the megathrust boundary because this portion of the subduction zone was not thought to be locked and accumulating significant stress. However, subsequent analysis suggested that such large slips may result when shear stress on the fault drops to very low levels during an earthquake.31

Scientists, long concerned with seismic haz-ards, have sought to mitigate their catastrophic effects by studying earthquake mechanisms. Participants in a 2009 ICDP international work-shop had proposed strategies and made technical recommendations for rapidly mobilizing a drill-ing program following future large earthquakes.32 The tragedy of the results of the Tohoku-Oki earthquake emphasized the need and provided an opportunity for realizing such a program.

The Japan Trench Fast Drilling Project (JFAST) emerged as a rapid response project for several reasons.33 Seismologists theorize that fric-tion controls the dynamics of large ruptures on faults. Friction generates heat during an earth-quake, and the fault temperature observed after an earthquake provides insight into the level of friction. One of the most direct ways to estimate the dynamic friction during the earthquake is to measure the residual heat at the fault zone. However, the temperature signal diminishes with time. To resolve residual temperature, geophysi-

cists theorize that measurements need to begin within two years following the earthquake. Important time-sensitive measurements needed for reliable estimates of friction include the fault temperature, fault zone permeability and chemi-cal properties of the fluids and rock.

The purpose of the JFAST project was to understand the reasons and mechanisms for the large fault slip of the Tohoku-Oki earthquake; investigators began by drilling to the plate bound-

29. Malinverno A and Saito S: “Borehole Breakout Orientation from LWD Data (IODP Exp. 334) and the Present Stress State in the Costa Rica Seismogenesis Project Transect,” paper T34C-07, presented at the American Geophysical Union Fall Meeting, San Francisco, December 9–13, 2013.

30. Lin et al, reference 18. 31. Lin W, Conin M, Moore JC, Chester FM, Nakamura Y,

Mori JJ, Anderson L, Brodsky EE, Eguchi N and the Expedition 343 Scientists: “Stress State in the Largest Displacement Area of the 2011 Tohoku-Oki Earthquake,” Science 339, no. 6120 (February 8, 2013): 687–690.

32. Brodsky EE, Ma K-F, Mori J, Saffer DM and the participants of the ICDP/SCEC International Workshop: “Rapid Response Fault Drilling Past, Present, and Future,” Scientific Drilling 8 (September 2009): 66–74.

33. For detailed information about the JFAST project: “Importance for Understanding the Devastating Tsunami from the 2011 Tohoku Earthquake,” JAMSTEC: Japan Trench Fast Drilling Project, http://www.jamstec.go.jp/chikyu/exp343/e/science.html (accessed April 18, 2014).

34. Expedition 343/343T Scientists: “Expedition 343/343T Summary,” in Chester FM, Mori J, Eguchi N, Toczko S and the Expedition 343/343T Scientists (eds): Proceedings of the Integrated and Ocean Drilling Program 343/343T, 2013, http://publications.iodp.org/proceedings/343_343T/EXP_REPT/CHAPTERS/ 343_101.PDF (accessed May 15, 2014).

> Stress directions in the CRISP drilling area. Directions of maximum horizontal compressive stress, σHmax, (red lines) were determined from a 16-track LWD density caliper at Sites U1378 and U1379 and from Formation MicroScanner and UBI ultrasonic borehole imager wireline logs at Site U1413. Sites at which holes were drilled during Expeditions 334 and 344 are shown with red and yellow dots, respectively. Sites at which operations occurred on both expeditions show both colors. The black arrow shows the direction of motion of the Cocos plate. A NNE deformation vector (not shown) was determined from GPS on land. Stress directions, including the vertical stress, are consistent with NNW–SSE compression at midslope Site U1378 near the frontal prism and with NNW–SSE extension at Sites U1379 and U1413 in shelf sediments on the upper slope. (Adapted from Malinverno and Saito, reference 29.)

8°15’

9°00’

8°45’

8°30’

84°45’ 84°30’ 84°15’ 84°00’

Relativeplate

motion

OsaPeninsula

Longitude

Latit

ude

83°45’ 83°30’ 83°15’

σHmax

Summer 2014 29

ary fault zone and measuring the rock properties and the frictional heat from the fault movement. Scientists expected to use estimates of dissipated heat energy to infer the forces that acted on the fault during the earthquake. Objectives of the drilling included using LWD to locate the fault that ruptured, taking core samples to character-ize the composition of the fault zone and the mechanisms of the slip and healing processes along the fault, and placing a temperature mea-surement observatory across the fault to estimate the frictional heat and stress within and around the fault zone. An international rapid response team achieved these objectives using the drilling vessel Chikyu as the drilling platform during IODP Expeditions 343 and 343T.34

The JAMSTEC Center for Deep Earth Exploration (CDEX) acted as the implementing organization for the JFAST project, conducting Expedition 343 from April to May 2012, 13 to 14 months after the Tohoku-Oki earthquake. Using riserless drilling, engineers made LWD measurements in a borehole drilled to 850.5 m [2,791 ft] bsf at a drilling site on the landward slope of the Japan Trench. Total depth of the hole was 7,740 m [25,394 ft] below sea level (bsl) (below). Geologists used LWD data from the geoVISION imaging-while-drilling service and arcVISION array resistivity compensated tools to identify and characterize the plate boundary fault zone. On this expedition, scientists also acquired 21 cores that spanned the two main fault targets in a separate coring hole drilled to 844.5 m [2,771 ft] bsf.

> Topography of Japan and bathymetry of the Pacific Ocean near the Japan Trench. The black star marks the epicenter of the 2011 Tohoku-Oki earthquake. The red dot marks the drilling site of IODP Expeditions 343/343T for the JFAST project. The Pacific plate (PAC) subducts under the North American plate (NAP) at the Japan Trench offshore of the northern part of Japan (inset).

–200

–1,500

–3,500

–5,000

Heig

ht re

lativ

e to

sea

leve

l, m

Latit

ude

Longitude

–8,000

–10,000

44°

43°

42°

41°

40°

39°

38°

37°

36°

35°

34°138° 139° 140° 141° 142° 143° 144° 145° 146° 147°

JapanTrench

Drilling siteTohoku-Oki, 2011

EP

PAC

PHS

NAP

> Interpreted seismic section from Line HD33B through JFAST Site C0019 near the Japan Trench. The Pacific plate subducts beneath the overthrust frontal sedimentary prism of the North American plate. At this site, three holes were drilled through the sedimentary prism and the décollement. LWD measurements were made in Hole C0019B (holes not shown). Coring was performed in Hole C0019E. Scientists installed an observatory in Hole C0019D to make temperature measurements in the seismogenic zone near the décollement. (Adapted from Expedition 343/343T Scientists, reference 34.)

Frontalsedimentary prism

Décollement

Horst GrabenBasalt top

Pacific plate: igneous oceanic crust

Trench fill,mass transport deposit

Normal faults

Bedded chert topPelagic claystone top

Hemipelagic mudstone

Line HD33B

Dept

h, m

bsl

mbsfSite C0019

0

200

400

600

800

1,000

6,500NW SE

7,000

7,500

8,000

8,500

10 mi

0 1km

30 Oilfield Review

Investigators have determined that the over-all structure at the drillsite consists of an over-riding prism of faulted and folded mudstones in fault contact with a thin sequence of sediments that were deposited on top of the incoming Pacific plate. A fault contact at about 820 m [2,690 ft] bsf is defined by a thin zone of highly sheared clay and is interpreted as the plate boundary décollement. Scientists concluded that the plate boundary décollement was the most likely location for the fault slip that occurred during the 2011 Tohoku-Oki event. Using fluid analysis, they identified another pos-sible location of recent fault motion, at about 700 m [2,300 ft] bsf. Comparing image logs with observations from cores, they identified a frac-tured and brecciated zone at roughly 720 m [2,360 ft] bsf that contained faults. Scientists

chose the zone at 720 mbsf and the décollement at 820 mbsf as the primary targets for the tem-perature measurement observatory.

Logging data, particularly resistivity images, helped scientists understand the local stress regime. Image logs acquired while drilling revealed borehole breakouts that reflected differ-ences in several in situ stress domains along the borehole. At shallow depths, the maximum hori-zontal compressive stress, σHmax, appeared vari-able. At deeper levels within the prism, 537 to 820 m [1,762 to 2,690 ft] bsf, σHmax displayed a single preferred orientation approximately 20° clockwise from the plate convergence direction. Faults and bedding throughout the upper sedi-mentary prism also showed a predominant north-east strike direction consistent with horizontal contraction approximately parallel to the plate convergence direction.

In July 2012, roughly 16 months after the Tohoku-Oki earthquake, the CDEX staff and additional scientists conducted Expedition 343T, a short technical extension of JFAST. During this expedition, engineers installed an observatory with temperature and pressure sensors to detect the residual temperature signature of fault displacement prior to its dissipation. During installation of the temperature observatory, the Chikyu crew drilled to 854.8 m [2,804 ft] bsf in water depth of 6,897.5 m [22,630 ft] for a total depth (TD) of 7,752.3 m [25,434 ft] bsl.35 Each of the 55 thermometers in the observatory has its own data logger. This was the first attempt to measure residual temperatures at a plate boundary fault soon after an ocean trench earthquake. In April 2013, the JAMSTEC research vessel Kairei returned to the site of the

> Subseafloor residual temperature field. The residual temperature is the difference between the recorded temperature and the temperature predicted from the background geothermal gradient. A time-depth map (left ) of data from the thermistor array (yellow circles) shows elevated temperatures in the depth interval 810 m [2,657 ft] to 820 m [2,690 ft] near the estimated depth of the décollement fault zone. Periods of no data collection are shown by white. Drilling disturbances caused temperatures to be cooler in August 2012. Temperatures increased at depth 760 m [2,493 ft] following the Mw 7.4 earthquake in December 2012. A time-lapse depth profile of residual temperatures between August 2012 and April 2013 (right ) shows the temperature progression at five discrete times. Scientists modeled the temperature anomaly to estimate the frictional heat from the 2011 Tohoku-Oki earthquake. (Adapted from Fulton et al, reference 38.)

Residual temperature, °C0.5–0.5 0

660

680

700

720

740

760

780

800

820

740

760

750

780

770

790

800

810

820

Dept

h, m

bsf

Dept

h, m

bsf

Date

MW 7.4 local earthquake

Aug 2012 Oct 2012 Dec 2012 Feb 2013 Apr 2013

Oct 1, 2012

Aug 1, 2012

Apr 1, 2013

Feb 1, 2013

Dec 1, 2012

Tem

pera

ture

, °C

0.50

0.25

–0.25

–0.50

0

Summer 2014 31

observatory, and scientists retrieved the sensor array after nine months of operation. The complete sensor string was recovered using a remotely operated vehicle, which located, detached and retrieved the array from a borehole that straddles the plate boundary.

Scientists directed an intensive effort toward analysis of LWD and core sample data from the JFAST expeditions. They concluded that a single major plate boundary fault had accommodated the large slip during the Tohoku-Oki earthquake rupture; they also determined that this was the site of most of the historical interplate motion that had occurred. Deformation was observed to be localized within a pelagic clay layer less than 5 m thick. This suggested that the frictional prop-erties of pelagic clay in this area serve as an important factor controlling the dynamic behav-ior of regional subduction earthquakes and of tsunamigenesis.36

Researchers conducted high-velocity friction experiments as well as geochemical, mineralogic and microstructural analyses on rock samples

retrieved from the plate boundary décollement and the neighboring sediments. They observed low stress drops associated with low peak and steady-state shear stresses and attributed these low stresses to the abundance of weak clay and thermal pressurization effects.37 Using data from the temperature sensor array, scientists observed a 0.31°C [0.56°F] temperature anomaly at the plate boundary fault (previous page). Combining these data with measured rock properties, they were able to estimate both the amount of energy dissipated during the earthquake and an appar-ent friction coefficient of 0.08. This very low coef-ficient of dynamic friction is thought to have facilitated the large slip that occurred on the Tohoku-Oki fault.38 These results of the JFAST project have helped the scientific community better understand the generation of earthquakes and tsunamis.

Challenges and the FutureSignificant advances over the last several decades in drilling, coring and logging technology have allowed IODP drillers to reach the depths of the seismogenic zone. However, additional technical challenges must be overcome before this achieve-ment can be extended to even deeper frontiers. IODP expeditions will soon explore candidate drilling sites in thick, hard rock formations, with water depths up to 4,500 m [14,800 ft] and bot-tomhole temperatures up to 250°C [480°F]. New spudding techniques or seafloor drilling systems may be necessary to initiate wells in bare rock crustal environments. Lightweight carbon fiber riser systems could extend the depth of riser drilling. An ultradeepwater electrohydraulic BOP may also be required. Drillers are considering the use of expandable casing to protect the wall sur-face of ultradeep wells.

Because of the increased time and cost asso-ciated with riser drilling to great depth, project teams aboard the Chikyu are gradually shifting away from scientific drilling’s tradition of full cor-ing to now make maximum use of LWD, MWD and mudlogging, with spot coring and sampling. To ensure high quality and quantity of recovered cores, ultrahard, polycrystalline diamond com-pact bits may be required for coring in ultrahard formations. Core barrels need to be developed for use at 300°C [572°F] and at 7,000 mbsf.39

Temperature estimates for existing seismo-genic zone projects do not exceed 150°C [300°F], but future projects will be more demanding. High-temperature MWD and LWD tools currently have 175°C [350°F] temperature limitations, which may need to be extended. Currently, stan-dard wireline logging tools are rated to 175°C,

and some to only 150°C. High-pressure, high-tem-perature (HPHT) logging tools rated to 260°C [500°F] and 207 MPa [30,000 psi] are available on limited bases, but not all measurements are available in these special tools. Some logging cables used to convey the tools are rated for oper-ation only up to approximately 200°C to 250°C [390°F to 480°F], although, like HPHT tools, spe-cial HPHT, high-strength cables are also in lim-ited supply.40 A continuing close partnership between the scientific community and service providers will be necessary to develop tools and processes up to the challenge.

Over the next decade, the IODP science plan envisions continued long-term focus on seismo-genic zones, earthquakes and natural disasters. Researchers will also embark on expeditions to answer questions related to global climate change and the occurrence and distribution of mineral and hydrocarbon resources, including gas hydrates. They will explore the deep bio-sphere and study the nature and extent of micro-bial populations. Potential benefits may include discovery of microorganisms useful for medical or other purposes such as carbon dioxide [CO2] sequestration or its conversion to methane [CH4].41 The greatest challenge, however, may come in fulfilling the long-held ambition of ultradeep drilling to the mantle, the unfulfilled goal of Project Mohole in the 1960s. The Mohorovicic discontinuity, or Moho, is the funda-mental seismic boundary that marks the transi-tion between the Earth’s crust and its upper mantle, but there is little knowledge of its geo-logic nature. Progress toward a Project Mohole for the 21st century is now well underway.42 —HDL

35. The total depth of Hole C0019D set a new record length for scientific drilling. It exceeded the total length of 7,049.5 m [23,128 ft] (a water depth of 7,034 m [23,077 ft] plus a subseafloor depth of 15.5 m [50.9 ft]) recorded by the Glomar Challenger at the Challenger Deep, the deepest known point in the Earth’s seabed, in the Mariana Trench in 1978.

36. Chester FM, Rowe C, Ujiie K, Kirkpatrick J, Regalla C, Remitti F, Moore JC, Toy V, Wolfson-Schwehr M, Bose S, Kameda J, Mori JJ, Brodsky EE, Eguchi N, Toczko S and the Expedition 343 and 343T Scientists: “Structure and Composition of the Plate-Boundary Slip Zone for the 2011 Tohoku-Oki Earthquake,” Science 342, no. 6163 (December 6, 2013): 1208–1211.

37. Ujiie K, Tanaka H, Saito T, Tsutsumi A, Mori JJ, Kameda J, Brodsky EE, Chester FM, Eguchi N, Toczko S and the Expedition 343 and 343T Scientists: “Low Coseismic Shear Stress on the Tohoku-Oki Megathrust Determined from Laboratory Experiments,” Science 342, no. 6163 (December 6, 2013): 1211–1214.

38. Fulton PM, Brodsky EE, Kano Y, Mori J, Chester F, Ishikawa T, Harris RN, Lin W, Eguchi N, Toczko S and the Expedition 343, 343T and KR12-08 Scientists: “Low Coseismic Friction on the Tohoku-Oki Fault Determined from Temperature Measurements,” Science 342, no. 6163 (December 6, 2013): 1214–1217.

39. For more on ultradeepwater technology development: “Ultra-Deep Drilling Technology,” JAMSTEC: Center for Deep Earth Exploration, http://www.jamstec.go.jp/chikyu/eng/developtech/deepdrill/ (accessed April 20, 2014).

40. For more on HPHT logging technologies: DeBruijn G, Skeates C, Greenaway R, Harrison D, Parris M, James S, Mueller F, Ray S, Riding M, Temple L and Wutherich K: “High-Pressure, High-Temperature Technologies,” Oilfield Review 20, no. 3 (Autumn 2008): 46–60.

For more on HPHT sampling and pressure: Avant C, Daungkaew S, Behera BK, Danpanich S, Laprabang W, De Santo I, Heath G, Osman K, Khan ZA, Russell J, Sims P, Slapal M and Tevis C: “Testing the Limits in Extreme Well Conditions,” Oilfield Review 24, no. 3 (Autumn 2012): 4–19.

41. For more on the application of drilling science to exploring the Earth’s interior: JAMSTEC: Research and Development Center for Ocean Drilling Science, http://www.jamstec.go.jp/ods/e/ (accessed July 16, 2014).

42. For more on the Mohole project: Pilisi N and Whitney B: “Initial Feasibility Study to Drill and Core the Ocean Mantle,” Scientific Drilling 12 (September 2011): 46–48.

32 Oilfield Review

Land Seismic Surveys for Challenging Reservoirs

Seismic surveys are instrumental in locating sweet spots in challenging reservoirs

and making them viable economic targets for completion and production.

Point-receiver technology is facilitating identification of sweet spots while offering

a cost-effective method of acquiring 3D land seismic surveys over large areas.

Gabriele BusanelloAbu Dhabi, UAE

Zhifeng ChenXue LeiRong LiBeijing, People’s Republic of China

Mark EganHouston, Texas, USA

Thomas HeesomDubai, UAE

Bo LiangChina National Petroleum Corporation Sichuan Geophysical CompanyChengdu, People’s Republic of China

Heloise Bloxsom LynnLynn IncorporatedLa Veta, Colorado, USA

Anastasia PooleGatwick, England

Peter van BaarenSneek, The Netherlands

Fusen XiaoPetroChina Southwest Oil and Gas CompanyChengdu, People’s Republic of China

Oilfield Review Summer 2014: 26, no. 2. Copyright © 2014 Schlumberger.For help in preparation of this article, thanks to Phillip Bilsby, John Kingston and Dominic Lowden, Gatwick, England; George El-Kaseeh and Scott Totten, Houston; David F. Halliday, Cambridge, England; Harvey F. Hill, Frisco, Texas, USA; Qinglin Liu, Beijing; Denis Sweeney, Perth, Australia; and Brian Toelle, Denver.SWAMI and UniQ are marks of Schlumberger.StarTrak is a mark of Baker Hughes Inc.

The best tool for identifying potential exploration and development targets before drilling is the 3D seismic survey. These surveys allow operators to image huge volumes of the subsurface and iden-tify potential hydrocarbon reservoirs. Advances in seismic acquisition and processing technology now allow geophysicists to peer within prospects in search of sweet spots. One key advance is point-receiver acquisition.1

Unconventional reservoirs, particularly shale formations, have become increasingly important to the industry and today are the focus of major exploration and production efforts. In these tight formations, optimal production comes from sweet spots, which have a unique combination of reservoir and geomechanical rock properties.

Sweet spots are characterized by excellent reservoir quality (RQ) and completion quality (CQ) and, when stimulated effectively by hydrau-lic fracturing, produce economic quantities of hydrocarbons. Consequently, the objective of seismic surveys has become more well defined. Survey results are used to identify drilling tar-gets, optimize drilling trajectories and locate placements of hydraulic fracture stages and com-pletions.2 To accomplish these objectives, seismic processors and interpreters require high-quality data to characterize RQ, CQ and vertical, hori-zontal and azimuthal variations within individual reservoir layers and compartments.3

This article describes state-of-the-art land seis-mic point-receiver technology that enables acqui-sition of full azimuth (FAZ) and long offset surveys with small bin sizes.4 These surveys “shine the light” on reservoir targets from many directions, and relative to conventional surveys, provide bet-ter illumination, higher signal-to-noise ratio (SNR) and improved seismic resolution.

A point receiver records an individual trace of raw data. In contrast, a conventional seismic trace is one that results from summing traces from an array, or a group, of receivers.5 The sum-ming improves the SNR of recorded data by attenuating ambient and coherent noise and accentuating signal. However, conventional prac-tice yields recorded data one step removed from the raw data, which reduces flexibility during later data processing. Point-receiver traces are the raw data, which provide the greatest process-ing flexibility.

Dense surveys with many recording channels are required to obtain full azimuth, long offset and high-fold data. The WesternGeco UniQ land seismic system offers this capability and is able to record more than 200,000 live channels simul-taneously. Datasets acquired with UniQ systems facilitate processing to characterize seismic anisotropy, preserve bandwidth, improve SNR and increase the temporal and spatial resolution of seismic images and attributes.6 The resulting data enable a precise definition of rock proper-ties for determining RQ and CQ; when these are determined, the chance of drilling a productive well is improved. Examples from China and the US illustrate how these data help resolve explora-tion, completions and production questions before drilling.

Wavelet PreservationA seismic trace is a recorded response of a wavelet, originating from an energy source, to the subsurface geology. It is the convolution, or combination, of the wavelet with the subsur-face’s reflectivity—the series of reflections the seismic wavelet makes as it encounters geologic interfaces. Across each interface, the physical

Summer 2014 3333

properties of the rock units change—in particu-lar, the compressional (P-wave) and shear (S-wave) velocities, density and P-wave and S-wave impedances. The impedance contrast across each interface, along with the wave’s angle of incidence, determines the amount of energy reflected and transmitted.

Geophysicists use impedance contrasts to ascertain properties of subsurface rocks by performing amplitude inversion. This process removes the wavelet from the seismic trace and inverts for the impedances between seis-mic reflectors. The procedure is challenging

because the recorded trace also includes coherent, scattered and environmental noise that must be removed. Wavelet extraction is even more difficult when imaging unconven-tional reservoirs because the earth response typically manifests itself as small and subtle amplitude and phase variations, which can be masked easily by noise or obliterated during seismic acquisition and processing.

To ensure extraction of the correct wavelet and obtain accurate inversion, all stages of seis-mic acquisition must be performed with great care. Geophone arrays, typically used for conven-

tional acquisition to improve SNR, are known to distort the amplitude and phase of the apparent wavelet. Using arrays is a major cause for inac-curate wavelet extraction because the recording response of the array varies with the azimuth, geophone spacing, ground surface elevation dif-ferences and individual geophone tilt within the array.

Closely spaced point sensors—in which each sensor records its own trace—eliminate these effects. Closely spaced single traces remove the need for arrays and facilitate noise removal and preservation of wavelet fidelity—the accuracy of

1. Ait-Messaoud M, Boulegroun M-Z, Gribi A, Kasmi R, Touami M, Anderson B, van Baaren P, El-Emam A, Rached G, Laake A, Pickering S, Moldoveanu N and Özbek A: “New Dimensions in Land Seismic Technology,” Oilfield Review 17, no. 3 (Autumn 2005): 42–53.

Bagaini C, Bunting T, El-Emam A, Laake A and Strobbia C: “Land Seismic Techniques for High-Quality Data,” Oilfield Review 22, no. 2 (Summer 2010): 28–39.

Barclay F, Bruun A, Rasmussen KB, Camara Alfaro J, Cooke A, Cooke D, Salter D, Godfrey R, Lowden D, McHugo S, Özdemir H, Pickering S, Gonzalez Pineda F, Herwanger J, Volterrani S, Murineddu A, Rasmussen A and Roberts R: “Seismic Inversion: Reading Between the Lines,” Oilfield Review 20, no. 1 (Spring 2008): 42–63.

2. Ajayi 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: “Stimulation Design for Unconventional Resources,” Oilfield Review 25, no. 2 (Summer 2013): 34–46.

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

3. Barclay et al, reference 1.4. A point sensor is a seismic source or receiver that has a

small footprint on the earth surface. Full azimuth is complete 0° to 360° azimuthal coverage. Long offset refers to large separation distances, corresponding to wide angles of incidence, between sources and receivers. A bin is a subdivision of a seismic survey; seismic traces are assigned to specific bins based on whether they are sorted by common midpoint, common reflection point, common offset or other criteria. Fold is the number of traces assigned to a bin.

5. An array is a group of seismic receivers from which all signals are combined and recorded on one data channel of a recording device.

6. Resolution is the minimum separation—two-way traveltime or distance—between two features for them to be distinguishable in a seismic section or volume. Temporal resolution is the minimum time separation that allows features to be distinguished in a time-based plot. Spatial resolution is the distance required to recognize differences between features separated horizontally and vertically.

For more on resolution: Egan MS: “The Drive for Better Bandwidth and Resolution,” in Doré AG and Vining BA (eds): Petroleum Geology: North-West Europe and Global Perspectives—Proceedings of the 6th Petroleum Geology Conference. London: The Geological Society, Petroleum Geology Conference Series 6 (2005): 1415–1424.

Egan MS, Seissiger J, Salama A and El-Kaseeh G: “The Influence of Spatial Sampling on Resolution,” CSEG RECORDER 35, no. 3 (March 2010): 29–36.

34 Oilfield Review

its amplitude and phase—during processing and seismic inversion.7

After the seismic wavelet is removed and the earth response is recovered, the result is a dataset of relative seismic impedances that geophysicists translate into physical properties representing reservoir lithologies. Geophysicists combine the relative impedances from inversion with an imped-ance model derived from a single well location, or if available, multiple well locations. They next con-struct absolute impedances by adding the low-fre-quency information from 0 Hz to the lowest frequencies contained in the seismic data. The absolute impedances may be converted to litho-logic physical properties such as density, porosity, Poisson’s ratio and P-wave and S-wave velocities. These properties are critical for classifying uncon-ventional reservoir lithologies, determining their RQ and CQ and identifying sweet spots.

Point-Receiver AcquisitionRecording a 3D seismic survey is a compromise between the ideal survey design and the acquisi-tion cost and time. The ideal design is a dense geometry that illuminates the reservoir from many directions. However, to acquire a dense sur-vey entails a lot of equipment and many people, leading to high cost and long duration. WesternGeco geophysicists and engineers have spent more than a decade perfecting the UniQ land seismic system to reduce survey costs by deploying less equipment—single sensors rather than arrays—and reducing acquisition time by acquiring data more quickly.8 In addition, the UniQ land system has built-in redundancies that mitigate recording equipment and communica-tion line failures.

During conventional seismic acquisition, the traces from an array of receivers are aggregated, summed together and then recorded. The intent

of array summation is to enhance vertically trav-eling seismic signals coming from the subsurface and to attenuate horizontally traveling ambient and coherent noise. However, sensors in an array are not clumped closely together but spread over a measurable area. Variations of topography and geology between receivers within an array affect the arrival time and amplitude of signals within individual traces. Consequently, summation of uncorrected, misaligned traces results in recorded traces that are smoother and contain less detail than the original traces and are less precisely correlated with neighboring recorded traces; summation of misaligned traces blurs the high-frequency information within traces. The practice of recording after trace summation removes any possibility of reconstructing the raw data to recover lost information.

Advances in computer storage, data processing and receiver technology enable geophysicists to use

> Geophone accelerometer (GAC) receiver unit. A seismic survey crew member positions a GAC (top left ). A cutaway of a GAC sensor unit (bottom left ) shows the acquisition electronics and geophone element. The frequency response (top right ) of a typical GAC is flat, or uniform, between 1.4 Hz and 250 Hz (red dotted lines) in which the amplitude response is down by 3 dB to 70% of the peak amplitude. A string of GAC sensors (bottom right ) is ready for the survey crew to deploy in the field.

Ampl

itude

resp

onse

, dB

Frequency, Hz0.1 1.0 10.0 100.0

–24

–21

–18

–15

–12

–9

–6

–3

0

Geophoneelement

Acquisitionelectronics

Summer 2014 35

point-receiver acquisition. In 2002, WesternGeco introduced its first-generation single-sensor seis-mic acquisition and processing system capable of recording 30,000 live data channels.9

WesternGeco introduced the second-genera-tion UniQ integrated point-receiver system in 2009.10 It is currently capable of recording more than 200,000 live channels, and its architecture has been designed to be scaled up even more. Today, hardware improvements such as graphical processing units and increases in processing speed and algorithm efficiency have eliminated the requirement to reduce the data volume after initial data processing—data preparation and noise attenuation—through digital group form-ing (DGF).11 Now, all traces can pass through the full data processing sequence.

Each point receiver is a geophone acceler-ometer (GAC), which is a high-sensitivity, low-distortion broadband sensor.12 The GAC records data containing frequencies from 0 Hz (station-ary) to more than 500 Hz; GACs have a uniform, or flat, frequency response from 1.4 to 250 Hz, which ranges over seven octaves (previous page).13 The GAC is designed to increase

dynamic signal response to low-frequency sig-nals relative to conventional geophones. Accurate measurement of low-frequency signal is essential for inverting seismic amplitudes to infer geologic and rock property information.

The high channel count of the UniQ system allows the wavefields to be finely sampled with-out aliasing.14 Seismic analysts can process the raw data and make trace-by-trace corrections to take into account any local elevation and near-surface geology variations before they apply noise removal techniques to enhance the vertically propagating signal.

Noise RemovalRecently, noise attenuation methods have been developed to reduce coherent, scattered and ambient, or environmental, noise using point-receiver datasets; these new methods have largely replaced DGF. Unlike conventional meth-ods, this state-of-the-art approach is not depen-dent on regular acquisition geometries such as uniform grids to gather and record the data (above). Each of these new noise attenuation techniques focuses on a specific type of noise

7. A wavefield is the amplitude response of a wave spreading through the subsurface.

8. For more on UniQ land seismic techniques: Ait-Messaoud et al and Bagaini et al, reference 1.

9. Ait-Messaoud et al, reference 1.10. Papworth S: “Stepping up Land Seismic,” E&P 82, no. 3

(March 1, 2009), http://www.epmag.com/Exploration-Geology-Geophysics/Stepping-land-seismic_31469 (accessed May 29, 2014).

11. Digital group forming (DGF) is a series of processing steps to combine many raw point-sensor measurements into a smaller number of measurements formed by groups of traces. For more on DGF: Ait-Messaoud et al, reference 1.

12. Ground motion may be measured as displacement, velocity or acceleration. A geophone is a seismic recording device that detects ground velocity, or change in ground displacement with time, produced by seismic waves. An accelerometer is a recording device that measures acceleration, or change in velocity with time. When used as a seismic recording device, an accelerometer detects ground acceleration. An accelerometer records a broader frequency range than that of a geophone.

13. An octave is a doubling of frequency content. A one-octave range is 2 to 4 Hz, 4 to 8 Hz, 8 to 16 Hz and so on. A two-octave range is 2 to 8 Hz, 4 to 16 Hz, 8 to 32 Hz and so on. A seven-octave range is 2 to 256 Hz.

14. Aliasing occurs when signals in waveforms cannot be distinguished because the waveforms are not sampled at small enough intervals—at high enough sample frequency. To prevent aliasing, the sample frequency should be greater than twice the highest frequency in the waveforms.

> Layout crew. A combined layout and survey crew ensures accuracy of GAC location coordinates and reduces head count and operational time. The combined operations ensure accurate sensor locations for nonuniform data acquisition and minimize disturbance to the environment because no prelayout survey, flags or stakes are required. The GPS operator navigates to the planned sensor location using a satellite-based augmentation system (SBAS) enhanced global positioning system (GPS). The digger digs a hole at the marked optimal sensor location, which is away from shrubbery that could generate noise from its movement by wind and away from hard ground that offers poor sensor coupling with the earth. The sensor unit string (SUS) handler carries the string of GAC sensors and drops a sensor in the hole, and the person performing GAC placement ensures the sensor is as close to vertical as possible and well planted for optimal coupling. One of the last two crew members (right ) carries a mobile positioning and testing system (MPTS) unit that injects the SBAS-corrected GPS coordinates directly into the sensor memory—a three-second operation—for inclusion into the seismic data headers during acquisition. The other person covers the sensor with soil and ensures that all wires are flat on the ground to minimize noise from the wind. Depending on the terrain, a combined crew can cover several kilometers a day.

GPS operator

DiggerSUS handler GAC placement

GPS coordinate injectionand GAC burial

Layout and Survey Crew GPS Coordinate Injection

36 Oilfield Review

present in land data and is applied methodically and progressively to peel away the various com-ponents of the complex surface wave energy con-taminating the data. The typical noise attenuation route, in order of application, is the following: nonuniform coherent noise suppression (NUCNS) to remove coherent ground roll, or surface wave, noise; model-driven interferometry (MDI) to remove scattered ground roll noise; and nonuni-form environmental noise suppression (NUENS) to attenuate environmental noise (left).15

Based on the assumption of spatially irregular input data, the NUCNS method removes coherent noise through subtraction of a coherent noise model from the original data. When the direct ground roll arrivals are aliased, geophysicists may use the SWAMI surface wave analysis, model-ing and inversion method for coherent noise sup-pression (CNS). The SWAMI method allows the requirement for dense point-receiver spacing to be relaxed.16

The NUCNS method works well for direct ground roll noise. However, when attenuating scattered ground roll, the method is limited to removing only the flanks of scattered ground roll noise and, similar to other velocity discrimina-tion techniques, is unable to attenuate the apexes.17 For this reason, the MDI method was developed to suppress scattered ground roll noise and can estimate the difficult-to-remove apexes of the scattered ground roll.18

After coherent and scattered ground roll have been subtracted from the data, noncoherent ambi-ent noise remains. The NUENS method is a local, environmental noise modeling technique based on a diversity principle and frequency separation of signal and noise. This method effectively removes ambient noise that remains in the data after NUCNS, or SWAMI CNS, and MDI processing.

Unconventional Reservoir DrillingEarly development of unconventional—tight oil and shale gas—reservoirs in North America was characterized by geometric drilling and comple-tions. Operators drilled horizontal wells in regu-lar patterns and stimulated the wells using equally spaced hydraulic fracture stages along the laterals.

This development strategy ignores the het-erogeneity that is characteristic of unconven-tional reservoirs. Retrospective studies of reservoirs that were developed in this manner showed that productivity varied significantly across each field, among neighboring wells and between fracture stages within the same well.19 The studies demonstrated that reservoir hetero-

>Noise removal workflow. A typical workflow for attenuating noise in seismic data includes removing coherent noise, scattered ground roll noise and ambient noise. The nonuniform coherent noise suppression (NUCNS) and SWAMI surface wave analysis, modeling and inversion methods (left ) remove coherent, direct ground roll noise. The NUCNS method works well when the direct ground roll can be identified easily and is not aliased. If it is aliased, then the SWAMI procedure is the better choice. The model-driven interferometry (MDI) method (middle) removes scattered ground roll noise. The nonuniform environmental noise suppression (NUENS) method (right ) removes any remaining ambient noise.

Removecoherent noise

Removeambient noise

Remove scatteredground roll noise

Nonuniform coherentnoise suppression

(NUCNS)

SWAMI surface waveanalysis, modeling

and inversion

Nonuniform environmentalnoise suppression

(NUENS)

Model-driveninterferometry

(MDI)

> Imaging prospects better. Seismic sections of conventional and UniQ acquisition from the same location show different features after anisotropic prestack depth migration (PSDM). Conventional acquisition (top left) shows little evidence of the planned well intersecting a fault. The UniQ 3D section (top right) shows the fault more clearly. A common image point (CIP) gather of traces from the UniQ dataset focuses on a target reflector. If the traces are sorted strictly by increasing offset (middle), the reservoir horizon (brightest blue and red) appears incoherent. If the traces are sorted into offset bands and then by azimuth (bottom), the target reservoir reflector horizon appears ordered and sinusoidal. The sinusoidal and color variations along the reflector result from seismic anisotropy. Vertical dotted lines and colors underneath the image delimit the offset bands. Within each band, the traces are arranged from left to right according to azimuth from 0° to 180° (see “In Situ Stress, Natural Fractures and Seismic Anisotropy,” page 44).

Trav

eltim

eTr

avel

time

Trav

eltim

e

Conventional Prestack DepthMigrated Section

Traces Ordered Strictly by Offset

Traces Ordered by Offset then by Azimuth

UniQ Prestack DepthMigrated Section

Reflection amplitude

– 0 +

Fault

Summer 2014 37

15. For more on noise attenuation applied to UniQ data: Xiao F, Yang J, Liang B, Zhang M, Li R, Li F, Xiao H, Lei X, Liu Q and Heesom T: “High-Density 3D Point Receiver Seismic Acquisition and Processing—A Case Study from the Sichuan Basin, China,” First Break 32, no. 1 (January 2014): 81–90.

16. Strobbia C, Zarkhidze A, May R, Quigley J and Bilsby P: “Attenuation of Aliased Coherent Noise: Model-Based Attenuation for Complex Dispersive Waves,” First Break 29, no. 8 (August 2011): 93–100.

For more on surface waves and the SWAMI method: Strobbia C, Vermeer PL, Laake A, Glushchenko A and

geneity, natural fracture systems and in situ stress were key determinants of geologically favorable reservoir compartments for which RQ and CQ were high.20

The 3D surface seismic survey is the best tool for identifying potential exploration and development targets before drilling. These sur-veys allow operators to image huge volumes of the subsurface and to locate potential reser-voirs. Advances in seismic acquisition and pro-cessing technology are allowing geophysicists to peer within these prospects with increased con-fidence (previous page, bottom). Although oper-ators use this technology in unconventional resource plays, it can also benefit operators in developing conventional reservoirs.21

Illuminating DetailsThe PetroChina Southwest Oil and Gas Company (SWOGC) is exploring for tight oil accumula-tions in the Gongshanmiao oil field, located in the north-central Sichuan basin in central China (right). From oldest to youngest, the tight oil reservoirs reside in fractured limestone of the lower Jurassic Daanzhai (Da) formation deposited in lacustrine environments, interbed-ded sandstone and shale of the lower Jurassic Lianggaoshan (Liang) formation deposited in fluvial and lacustrine environments and chan-nel and sheet sandstone of the middle Jurassic Shaximiao (Sha) formation deposited in fluvial environments.22 Collectively, these reservoir zones have porosities ranging from 1% to 5% and ultralow permeabilities ranging from 0.0001 to 0.5 mD.23 The gross thickness of the units ranges from 10 to 230 m [33 to 750 ft]. Individual deposits may be thinner and could not be ade-quately imaged and delineated using legacy 3D seismic data from 1999 and 2003.

As part of its exploration effort, SWOGC com-missioned the China National Petroleum Corporation Sichuan Geophysical Company in early 2013 to acquire a high-density 3D seismic survey over the Gongshanmiao oil field using the UniQ integrated point-receiver land seismic sys-tem.24 The SWOGC exploration team wanted

> Gongshanmiao oil field. An exploration team from PetroChina Southwest Oil and Gas Company conducted a 3D UniQ integrated point-receiver land seismic survey over the Gongshanmiao field. The tight oil reservoirs in this field were not adequately imaged by earlier 3D surveys.

Sichuan basin

Gongshanmiaooil field

CHINA

Shanghai

Beijing

Xi’an

Chongqing

Basin Gas field Oil fieldSouth China Sea

Re S: “Surface Waves: Processing, Inversion, and Attenuation,” First Break 28, no. 8 (August 2010): 85–91.

Bagaini et al, reference 1.17. The apex of scattered ground roll is the apparent origin

point. Scattered energy spreads away, or fans out, from this point as traveltime increases. This spreading energy is the flank of scattered ground roll.

18. For more on model-driven interferometry: Halliday D: “Adaptive Interferometry for Ground-Roll Suppression,” The Leading Edge 30, no. 5 (May 2011): 532–537.

Halliday DF, Curtis A, Vermeer P, Strobbia C, Glushchenko A, van Manen D-J and Robertsson JOA: “Interferometric Ground-Roll Removal: Attenuation of Scattered Surface Waves in Single-Sensor Data,” Geophysics 75, no. 2 (March–April 2010), SA15–SA25.

Bilsby P, Halliday DF and West LR: “Case Study—Residual Scattered Noise Attenuation for 3D Land Seismic Data,” paper I040, presented at the 74th European Association of Geoscientists and Engineers Conference and Exhibition, Copenhagen, Denmark, June 4–7, 2012.

19. 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, USA, November 2–3, 2010.

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, 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.

20. A fracture system is a set of fractures that formed at the same time under the same in situ stress regime. A fracture trend is a linear set of multiple open fractures that are more or less aligned. In the literature, a fracture trend is also referred to as a fracture corridor.

For more on fractured reservoirs: Bratton T, Canh DV, Que NV, Duc NV, Gillespie P, Hunt D, Li B, Marcinew R, Ray S, Montaron B, Nelson R, Schoderbek D and Sonneland L: “The Nature of Naturally Fractured Reservoirs,” Oilfield Review 18, no. 2 (Summer 2006): 4–23.

Natvig JR, Skaflestad B, Bratvedt F, Bratvedt K, Lie K-A, Laptev V and Khataniar SK: “Multiscale Mimetic Solvers for Efficient Streamline Simulation of Fractured Reservoirs,” SPE Journal 16, no. 4 (December 2011): 880–888.

21. van Baaran P, Baioumy M, Cunnell C, Mohamed G, Zarkhidze A, Zubay E and Al Quadi A: “Integrated High-Density Point-Source, Point-Receiver Land Seismic,” E&P 86, no. 6 (June 2013), http://www.epmag.com/item/Integrated-high-density-point-source-point-receiver-land-seismic_116753 (accessed May 29, 2014).

22. Zou C, Shizhen T, Fan Y and Xiaohui G: “Characteristics of Hydrocarbon Accumulation and Distribution of Tight Oil in China: An Example of Jurassic Tight Oil in Sichuan Basin,” Search and Discovery Article 10386 (2012), adapted from an extended abstract prepared in conjunction with an oral presentation at the AAPG International Conference and Exhibition, Milan, Italy, October 23–26, 2011.

Liang B, Zhang M, Xiao F, Yang J, Lei X, Liang D, Li R, Li F, Liu Q, Qian H and Zhan L: “Characterizing Tight Thin Reservoirs Using High-Density 3D Seismic—A Case Study from the Central Sichuan Basin,” First Break 32, no. 5 (May 2014): 85–93.

23. Zou et al, reference 22.24. For more on the Gongshanmiao oil field: Xiao et al,

reference 15. For more on the UniQ land seismic system:

Papworth, reference 10.

38 Oilfield Review

improved definition of the Jurassic reservoir zones to support decisions for developing the resources and placement of horizontal wells. The survey imaging objectives were the thin Sha and Liang reservoir sandstones and the thin Da frac-tured limestone units beneath them. To achieve both objectives, geophysicists planned a full azi-muth, long offset, high-fold and symmetrically sampled survey design (below).

At any given time during the survey, there were more than 40,000 channels deployed to ensure efficient recording operations. However, each shot gather consisted of 25,920 live chan-nels of GAC receivers deployed in a characteris-tic square acquisition pattern arranged along 36 parallel receiver lines separated by 200 m [660 ft] and containing 720 receivers spaced 10 m apart.

The seismic sources, 1- to 1.5-kg [2.2- to 3.3-lbm] dynamite charges, were placed in rela-tively shallow 6-m [20-ft] drill holes. The source pattern, identical to the receiver pattern, was oriented crossline, or perpendicular, to it. The

maximum inline and crossline offsets were 3,600 m [11,800 ft]. The trace density that resulted from the survey was about 13 million traces/km2 [34 million traces/mi2].25

Because the dynamite charge size and shot hole depth were small relative to those customar-ily used in the area of operations, holes could be drilled in less time, which led to early completion of the acquisition program. The operator was assured that excessive noise generated by the shal-low charges could be removed during data process-ing because of the high trace density and the options for noise processing afforded by recording a symmetrically sampled point-receiver dataset.

Geophysicists began the data processing by merging the geographic locations and elevations of sources and receivers with the recorded data. They then picked first-break arrivals and calcu-lated first-arrival statics.26 In this process, unwanted noise must be removed without affect-ing the geologically significant reflection signals used to image the reservoir zones. To account for such noise and to safeguard against aggressive

noise removal, geophysicists applied surface-consistent amplitude compensation to the data before noise removal.27 Noise attenuation was accomplished using the NUCNS, MDI and NUENS processing techniques (next page).28

The geoscience team wanted to use the ampli-tude variation with offset and azimuth (AVOAZ) method for reservoir imaging and for analysis of seismic anisotropy and hydrocarbon potential.29 This method requires retaining full azimuthal information and taking complete advantage of the long offsets and azimuth coverage in the dataset. Seismic processors sorted and binned the data into offset vector tile (OVT) gathers.30 The OVT sorting preserves the offset and azimuth information of the prestack data for anisotropy, fracture and AVOAZ analyses.

Seismic reflection AVOAZ data processing, inversion and analysis assume that reflection amplitude is proportional to the seismic, or elas-tic, impedance contrast across the interfaces between the reservoir zone and the intervals above and below it. This contrast is quantified by

> Point-receiver land seismic survey design. The UniQ receiver acquisition template (left ) consisted of 25,920 live channels arranged in a 7,200-m × 7,200-m [23,600-ft × 23,600-ft] square pattern formed by 36 densely sampled receiver lines (blue). For comparison, a receiver template from a legacy survey is represented by the area covered by black lines. The receiver template is centered on a shot location, shown in red. The green dashed circles show angles of incidence for reflections at the target depth and progress outward from the center in increments of 11°; horizontal distances correspond to offsets of receivers from the sources in the center. The source template was identical to the receiver template but was oriented crossline to the receivers. The offset-azimuth plot (right ) indicates the offsets and azimuths acquired by the survey. Offset corresponds to the distance from the center of the circle. Azimuth corresponds to the angle clockwise from the reference direction at the top of the circle. Colors range from purple for a low number of traces to green, yellow and red for a high number of traces. The survey design provided full azimuth and four-quadrant symmetrical coverage.

–4,000

–3,000

–2,000

–1,000

0

4,000

3,000

2,000

1,000

–4,000

Cros

slin

e of

fset

, m

–5,000

–6,000

–3,000

–4,000

–1,000

–2,000

1,000

0

3,000

2,000

5,000

6,000

4,000

6,0005,0004,0003,0002,0001,0000

0°

180°

150°

120°

90°270°

60°

240°

300°

30°

210°

330°

–1,000–2,000–3,000–4,000–5,000–6,000

Cros

slin

e of

fset

, m

Inline offset, m

W

N

S

E

0–3,000 –2,000 –1,000 4,0003,0002,0001,000

Inline offset, m

UniQ spreadLegacy spreadShot location

0° to 11°11° to 22°22° to 33°

Trace density per bin

Low High

Summer 2014 39

25. For more on the data acquisition in the Gongshanmiao oil field: Xiao et al, reference 15.

26. A static, or statics correction, is the time shift to the traveltimes within each seismic trace to account for traveltime perturbations caused by local near-surface conditions such as variations in elevation and in seismic velocity—from geology and weathering—at each source and receiver position. First-arrival statics and reflection residual statics are methods to correct for these effects.

27. Various surface-consistent processing methods remove source and receiver traveltime, amplitude and the wavelet variations caused by acquisition-related effects and not by geology. Specifically, surface-consistent amplitude compensation accounts for source and receiver amplitude variations.

28. Xiao et al, reference 15.

the seismic reflectivity, or reflection coefficient, at the lithologic contact. The reflectivity depends on formation bulk density, P-wave velocity and S-wave velocity; knowledge of the reflectivity con-trast allows geophysicists to deduce changes in fluid and rock type. Reflectivity varies with the angle of incidence, which is determined by the source-receiver offset; this variation is the basis of the amplitude variation with offset (AVO) method. In addition, in an anisotropic medium, reflectivity varies with the azimuth from which the seismic wave strikes the interface because seismic velocity varies with azimuth within the layer above and below the interface. This phe-

nomenon is the basis for AVOAZ analysis, which is used to infer the anisotropic characteristics of a reservoir. Such anisotropy may result from the predominant orientation of fractures, layering and in situ stresses and may express itself in the direction-dependent values of physical, geome-chanical and fluid-flow properties.

To obtain optimal reflection images that may be used for AVOAZ inversion, geophysicists must determine the seismic amplitude effects result-ing from geologic complexities above the target. For accuracy, this determination should charac-terize the lithologic, petrophysical and anisotro-pic nature of the overburden. Of the 30 wells in a

100-km2 [40-mi2] area that surrounded the Gongshanmiao oil field survey area, only 7 wells were inside the survey area. Well log data were available from 12 of the 30 wells; S-wave velocity information was available from 4 wells located outside the survey area.

Geoscientists constructed a calibrated geo-logic model constrained by available well data and then processed the OVT prestack migrated data using simultaneous seismic inversion to extract porosity, lithology, hydrocarbon satura-tion and geomechanical parameters. They then benchmarked the results against well logs.31 This process allowed definition of the geometry of

>Noise suppression. Data from a common source gather (left ) show vertical sections in the source and receiver line directions and a seismic horizon (horizontal plane) before noise processing. After NUCNS, MDI and NUENS processing (right ), coherent and environmental noise have been suppressed from the apex and flanks of the noise cone to reveal more detail of reflections, which can be traced with greater confidence under the source.

Receiver line Source line Receiver line Source line

Before Noise Suppression Processing After Noise Suppression Processing

Reflections

29. Anisotropy is characterized by having physical properties whose values depend on their measurement direction.

30. An offset vector tile (OVT) is a single-fold dataset that, after sorting and binning, contains a restricted range of offsets and azimuths, or offset vectors.

For more on OVTs: Vermeer GJO: 3D Seismic Survey Design, 2nd ed. Tulsa: Society of Exploration Geophysicists (2012): 15–58.

31. 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.

Rasmussen KB, Bruun A and Pedersen JM: “Simultaneous Seismic Inversion,” paper P165, presented at the 66th European Association of Geoscientists and Engineers Annual Conference and Exhibition, Paris, June 7–10, 2004.

Ma X-Q: “Simultaneous Inversion of Prestack Seismic Data for Rock Properties Using Simulated Annealing,” Geophysics 67, no. 6 (November–December 2002): 1877–1885.

Bachrach R, Beller M, Liu CC, Perdomo J, Shelander D, Dutta N and Benabentos M: “Combining Rock Physics Analysis, Full Waveform Prestack Inversion and High-Resolution Seismic Interpretation to Map Lithology Units in Deep Water: A Gulf of Mexico Case Study,” The Leading Edge 23, no. 4 (April 2004): 378–383.

Bachrach R: “Joint Estimation of Porosity and Saturation Using Stochastic Rock-Physics Modeling,” Geophysics 71, no. 5 (September–October 2006): O53–O63.

40 Oilfield Review

hydrocarbon-bearing channel sands of the upper Jurassic Sha reservoir interval and qualitative categorization of their RQ into hydrocarbon-bearing sand, wet sand and shale (above).

In addition, geoscientists combined AVOAZ inversion and poststack fault and fracture detec-tion via ant tracking to characterize the natural fracture systems within the middle Jurassic Da limestone. The results indicated that the domi-nant fracture trend generally strikes NW–SE (next page).32

To confirm the seismic interpretation, the exploration team performed a blind comparison of the seismic results with well data that had not been included in developing those results.33 For the case of the Sha channel sands, the scientists used three wells. The well data substantiated the seismic forecasts that good RQ coincided with high values of porosity, hydrocarbon content and shear velocity anisotropy.

For the case of the Da fractured limestone, the scientists used wellbore images from logging while drilling (LWD) in a horizontal section of a well drilled in the tight limestone of the lowest Da member, the Da3. The image interpretation corroborated the AVOAZ and ant tracking results; the dominant fracture trend coincided with areas of high S-wave velocity anisotropy and the frac-tures tend to strike in the NW–SE direction and dip in the NE direction at 60° to 90°.

These tests support the value of the informa-tion gained by conducting 3D seismic surveys using point sensors in high density configura-tions. Such surveys give geophysicists rich data-sets to work with and, when combined with geologic and petrophysical information, allow geoscientists to develop more precise models of reservoir zones that help guide development, drilling and completion strategies.

> Lithology classification. A seismic time horizon near the bottom of the Shaximiao (Sha) formation (left ) is color coded by lithology. The pattern of hydrocarbon (HC)-bearing sand (orange), wet sand (blue) and shale (black) reveals fluvial, braided and isolated channels within the Sha formation. Lithology classification is the result of defining the seismic, or elastic, characteristics of rocks and fluids based on calibrations of seismic data against well logs using rock physics and statistics methods (right ). This method classifies lithology of each point in the image based on probability density functions (PDFs) of the P-wave impedance and the P- to S-wave velocity ratio from simultaneous AVO inversion. The PDFs combine into an occurrence probability contour map. Probability increases from the outer contour (zero) toward the innermost contour.

P- to

S-w

ave

velo

city

ratio

2.0

1.5

2.5

P-wave impedance, 106 kg/m2/s8 12 16

HC-bearing sand

Wet sand

Shale

HC-bearing sand Wet sand Shale

km0 5

0 mi 3

N

Lithology Occurrence Probability Contours

Rejuvenating a Fractured Carbonate Oil FieldProduction from wells in an onshore field in Texas, USA, was on the decline. The reservoir formation is a thick, naturally fractured carbon-ate characterized by low matrix porosity and low matrix permeability; the carbonate is over-lain by an organic-rich shale unit. The best oil production is found in zones that have well-con-nected natural fracture networks. Seeking to increase production, the operator wanted to use results from seismic technology to design field development plans, drill horizontal wells, com-plete them using hydraulic fracture stimula-tions and connect the wells to the natural fractures. Because existing 3D P-wave seismic data in the area were not of sufficient quality for this purpose, the operator contracted with WesternGeco to acquire a 3D P-wave UniQ point-receiver survey over the field.

Summer 2014 41

The data were acquired in 2011 using vibro-seis sources and point-receiver GACs.34 For each vibration point (VP), 16,900 GAC receivers were active. The receiver pattern consisted of parallel lines at 1,000-ft [300-m] spacing, and each line contained 1,300 GACs separated by 20 ft [6 m]. The source pattern was crossline to the receiver pattern; it had lines spaced 1,200 ft [370 m] apart and contained VPs separated by 200 ft [60 m]. The crossing source and receiver patterns resulted in symmetric, full azimuth, long offset coverage in which the maximum inline and crossline offsets were 13,200 ft [4,020 m] and 13,000 ft [3,960 m], respectively. These long off-sets are important for characterizing amplitude

variation with offset (AVO) effects because they exceeded the 8,000- to 10,000-ft [2,400- and 3,000-m] target depth.35

Data processing included attenuating random and coherent noise, conducting velocity analysis and developing a first-order anisotropic layered velocity model using tomography and Kirchhoff prestack depth migration (PSDM). Next, the data were sorted into OVTs, processed using OVT PSDM and analyzed for the following:• azimuthal P-wave velocity anisotropy using fitted

elliptical anisotropy from traveltimes (FEATT)• natural fractures using amplitude variation

with azimuth (AVAZ)• orthorhombic fabric using AVOAZ.36

The initial anisotropic velocity model repre-sented horizontal layers of the sedimentary rocks. In this relatively simple type of anisotropy, the elastic velocities did not vary in the horizon-tal direction but did vary in the vertical direction. A velocity model of this type is said to be trans-versely isotropic with a vertical axis of symmetry (TIV). Seismic waves generally travel faster along horizontal layers than vertically across them.

Another type of anisotropy, azimuthal anisot-ropy, may result from a natural fracture system approximated by a single set of vertical parallel fractures. Unequal horizontal stresses also cause azimuthal anisotropy, generating stress-aligned microfractures, whose apertures are too small to

>Natural fracture system within the Daanzhai (Da) limestone formation. A comparison of the combined legacy datasets from 1999 and 2003 (left ) and the UniQ dataset (right ) from 2013 shows distinct improvements in resolution brought by UniQ acquisition. Both displays are the result of modern prestack time migration (PSTM). The crossing sections in each image show reflection amplitude. The time-slice horizon in each image is at the same level within the Da formation and displays the variance of seismic reflection amplitude. The legacy datasets lack detailed resolution. The UniQ dataset shows rich detail about the pattern of the natural fracture system.

m0 1,000

0 ft 3,000

m0 1,000

0 ft 3,000

Reflection amplitude

– 0 +

Variance

Low High

32. For more on seismic characterization of natural fracture systems: Worthington MH: “Interpreting Seismic Anisotropy in Fractured Reservoirs,” First Break 26, no. 7 (July 2008): 57–63.

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.

Paddock D, Stolte C, Young J, Kist P, Zhang L and Durrani J: “Seismic Reservoir Characterization of a Gas Shale Utilizing Azimuthal Data Processing, Pre-Stack Seismic Inversion and Ant Tracking,” Expanded Abstracts, 78th SEG Annual International Meeting and Exposition, Las Vegas, Nevada, USA (November 9–14, 2008): 2777–2781.

33. Liang et al, reference 22.

34. The vibroseis technique was introduced by Conoco in 1952 and uses a truck-mounted vibrator plate coupled to the earth surface to produce ground motion from a sweep of frequencies that propagates into the subsurface. For more on the vibroseis technique: Bagaini et al, reference 1.

35. For more on AVO: Chiburis E, Franck C, Leaney S, McHugo S and Skidmore C: “Hydrocarbon Detection with AVO,” Oilfield Review 5, no. 1 (January 1993): 42–50.

36. Vermeer, reference 30. Stein et al, reference 31.

42 Oilfield Review

accommodate fluid flow. Stress-aligned micro-fractures extend in the direction of the maximum horizontal principal in situ stress and open in the direction of the minimum horizontal principal in situ stress. Seismic P-waves generally travel faster parallel to fracture planes or to the maxi-mum horizontal stress. If the fractures are open and large enough to allow fluid flow, they will dominate reservoir permeability. The simplest velocity model of this type is said to be trans-versely isotropic with a horizontal axis of symme-try (TIH); the open fractures are parallel to today’s maximum horizontal stress. In TIH mod-els, the effect of any TIV anisotropy is ignored.

Orthorhombic anisotropy is produced by superimposing the TIV (layer anisotropy) and TIH (fracture anisotropy) symmetries. This superposi-tion occurs when layered rocks are subjected to unequal horizontal stresses or when they contain a single set of vertical fractures. Anisotropy of

higher complexity (monoclinic or triclinic) may ensue from the superposition of other effects such as dipping layers or rocks with multiple fracture sets. Complex seismic anisotropy may result if current maximum horizontal stress azimuth were skewed to an older natural fracture system that formed in an earlier stress regime.

Prestack depth migration removes the effects of heterogeneity and layer anisotropy in the over-burden above the reservoir level targeted for imaging. When a PSDM record is sorted by offset and azimuth, the variation of traveltimes and reflection amplitudes with azimuth becomes clear (see “In Situ Stress, Natural Fractures and Seismic Anisotropy,” page 44).

Azimuthal velocity anisotropy may be associ-ated with the horizontal principal stress orienta-tions in the target reservoir. Fitted elliptical anisotropy from traveltimes (FEATT) analysis

enables geophysicists to find the directions of the fast and slow horizontal velocities. The OVT com-mon image point (CIP) gathers (CIGs) from PSDM were converted from depth to two-way traveltime, which was then converted to interval velocity.37 Using the FEATT method, geophysicists determine best-fit ellipses to the interval velocity data. The major and minor axes of the ellipses provide the fast (VP, int, fast) and slow (VP, int, slow) P-wave interval velocities and their directions.

The P-wave velocity anisotropy characteris-tics derived from these quantities may be dis-played on maps (left). If these maps have been properly calibrated to the local in situ stress con-ditions and natural fracture system, drilling engi-neers can use them to plan horizontal well trajectories that follow the direction of the mini-mum horizontal stress, and completion engineers can place hydraulic fracture stages in locations of low minimum stress.

Analyzing for AVAZ and AVOAZ effects is another method that geophysicists use to charac-terize reservoir fabric. The PSDM CIGs were sorted into four overlapping azimuthal sectors 60° wide, which from north were 15° to 75°, 55° to 115°, 95° to 155° and 135° to 195°. Within each sector, the data were sorted into four overlapping 15° sets of angle of incidence, which range from vertical to 45° incidence—from zero, or near, to far offset—and ordered 0° to 15°, 10° to 25°, 20° to 35° and 30° to 45°. The resulting data deliverables con-sisted of 16 volumes of imaged data. From these volumes of limited-azimuth and limited-offset data, the top-of-the-reservoir reflection amplitude was extracted for each CIG, in which the bin size was 100 ft × 100 ft [30 m × 30 m].

The FEATT velocity anisotropy results corre-lated to the in situ stress orientation data obtained from borehole and hydraulic stress analyses. The fast interval velocity direction aligned with the present-day maximum principal stress direction, as verified by monitoring micro-seismicity. Areas characterized by high ΔVP, int, defined as VP, int, fast minus VP, int, slow, and by low VP, int, slow coincided with zones characterized by low minimum horizontal principal stress. These

> P-wave velocity anisotropy. This horizontal slice at reservoir depth (inset ) is color coded with azimuthal velocity anisotropy, calculated as the difference between VP, int, fast and VP, int, slow divided by VP, int, fast. The direction and length of the red lines indicate the orientation of VP, int, fast and P-wave velocity anisotropy expressed as a percentage, respectively. Velocity anisotropy in this field has been shown to be directly proportional to the local in situ principal stress anisotropy. The black lines are existing horizontal wells. The two wells on the left are oriented in the direction of minimum stress predicted by seismic azimuthal anisotropy; the well on the right was drilled almost parallel to the maximum stress direction and hydraulic fractures are likely to propagate parallel to the well. The dark blue zone (white oval) to the left of the leftmost well is a possible fault. On the west side of this fault, the maximum stress orientation is SW–NE and parallels the fault plane; on the east side, it is NW–SE and orthogonal to the fault. Geomechanical studies have revealed that faults may alter the stress field near them.

Azimuthal P-wave velocity anisotropy, %

0 20

m0 500

0 ft 2,000

37. Common image point (CIP) gathers (CIGs) are subsets of an image that have fixed surface locations. They facilitate the measurement of variations in seismic attributes and petrophysical properties at fixed locations, or image points. CIGs may be analyzed as a function of offset or of reflection angle. An alternative term for CIG is common reflection point (CRP) gather.

38. Lynn HB, Lynn W, Obilo J and Agarwal V: “Azimuthal Pre-Stack Depth Migration for In-Situ Stress Evaluation in a Fractured Carbonate Oil Reservoir: Predrill Prediction of Instantaneous Shut-In Pressure Gradients,” Expanded Abstracts, 84th SEG Annual International Meeting and Exposition, Denver, October 26–31, 2014 (in press).

Summer 2014 43

locations would be prime candidates for hydrau-lic fracture stimulations if optimally oriented horizontal wells were drilled into them.

Based on FEATT data, the geophysicists were able to predict hydraulic fracturing behavior for horizontal Well B (above). Fracturing would be easier in the north, at the toe of the well, where ΔVP, int was highest and VP, int, slow was lowest. Conversely, fracturing would be more difficult and variable in the south, at the heel, where ΔVP, int was

low and VP, int, slow was high. Observations of micro-seismicity from hydraulic stimulations corrobo-rated these predictions. The instantaneous shut-in pressures (ISIPs) from the first nine hydraulic fracture stimulations in Well B were directly pro-portional to VP, int, slow. Together with well log deter-minations of stress gradients, these results suggest that calibrated VP, int, slow may be used as a predic-tor of relative minimum horizontal stress magni-tudes and, therefore, ISIP levels.38

Using images from a Baker Hughes StarTrak high-definition LWD resistivity imaging log in Well B to identify the orientation of open fractures, the geophysicists found from AVAZ results that the near-offset data showed the highest reflection amplitudes (AP, max) coincided with the strike of open preexisting natural fractures. In some areas of the field, these azimuths of AP, max coincided with the SW–NE present-day maximum in situ compressive stress direction—that of VP, int, fast.

> Calibrating P-wave velocity anisotropy to in situ stress. Horizontal Well B (top left ) was drilled from SSE to NNW. From two monitor wells (circles), seismic engineers monitored microseismicity during 12 hydraulic fracture stages. P-wave velocity anisotropy is superimposed on the top-of-the-reservoir surface (contours), which dips NE; the contour interval is 100 ft [30 m]; the size of the underlying grid cells is 2,000 ft × 2,000 ft [610 m × 610 m]. The tack icons point in the direction of fast P-wave interval velocity (VP, int, fast), their color indicates slow interval velocity (VP, int, slow) and their length is the difference (ΔVP, int) between the fast and slow interval velocity. After the well was stimulated (top right ), engineers measured the magnitudes of the prestimulation (black squares) and poststimulation (blue squares) instantaneous shut-in pressure (ISIP). The icons indicate the same properties as those in the figure on the left, and the background color is VP, int, slow. The ISIP is a measure of the magnitude of the minimum horizontal principal compressive stress, which increases in magnitude from Stage 1 at the toe to Stages 10 and 11, toward the heel. The ISIP is smallest where VP, int, slow is low (red), ΔVP, int is large (long tacks) and the VP, int, fast direction is NE–SW, about 45° from the trajectory of Well B. The ISIP is highest where VP, int, slow is relatively high (green), ΔVP, int is relatively small (short tacks) and the VP, int, fast direction is more closely aligned with the borehole. A comparison of VP, int, slow with the ISIP gradient (bottom, black squares) shows a direct proportionality between them, except for Stages 10 and 11 (blue squares), which had high ISIP gradients and were difficult to fracture. These combined datasets suggest that ΔVP, int magnitudes correlate with in situ horizontal stress anisotropy, the VP, int, fast direction correlates with the maximum horizontal principal compressive stress direction and the VP, int, slow magnitudes correlate with the minimum horizontal principal compressive stress magnitude.

Monitor Well 1

Toe

Heel

Monitor Well 2

Well B

Monitor Well 1

Monitor Well 2

Well BN N

Slow

P-w

ave

inte

rval

vel

ocity

, ft/

s

15,500

15,000

14,500

14,000

13,500

ISIP, or minimum stress, gradient along the wellbore, psi/ft0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00 1.05

Stages 1 through 9Stages 10 and 11Regression line

Toe

Heel

P-wave velocity anisotropyFracture stage and sizePrejob ISIP gradientPostjob ISIP gradient

Slow P-wave interval velocity, ft/s

13,000 14,000 15,000 16,000 17,000 18,000

(continued on page 46)

The economic success of an unconventional play depends largely on the effectiveness of the fracture stimulation. Primary factors that govern rock fracture are the rock strength, the minimum horizontal principal compressive stress magnitude and the orientation and den-sity of open natural fractures. Operators can extract this information from seismic data to plan oilfield drilling, completion and stimula-tion operations.1

In Situ StressIn situ stresses, which can be resolved into the three principal stresses, affect seismic velocity anisotropy in rock because they determine which microfractures are open and which are closed. The vertical stress is governed by the weight of the overlying rock. Because the pres-ent-day horizontal stresses usually are not equal to each other, the microfractures ori-ented nearly perpendicular to the maximum horizontal stress tend to squeeze shut, while microfractures oriented nearly parallel to the maximum horizontal stress tend to remain open. These latter fractures are stress-aligned

microfractures, and their orientation deter-mines how seismic traveltimes, and their inverse—velocities—vary with azimuth.2 The maximum horizontal stress direction, which is parallel to the open stress-aligned microfrac-tures, is the fast direction—it has the shortest traveltime and highest velocity. The minimum horizontal stress parallels the slow direction—it has the longest traveltime and lowest veloc-ity. These phenomena characterize a common trait seen in nearly all formations, which is known as velocity anisotropy.

A common image point (CIP) gather shows this phenomenon (below left). The seismic sec-tion displays a CIP gather of data focused on a target reflector (below). The waviness indi-cates variations in seismic traveltime and the colors depict seismic reflection amplitude. The far-offset traces are diagnostic and show the azimuthal variations of traveltime and ampli-tude best because there are more far-offset traces than near, and the far-offset seismic ray-paths are more horizontal than the near ray-paths. In the CIP section, the fast direction is in the SE quadrant at the azimuth of minimum traveltime and maximum velocity. The slow direction is in the NE quadrant at the azimuth of maximum traveltime and minimum velocity.

The maximum amplitudes—the darkest blues and reds—are in the ENE quadrant. Although this occurrence is most evident in the far-offset bands, it is also visible at near offsets. In addition, the maximum reflection ampli-tudes do not coincide with either the fast or the

slow velocity direction. Reflection amplitudes need not respond to the same characteristics of the rock as seismic velocities.

For each CIP within a 3D survey, the CIP gather is used to measure the azimuthal travel-time variation around it. Geophysicists use the fitted elliptical anisotropic traveltime (FEATT) workflow to examine azimuthal traveltimes and output the fitted ellipse to the velocities indi-cated by each source-receiver traveltime and for each event of interest. After application of FEATT and removal of the elliptical residual moveout from the reflections, the fast and slow interval velocities and azimuths are available for further analysis. Geophysicists plot these FEATT results on maps (next page, top left).

Two hydraulic fracture stages are featured on the map. The general trend of microseismic-ity from Stage 2 is summarized by an ellipse aligned SW–NE, indicating the orientation of the maximum horizontal stress. This azimuth is also the azimuth of VP, int, fast. Microseismic events in Stage 1 were anomalous and trended NW–SE. Further analysis and calibration of azi-muthal P-wave reflection amplitude anisotropy indicated that the Stage 1 microseismicity fol-lowed a natural fracture trend.

44 Oilfield Review

In Situ Stress, Natural Fractures and Seismic Anisotropy

> Understanding a common image point (CIP) gather. The image point, or reflection point, is at the center. Squares are locations of sources and receivers of traces contributing to the image point. Colors and numbers show regions of similar offset. Only the eastern half is colored; assuming reciprocity, the traces in the NW quadrant are the same as those in the SE quadrant, and those in the SW quadrant are the same as those in the NE quadrant.

NCIP Gather

S

EW 1 2 3 4 5 6 7

> Common image point section. From left to right, the traces in the section are arranged in offset bands from near to far offset and then within each band by azimuth. The vertical lines separate each band. The colors and numbers below the section are keyed to the CIP gather in the figure on the left. The number of traces within a band controls its width; there are fewer near-offset traces than far-offset traces. Within each band, from left to right, trace azimuth ranges from north to south.

Trav

eltim

e

Traces Ordered by Offset then by Azimuth

Reflection amplitude

– 0 +

1 2 3 4 5 6 7

1. Rich JP and Ammerman M: “Unconventional Geophysics for Unconventional Plays,” paper SPE 131779, presented at the SPE Unconventional Gas Conference, Pittsburgh, Pennsylvania, USA, February 23–25, 2010.

2. Crampin S: “Geological and Industrial Implications of Extensive-Dilatancy Anisotropy,” Nature 328, no. 6130 (August 6, 1987): 491–496.

3. The AVO gradient is the rate, or slope, of the reflection amplitude variation with offset, or angle of incidence.

Natural FracturesNatural fractures formed in a paleostress regime, either as tensile or shear fractures, in accord with extant in situ stresses and rock strengths. In some geologic settings, the paleostress is similar to present-day stress.

Evidence of natural fractures appears in the azimuthal variation of reflection amplitudes. Seismic reflections have higher spatial and temporal resolution than those of seismic trav-eltimes. Reflection amplitude depends on the P-wave and S-wave impedance of the rock above and below a reflector and is a function of

incidence angle. This phenomenon is called amplitude variation with offset (AVO) or ampli-tude variation with angle of incidence (AVA).

Originally, geophysicists studied the azi-muthal variation of the AVO gradient, or amplitude variation with offset and azimuth (AVOAZ).3 In the presence of anisotropy, such as a high fracture density of one set of frac-tures either above or below a reflector, then the AVO gradient will vary with azimuth. However, in the absence of azimuthal anisot-ropy across the interface, the AVO gradient will not vary with azimuth.

Recently, however, geophysicists have started focusing on the amplitude variation with azimuth (AVAZ) for a fixed range of offsets, in particular near-offset and midoffset traces because they can show azimuthal variation.

The AVAZ of a reflector in a CIP gather is most pronounced on the far offsets and is a reason industry experts have conventionally looked only at the far offsets to see the AVOAZ effect. However, all offset bands show the AVAZ effect. Near-offset AVAZ results are sum-marized on a map (above). The data were

Summer 2014 45

> P-wave interval velocity anisotropy. Results of FEATT analysis are superimposed on a contour map of the top-of-the-reservoir horizon; the contour interval is 50 ft [15 m]. The tacks display azimuthal P-wave velocity anisotropy and are on a grid spacing of 200 ft × 200 ft [60 m × 60 m]. At each grid point, the tack points in the direction of the fast P-wave velocity (VP, int, fast), its length indicates the magnitude of the velocity anisotropy (ΔVP, int) and its color indicates the magnitude of the slow P-wave velocity (VP, int, slow). The background colors are VP, int, slow. The squares represent prejob (black) and postjob (blue) instantaneous shut-in pressure (ISIP) gradients measured before and after each hydraulic fracture (HF) stimulation stage (large black numbers) in Well B. Microseismicity (circles and dashed ellipses) from the Stage 1 (red) and Stage 2 (black) HFs are shown. Most events from the Stage 2 HF trend SW–NE, which is similar to the direction of the azimuth of VP, int, fast. Many HFs from this well propagated along the SW–NE trend. The Stage 1 HF propagated along a NW–SE trend, attributable to preexisting natural fractures.

Well B

N

P-wave velocity anisotropy

m0 300

0 ft 1,000

Slow P-wave interval velocity, ft/s( and background)

13,000 14,000 15,000 16,000 17,000 18,000

> P-wave reflection amplitude anisotropy. Results of near-offset azimuthal variation with azimuth (AVAZ) analysis are superimposed on a contour map of the top-of-the-reservoir horizon; the contour interval is 50 ft [15 m]. The triangles display azimuthal P-wave amplitude anisotropy and are on a grid spacing of 100 ft × 100 ft [30 m × 30 m]; there are no icons where the data are unreliable. The triangles represent the near-offset AVAZ data, which include angles of incidence from 1° to 15°. The icon color represents the magnitude of the amplitude anisotropy (ΔAP), which is the maximum (AP, max) minus the minimum (AP, min) amplitude and is interpreted as the azimuthal variation in the P-wave impedance of the reservoir layer. The triangles point in the direction of AP, max, and their length is the reliability of the anisotropy determination. The bar icons display natural fracture trends from StarTrak resistivity image logs run in Well C; their length represents fracture density, their color is fracture aperture (from blue to purple to red signifies small to medium to large) and their azimuth is fracture orientation. These resistivity data were used to calibrate and confirm that the AVAZ data were indicative of trends of open natural fractures. The background color is the mean reflection amplitude of the near offsets. The squares are proportional to the prejob (black) and postjob (blue) ISIP gradients.

Well AWell C

N

P-wave amplitude anisotropyNatural fracture orientation

Mean amplitude (background)

– +0

m0 100

0 ft 400

Amplitude anisotropy, percent mean amplitude

0 100 200 300

( )

46 Oilfield Review

However, in other areas, these azimuths of AP, max varied from oblique to perpendicular to the pres-ent-day maximum stress direction. Geophysicists interpreted this observation to indicate that in these areas, the fracture system formed in an older in situ stress regime whose orientation dif-fered from that of the present day.

The geoscientists have concluded from micro-seismic observations that open natural fractures appear to have aided the propagation of hydraulic fractures for Stage 1 in Well B (next page). Stage 1 occurred at the well’s toe, where AVAZ analysis suggested the existence of a NW–SE fracture set oblique to the well trajectory. Microseismicity pat-terns indicated that the hydraulic fractures fol-lowed this NW–SE open natural fracture set rather than the present-day SW–NE maximum stress direction. During Stages 2 through 8, microseismic monitoring showed that the hydraulic fractures followed both the SW–NE maximum stress direc-tion indicated by FEATT and the natural fracture strikes indicated by AVAZ.39

The 3D UniQ land seismic dataset was of suffi-ciently high quality and resolution that it yielded a high-fidelity PSDM result. Geophysicists were able to process and analyze the data for anisotropy parameters, or attributes, that related P-wave velocity anisotropy to the present-day in situ hori-zontal stress directions and P-wave amplitude anisotropy to natural fracture orientations.40 The operating company engineers are using these anisotropy attributes to plan drilling, completions and hydraulic fracture stimulation operations.

UniQ FutureThe UniQ land seismic acquisition point-receiver technology enables acquisition of FAZ and long offset surveys that have small bin sizes of high fold. These surveys illuminate reservoir targets from all directions, which ensures high SNR and target definition. UniQ point receivers deliver the amplitude and phase fidelity required for charac-terizing sweet spots. These qualities enable geol-ogists, geophysicists and reservoir engineers to plan drilling and completions operations.

39. Lynn HB: “Fracture Densities and Fracture Azimuths Evident in the Azimuthal Amplitudes from the Top of a Fractured Carbonate Oil Reservoir,” Expanded Abstracts, 84th SEG Annual International Meeting and Exposition, Denver, October 26–31, 2014 (in press).

Lynn HB: “Field Data Evidence of Orthorhombic Media: Changes in the P-P Bright Azimuth with Angle of Incidence,” Expanded Abstracts, 84th SEG Annual International Meeting and Exposition, Denver, October 26–31, 2014 (in press).

40. Lynn HB: “Azimuthal Anisotropy: Distinguishing Between Unequal Horizontal Stress and Vertical Aligned Macro-Fractures, as Demonstrated in Thirty Years of Field Data Analysis,” Expanded Abstracts, 84th SEG Annual International Meeting and Exposition, Denver, October 26–31, 2014 (in press).

calibrated against borehole image data. The AVAZ amplitudes showed the highest ampli-tudes when measured parallel to open natural fractures observed in the image logs.

Microseismicity from Stage 1 (above) fol-lowed a NW–SE natural fracture set. Stage 2 and subsequent stages tended to follow natu-ral fractures trending SW–NE. In this dataset, the AVAZ and the AVOAZ results were consis-tent with the calibration data.4

Seismic Anisotropy for Engineering ApplicationsThe effectiveness of hydraulic fracture stimu-lations depends on the rock strength, mini-mum horizontal principal compressive stress and the orientation and density of open natu-ral fractures. Drilling, completion and stimula-tion engineers would like to know these quantities before spudding wells.

Properly acquired and processed seismic data can tell geophysicists the VP, int, fast direc-

tion, ΔVP, int magnitude and the VP, int, slow mag-nitude. These quantities may be calibrated and correlated to the local maximum horizon-tal stress direction, the horizontal stress anisotropy and the local minimum horizontal stress magnitude.5 In addition, seismic data can tell geophysicists the ΔAP, max direction and magnitude, which may be related to the orientation and intensity of the local natural fracture trend.6 With maps based on these cal-ibrated quantities, engineers can plan drilling, completion and stimulation operations.7 The data example shown here illustrates how care-fully acquired, processed and interpreted data can be of great value to engineers.

> P-wave reflection AVAZ. Results of AVAZ analysis are superimposed on a contour map of the top-of-the-reservoir horizon; the contour interval is 50 ft [15 m]. The triangle icons represent near-offset AVAZ data. Microseismicity (circles and ellipses) from the Stage 1 (red) and Stage 2 (black) HFs in Well B are shown. The microseismicity from Stage 1 propagated northwest along a NW–SE trend, and that from Stage 2 propagated southwest along a SW–NE trend. Both propagation directions are parallel to the maximum reflection amplitude (AP, max) direction, interpreted as indicating the local trend of open natural fractures. The background color is the mean reflection amplitude.

Well B

N

m0 300

0 ft 1,000

P-wave amplitude anisotropy

Mean amplitude (background)

– +0

Amplitude anisotropy, percent mean amplitude

0 100 200 300

( )

4. Lynn et al, reference 38, main text.5. Simon YS: “Stress and Fracture Characterization in a

Shale Reservoir, North Texas, Using Correlation Between New Seismic Attributes and Well Data,” MS thesis, University of Houston (2005).

6. Lynn, reference 39, main text.7. Ajayi et al, reference 2, main text.

Summer 2014 47

After the appropriate seismic processing, azimuthal and offset analyses of calibrated P-wave data from UniQ surveys can reveal infor-mation about natural fractures and in situ stress within target horizons. The case studies demon-strate how seismic attributes, such as velocity and amplitude anisotropy, P- to S-velocity ratio, P-wave impedance, reflection amplitude vari-

ance and ant tracking, can be associated with RQ and CQ factors for distinguishing lithology, delineating natural fracture systems and char-acterizing the in situ stress regime. These prop-erties are critical for planning pilot exploration wells, horizontal appraisal wells and horizontal production wells in both conventional and unconventional reservoirs.

In the future, 3D seismic surveys containing closely spaced point receivers and point sources will expand the usefulness of seismic data—from a tool used mainly for exploration to one used for res-ervoir engineering—leading to more efficient exploitation of challenging reservoirs. For opera-tors, this offers a reduction in exploration, develop-ment and production costs and risks. —RCNH

> Calibrating P-wave velocity attributes to microseismicity. Microseismicity (left ) observed during hydraulic fracturing in Well B showed that all stages except Stage 1 (red circles) propagated fractures along a SW–NE trend as indicated by the ellipses for Stage 2 (dashed black ellipse, black circles), Stage 5 (dashed green ellipse, green stars) and Stage 8 (dashed brown ellipse, brown circles). The ellipses are elongated parallel to the VP, int, fast direction (colored tacks), which is interpreted to be the maximum horizontal stress direction. The tacks, which display P-wave velocity anisotropy, are oriented in the VP, int, fast direction, their color indicates VP, int, slow and their length corresponds to ΔVP, int. Analysis of near-offset AVAZ for P-wave reflection amplitude (AP) anisotropy (right, colored triangles) indicates that one set of natural fractures dominates the results; the maximum amplitude (AP, max) direction coincides with the local, natural fracture strike direction. The triangles display AP anisotropy (ΔAP); their direction points in the direction of AP, max, their color describes the ΔAP and their length is the reliability of the measurement at their location. Most fracture stages, represented by Stages 2, 5 and 8, follow the SW–NE AP, max trend, which corresponds to the direction of the maximum horizontal stress direction. Stages 10 and 11 were difficult to fracture and correspond to no reliably identifiable ΔAP trend (no triangles plotted). The gap in AVAZ results at this location is interpreted as a lack of a dominant set of open natural fractures. The Stage 1 fracture (inset) propagated along a NW–SE trend, parallel to the direction of AP, max, perpendicular to the VP, int, fast direction, at about 45° from the trajectory of Well B, and appears to be following a preexisting open natural fracture system.

Monitor Well 1

Monitor Well 2

Monitor Well 1

Monitor Well 2

Slow P-wave interval velocity, ft/s (background)

13,000 14,000 15,000 16,000 17,000 18,000

Monitor Well 1

N N

P-wave velocity anisotropyFracture stage and sizePrejob ISIP gradientPostjob ISIP gradientP-wave amplitude anisotropy

Slow P-wave interval velocity, ft/s( and background)

13,000 14,000 15,000 16,000 17,000 18,000

Slow P-wave interval velocity, ft/s(background)

13,000 14,000 15,000 16,000 17,000 18,000

Amplitude anisotropy, percent mean amplitude

0 100 200 300

( )

Amplitude anisotropy, percent mean amplitude

0 100 200 300

( )

48 Oilfield Review

PDC Bit Technology for the 21st Century

Polycrystalline diamond compact bits have led the way to greater drilling efficiencies

in many recent plays. However, as operators push the limits of depth, temperature and

distance in pursuit of new reserves, they also push the limits of drillbit life and

efficiency. Recent advances in cutter technology are enhancing bit performance and

durability across a wider range of lithologies than was previously possible.

Greg BrutonChesapeake Operating, Inc.Oklahoma City, Oklahoma, USA

Ron CrockettMalcolm TaylorNovatekProvo, Utah, USA

Dave DenBoer Jeff LundProvo, Utah

Craig FlemingRobert FordGary GarciaAllen WhiteSmith BitsHouston, Texas, USA

Oilfield Review Summer 2014: 26, no. 2. Copyright © 2014 Schlumberger.For help in preparation of this article, thanks to Diane Jordan, Mark Teel, Rick von Flatern and Eric Wilshusen, Houston; and Maurizio Scordella, Milan, Italy. DRS, IDEAS, ONYX 360 and Stinger are marks of Schlumberger. STRATAPAX is a mark of General Electric.

The modern drill bit is a product of years of refinement in materials and designs aimed at increasing rate of penetration, improving wear resistance and extending bit life. One of the most significant changes occurred in the 1970s, when a synthetic diamond material was used to create

the polycrystalline diamond compact (PDC) bit. Man-made, polycrystalline diamonds possess extreme hardness—similar to that of natural dia-mond, the world’s hardest naturally occurring substance—with the strength and durability of tungsten carbide, which is used extensively in roller cone bits.

> End of a PDC bit run. Significant declines in rate of penetration (ROP) and drilling torque inevitably lead to increased weight on bit (WOB) and bit wear. (Adapted from Warren et al, reference 3.)

0

3,000

6,000

9,000

12,000

Wei

ght o

n bi

t, lb

f

0

1,000

2,000

3,000

4,000

Torq

ue, f

t/lb

f

00 1 2 3 4

Time, min

Bit Performance Data

5 6 7 8 9

100

200

300

400

Rate

of p

enet

ratio

n, ft

/h

1. Durrand CJ, Skeem MR, Crockett RB and Hall DR: “Super-Hard, Thick, Shaped PDC Cutters for Hard Rock Drilling: Development and Test Results,” paper SGP-TR-188, presented at the 35th Workshop on Geothermal Reservoir Engineering, Stanford University, Stanford, California, USA, February 1–3, 2010.

2. Durrand et al, reference 1.3. Warren TM, Brett JF and Sinor LA: “Development of a

Whirl-Resistant Bit,” SPE Drilling Engineering 5, no. 4 (December 1990): 267–274.

4. Brett JF, Warren TM and Behr SM: “Bit Whirl: A New Theory of PDC Bit Failure,” paper SPE 19571, presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, USA, October 8–11, 1989.

5. Bit bounce most frequently occurs while drilling vertically through hard formations as slight axial displacements repeatedly lift the bit off bottom and slam it back down. Stick-slip is caused by friction buildup between the BHA and formation, which compels the bit to momentarily slow or stop turning. When torque within the drillstring overcomes these frictional forces, the BHA releases from the wellbore wall, causing the BHA and bit to spin as the drillstring rapidly unwinds. Bending, caused by placing too much downward force on the drillstring, creates lateral shocks when the drillstring is deformed enough to make contact with the wellbore.

For more on downhole shock and vibration and their effects on drillbit design: Centala P, Challa V, Durairajan B, Meehan R, Paez L, Partin U, Segal S, Wu S, Garrett I, Teggart B and Tetley N: “Bit Design—Top to Bottom,” Oilfield Review 23, no. 2 (Summer 2011): 4–17.

Summer 2014 4949

The introduction of PDC cutters and associ-ated bit designs fundamentally changed the mechanics of drilling. Instead of gouging and crushing the rock, as does a roller cone bit, PDC bits use a transverse shearing motion. Although the three-cone roller bit dominated the industry for much of the 20th century, it may not be par-ticularly suited for some of the challenges facing drillers today. High temperatures typical of deep formations and high rotary speeds produced by downhole drilling motors led to damaged seals, worn bearings and bit failure. Since the introduc-tion of the fixed blade PDC bit, the industry has gradually moved away from roller cone bits.1 By 2004, footage drilled by the PDC bit surpassed that of the roller cone bit.

Although PDC bits were used extensively throughout a range of drilling environments, in certain drilling applications, PDC bit perfor-mance pales in comparison to that of roller cone bits. For example, hard carbonates and abrasive sandstones can be problematic for any bit. In such formations, PDC bits tend to drill at a higher rate of penetration (ROP) than either roller cone or diamond-impregnated bits, but at some point, shearing efficiency declines and ROP decreases abruptly, which typically prompts drillers to increase weight on bit (WOB) to main-tain ROP.

As WOB increases, PDC cutter edges wear flat, and drilling efficiency decreases further as

the worn bit starts crushing the rock rather than shearing it. Frictional energy generated by increased WOB heats the cutter, which leads to thermal degradation of the PDC.2 When torque and ROP abruptly decline, the driller is forced to pull out of the hole to replace the worn bit (previous page).3

As manufacturers worked to improve PDC thermal characteristics, Amoco Production Company field tests conducted during the late 1980s revealed that impact loading caused by excessive downhole vibration was another signifi-cant contributor to PDC cutter wear and failure.4 Downhole vibrations are often linked to the phe-nomena of bit bounce, stick-slip and bending.5 These problems can be monitored and corrected

50 Oilfield Review

from the drill floor. The same is not the case for bit whirl—another important contributor to downhole shock and vibration. To address this problem, bit manufacturers turned to their design engineers for a solution.

Bit whirl occurs when a bit’s axis of rotation is not in line with the bit’s physical center. Bit whirl produces severe lateral movement of the bit. During whirl, one of the cutters becomes an instantaneous center of rotation, such that the bit rotates about this contact point rather than about the central axis. The resulting asymmetric cutting action pushes one side of the bit against the borehole wall, producing an overgauge hole and additional friction. As the bit rotates about this contact point, friction builds, and torque in the drillstring increases, which can force the bit to move backward relative to the surface rotation of the drillstring, or laterally, creating high-impact loads on the bit and BHA.6

To alleviate this problem, manufacturers ini-tially developed variations on a PDC bit that used a large noncutting wear pad to slide the bit into the borehole wall and prevent walking, thus decreasing bit whirl.7 Since then, antiwhirl bit design has evolved considerably.

The drive to increase bit longevity led to numer-ous design modifications: thicker diamond layers, additional cutters, more blades, smaller cutters and increased back rake, or decreased contact angle, of the cutters. Materials research produced advances in diamond formulations and helped increase thermal stability. Improvements in the manufacturing process led to changes in diamond press engineering and sintering processes.8

Smith Bits, a Schlumberger company, recently introduced two innovations that are changing the ways in which PDC bits drill: • a single conical diamond element placed at the

center of the bit to create a PDC bit that pro-duces a combination of shearing and crushing action

• rotating cutting elements, which distribute wear evenly on the cutter edge to extend cutter life beyond that of premium fixed PDC cutters.

This article discusses the development of the Stinger conical diamond element and the ONYX 360 rolling PDC cutter. Case studies demonstrate how these new bit designs are expanding the application of PDC bits into challenging drilling environments while helping operators improve drilling efficiency and extend bit life.

PDC Basics: Design and TerminologyThe PDC bit was a radical departure from the conventional roller cone bit. To capitalize on the shearing action of the polycrystalline diamond compact, bit designers developed a specialized cutting structure.

In a PDC bit, the compact is a disk made of polycrystalline diamond, synthesized by sintering diamond grit with a catalyst under high pressure and temperature. During the manufacturing pro-cess, the diamond grit—an aggregate of ran-domly oriented fine and ultrafine synthetic diamond particles—is fused with cobalt [Co] under extreme pressure and heat to produce a cylinder of polycrystalline diamond (see “One Step Further: PDC Manufacturing,” page 52). Unlike natural diamond [C], which fractures along crystallographic planes, polycrystalline diamond, with its randomly oriented synthetic diamond matrix, has no preferred cleavage planes, thus making the PDC cutter extremely hard and resistant to impact and wear. A shift to multimodal diamond grit—using a range of grit sizes that permits smaller particles to fill voids between large particles—has helped further increase wear resistance.

The cutter consists of two parts: a polycrystal-line diamond table and its substrate (above left). PDC manufacturers refer to the flat cylinder of synthetic diamond as a table.9 The table is the part that comes in contact with the formation. Tables typically range in thickness from 2 to 4 mm [0.08 to 0.16 in.]. Some tables have a slight bevel that reduces stress on the cutter as it makes ini-tial contact with the rock at the instant it starts to cut. Although the beveled edge can reduce bit aggressiveness, it helps increase durability and diamond table impact resistance.

The diamond table is sintered to a hard sub-strate composed of tungsten carbide [WC]. Unlike many substances, tungsten carbide is able to bond to diamond, so while the substrate gives structural support to the diamond table, it also provides a medium that can withstand brazing—a process used to mount the cutter to the bit.10 The substrate diameter conforms to that of the table and is typically about 1.3 cm [0.5 in.] across. The development of a nonplanar interface between the table and substrate helped reduce stress and strengthened the bond between the diamond and tungsten carbide.11

A step change in extending longevity came with the invention of the leached PDC. The leach-ing process removes interstitial cobalt—a cata-lyst used in sintering the conventional PDC—from several microns of the diamond’s outer surface. The Co is introduced to the PDC through the WC substrate during the high-pressure, high-tempera-ture (HPHT) sintering phase of the manufactur-ing process. During sintering, Co melts and is forced into the diamond porosity, where a cata-lytic reaction produces intergranular diamond bonds. At high temperatures, however, Co also catalyzes the reversion of diamond to graphite, which weakens the PDC. Leaching Co from syn-thetic diamond improves its resistance to abra-sion while reducing the effects of differential thermal expansion between diamond and cobalt.12

The cutters are mounted along the surface of the PDC bit by brazing the substrate to a blade. Three to eight blades—sometimes more—are arrayed along the outside of the bit, radiating outward from its nose (next page, top left). The number of blades varies according to the intended application, as does the number of cut-ters mounted to each blade. Bit designers must consider cutter number and placement when specifying depth of cut. Reducing the number of

6. Brett et al, reference 4. 7. Warren et al, reference 3. 8. Durrand et al, reference 1. 9. Gemologists define a table as the largest facet on a gem.10. Brazing is a process, similar to soldering, in which a

melted filler is used to join metals or ceramics together. Two work pieces are heated to a temperature higher than the melting point of the filler, but lower than the melting points of the work pieces. The melted filler is distributed between the close-fitting work pieces by capillary action. When the filler cools and solidifies, it joins the pieces together.

11. Clegg J: “Faster, Longer, and More-Reliable Bit Runs with New-Generation PDC Cutter,” paper SPE 102067, presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, September 24–27, 2006.

12. Durrand et al, reference 1.13. Hardbanding is a manufacturing process in which an

alloy coating is applied to an exposed metal surface to protect the metal from abrasive wear. The alloy used must be harder than the metal it protects.

> Polycrystalline diamond compact cutter. Each polycrystalline diamond compact is made up of a diamond table (black) and a tungsten carbide substrate (gray). A nonplanar interface (not shown) between the substrate and the diamond table creates a strong bond between the two when the diamond table is sintered to the substrate.

Diamond tableSubstrate

Summer 2014 51

cutters tends to increase the depth of cut but also increases cutter wear. As a bit rotates, cutters near the side, or gauge, of the bit travel a greater distance than those near the bit center; there-fore, to extend bit life, some designs reduce the spacing between cutters placed near the side of the bit. On some bits, two rows of cutters are brazed to each blade; backup, or secondary, cut-ters ride behind the primary cutters and may be

set deeper into the bit body. The secondary cut-ters engage the formation after the primary cut-ters have started to wear (above right).

Cutters are placed at an angle along the lead-ing edge of the blade. This attack angle, or back rake, controls how aggressively the cutters engage the rock (right). Smaller angles of back rake are used in soft formations, but hard forma-tions require a less aggressive bit and, thus, greater back rake. The back rake angle may also

vary between cutters, depending on their posi-tion along the length of a blade. While back rake can limit depth of cut and penetration rate, it can be helpful in reducing bit vibration and wear.

The body of a PDC bit is made from steel or from a matrix of tungsten carbide and an alloy to bind the matrix, which is sintered to a steel core. The selection of a steel or matrix bit body usually depends on the operator’s intended application. The ductility and strength of steel make steel PDC bodies resistant to impact loading, but a steel body is less resistant to abrasion than a matrix body. Because steel is softer than tung-sten carbide, hardbanding or other wear-resis-tant applications may be installed on certain parts of the bit to resist wear.13 The tungsten car-bide matrix is available in a variety of formula-tions to provide resistance to abrasion or can be > PDC bit components. The most prominent features of the PDC bit are its

blades and cutters. Various types of cutting structures are concentrated along the cone, nose, shoulder and gauge areas of the bit (top). Nozzles, junk slots and fluid courses (bottom) aid in removal of cuttings from the bit face.

Face cutter

Gauge cutter

Breaker slot

Shank bore

Fluid course

PDC cutter

Cone

Nose

Gauge pad

Weld groove

API pinconnection

Junk slot

Blade

Shoulder

Interchangeablenozzle

Interchangeablenozzle

Gauge insert

Back reamingcutter

> Backup cutters. Some PDC bit designs include a second row of cutters. The two rows of cutters—primary and backup—reinforce each other to increase bit durability.

Primarycutters

Backupcutters

> Back rake of PDC cutter. PDC bits drill through formations by shearing the rock. Back rake angle controls how aggressively the cutter engages the rock.

Bit body Mounting post

FormationPDC cutter

Back rakeangle

(continued on page 54)

1. Hall HT: “Ultra-High-Pressure, High-Temperature Apparatus: The ‘Belt’,” The Review of Scientific Instruments 31, no. 2 (February 1960): 125–131.

2. General Electric researchers succeeded in transforming carbon to diamond using graphite, carbon black and sugar charcoal as the carbon source; a wide range of catalysts were used, including chromium, manganese, iron, cobalt, nickel, ruthenium, rhodium, palladium and platinum.

Bovenkerk HP, Bundy FP, Hall HT, Strong HM and Wentorf RH: “Preparation of Diamond,” Nature 184, no. 4693 (October 10, 1959): 1094–1098.

3. Besson A, Burr B, Dillard S, Drake E, Ivie B, Ivie C, Smith R and Watson G: “On the Cutting Edge,” Oilfield Review 21, no. 3 (Autumn 2000): 36–57.

4. Bellin F, Dourfaye A, King W and Thigpen M: “The Current State of PDC Bit Technology,” World Oil 231, no. 9 (September 2010): 41–46.

A compact is a solid form created by pressing, or compacting, a finely ground powder inside a die. Some of the most common compacts are created by the pharmaceutical industry: Aspirin, vitamins and various medications are often manufactured in tablet form. These tablets start as powders, which are then compacted in a cold press.

For ceramics or metals, the process of cre-ating compacts is taken one step further—in addition to pressure, heat is used to transform powder to solid. A blend of fine metallic or ceramic powder is placed in a die and com-pressed under high pressure. The compressed particles fuse together as heat is applied in a process known as sintering. Given sufficient heat, partial melting results in diffusion of atoms between granules, reduction of porosity and increase in density. The powder granules adhere and form a solid when cooled. The pressures and temperatures needed to make such compacts are created within the confines of a hot press.

Most metals can be sintered, as can many nonmetallic substances such as silica and even diamond. Some substances require an additional step beyond that of the con-ventional sintering process. In liquid-state sintering, a second component interacts with the powder during heating. The melt-ing point of the second component is lower than that of the primary component. During sintering, the powder granules form a matrix while the second component melts to fill in the pore space between granules.

The production of synthetic diamonds also involves sintering. An ultrahigh-pressure, high-temperature belt press, invented by sci-entists at General Electric in 1954, led to the first commercial production of synthetic dia-monds.1 This press supplied the pressure and heat needed to convert carbon to diamond. Temperature and pressure varied depending on the carbon source and catalyst, but ranged

from 1,200°C to 2,400°C [2,200°F to 4,350°F] and 55,000 to 100,000 atm [5,570 to 10,130 MPa or 808,200 to 1,469,600 psi].2

General Electric invented the PDC cutter in 1971 and, after years of field testing, intro-duced the STRATAPAX line of PDC cutters in late 1976. To make the diamond powder used in the sintering process, thin circular layers of alternating graphite and cobalt were stacked in small cans and pressed to about 13,800 MPa [2 million psi] followed by resistive heating to about 1,500°C [2,700°F]. Molten cobalt, acting as a catalyst and solvent, dissolved the graphite and deposited a monocrystalline diamond grit.3

To produce a polycrystalline diamond com-pact, the diamond grit is packed in the press against a tungsten carbide–cobalt [WC–Co] substrate. The pressure is raised inside the press, thus compressing the diamond aggre-gate and increasing its density. The tempera-ture increases, and when the cobalt in the substrate reaches its melting point, it is instantaneously squeezed into the pores between the diamond particles. The cobalt serves as a catalyst to create bonds between diamond particles and bind the diamond table to the substrate.4

In the years that followed, numerous com-panies developed synthetic diamonds, diamond bearings, polycrystalline diamond inserts, cut-ters and associated product lines. Two such com-panies were founded by a member of the team that invented the original belt press. In 1955, H. Tracy Hall left General Electric to launch Novatek, which now develops specialized electronics, metals and supermaterials. In 1966, he founded MegaDiamond, which now manufactures ultrahard products using diamond technology. MegaDiamond was later acquired by Smith Bits, which in turn was acquired by Schlumberger. Both Novatek and MegaDiamond supply diamond bit cutters to Schlumberger through its Smith Bits subsidiary.

At Novatek, Hall developed the tetrahedral press in 1957 and the cubic press in 1966. A solid unitary-frame press, developed in 1999, is now in its fifth generation (next page). This computer-controlled press has a cubic frame and six instrumented cartridges—each with an anvil capable of exerting more than 35 million N [8 million lbf]. The 4,000-tonUS [3.6-million-kg] press can subject work-pieces—cubic cells with raw materials embedded in a pyrophyllite shell—to temper-atures up to 2,300°C [4,200°F] and pressures exceeding 7,500 MPa [1 million psi]. A com-puter interface allows the operator to control the press from a separate location and pre-cisely monitor anvil positions. Novatek uses this advanced press technology to manufac-ture the Stinger conical diamond element. The polycrystalline diamond on this conical element is approximately twice as thick as the diamond layer on a conventional PDC cutting element; its shape is optimized for strength in axial compression to deliver high point load-ing on the formation.

The ONYX 360 rolling PDC cutter is pro-duced by MegaDiamond. The company pro-duces an extensive line of engineered polycrystalline diamond products for drilling and mining, including PDC shear elements for fixed cutter bits and diamond-enhanced

52 Oilfield Review

One Step Further: PDC Manufacturing

inserts for roller cone bits and percussive mining bits.

The company operates three types of HPHT presses: the belt press, piston-cylinder press and cubic press. To manufacture polycrystalline diamond components, MegaDiamond engineers determine spe-cific material and wear demands for each customer’s application, then apply the appropriate press technology to produce the diamond grades needed for the job. Each press is capable of generating the ultra-high pressures—greater than 5,500 MPa [800,000 psi]—and high temperatures—1,500°C [2,700°F]—required to sinter polycrystalline diamond products, and each has advantages regarding sintering charac-teristics and the properties imparted to the final product.

The belt press is modified from the origi-nal HPHT design of the 1950s. To generate the extreme pressures required to sinter polycrys-talline products, the press utilizes two carbide punches that converge on a high-pressure cap-sule contained within a carbide die. The mod-ern belt press is well suited for producing large diameter products or multiples of smaller products. The name is taken from the concentric, shrink-fitted steel “belts” that pre-stress the inner carbide die, allowing it to withstand the immense internal pressure.

The piston-cylinder press, similar to a belt press, uses a high-pressure capsule contained within a cylindrical bore. Two free-floating carbide pistons pressurize the capsule when a load is applied by conical carbide anvils. The carbide die is supported by radial hydraulic pressure rather than by a series of steel belts.

The cubic press is the most current design. It relies on six carbide anvils attached to massive hydraulic cylinders converging simul-taneously on a cube-shaped, high-pressure capsule. This triaxial system generates isostatic high pressures suited for sintering products with complex 3D geometries. As with all MegaDiamond presses, the cubic system is integrated with a computerized control system to ensure optimal and consistent pres-sure, temperature and time during sintering.

Laboratory facilities at Novatek and MegaDiamond permit scientists and engineers to evaluate polycrystalline diamond perfor-mance in a variety of functional tests, simulat-ing real-world conditions. Data gathered in the test laboratory are fed to an iterative design cycle process, culminating in improved performance at the wellsite.

Summer 2014 53

> Cubic press. The center of the computer-controlled press is surrounded by six anvils (five of which are shown), each capable of exerting more than 35 million N. A workpiece (green, inset ) is placed within the center of the press, then the anvils simultaneously press against it.

54 Oilfield Review

manufactured to suit a particular formation or drilling application. The matrix material can withstand relatively high compressive loads but is more brittle than steel and has a lower resis-tance to impact loading than steel.

A Central ElementTo improve bit performance in a wider range of environments, bit manufacturers are testing PDC bits in abrasive and interbedded formations, where penetration rates and bit life commonly cause problems for drillers. These types of forma-tions are difficult to drill, regardless of the type of

bit used. Because vibration-induced impact dam-age is the primary mechanism for reducing bit life and ROP, bit manufacturers sought to develop a more stable fixed cutter bit that would reduce vibration and shock in these types of formations. Engineers with Smith Bits, working in conjunc-tion with bit designers from Novatek, experi-mented with cutter placement and count to improve drilling efficiency and mitigate vibration. They focused on the problematic cone, or central cutting area, of a conventional PDC bit face.

When a conventional PDC bit is used, the cen-termost portion of the borehole can be difficult to

remove. Because the velocity of conventional PDC cutters decreases with proximity to the center of the cutting structure, the cutters are least effec-tive at removing rock from the center of the bore-hole, especially in hard formations.14 Central cutters are subjected to the highest axial loads on the bit and can create large bit torque fluctua-tions. Changes in depth of cut cause these fluctua-tions, which occur when drillers change weight on bit or rotary speed, or when they drill through changing lithologies that have differences in unconfined stress. The torque variations alter the dynamic response of the bit, exposing it to damag-ing shock and vibration.15 These phenomena cre-ate an inefficient shearing mechanism at the center of the conventional PDC bit (left).

This characterization of the cone area led to development of the Stinger conical diamond ele-ment (below left). This conical diamond element (CDE) has an ultrathick polycrystalline diamond layer. The CDE is positioned at the center of the bit with the conical tip facing the rock. Its conical geometry and thick diamond structure create a robust and durable cutting element that delivers high point loading for effective formation fracture.

To evaluate CDE potential for improving pen-etration rates and increasing total footage, bit engineers subjected the CDE to testing on a verti-cal turret lathe (VTL). This apparatus uses a test bed of granite or quartzite to measure the cut-ter’s ability to fracture rock under varying condi-tions. Testing on the lathe showed the CDE had greater cutting efficiency and wear resistance than standard PDC cutters. For example, at a 5,300-N [1,200-lbf] threshold, a 0.5-mm [0.02-in.] depth of cut resulted in a 70% increase in cutting efficiency compared with that of the baseline PDC cutter; at 1.3-mm [0.05-in.] depth of cut, the CDE cutter showed a 35% increase.16

The next challenge was to incorporate the CDE into a PDC bit design. Smith engineers used finite element analysis (FEA) to design the bit. The IDEAS integrated drillbit design platform helped engineers selectively remove inefficient

> Cutter force and distance from the bit center. Cutters at the cone of a bit (orange circles) are subjected to the highest force (green), experience the lowest velocity (brown) and typically remove the smallest volume of rock.

Nor

mal

ized

cutte

r for

ce

Distance from bit center, in.

Nor

mal

ized

cutte

r vel

ocity

Relative cutter velocity

Relative cutter force

00 0.5 1.5 2.5 3.5 4.51.0 2.0 3.0 4.0 5.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Highest load cutters

> Conical diamond element (CDE). The Stinger CDE (left ) is manufactured in an advanced synthetic diamond press to produce a layer of diamond that is substantially thicker than that of a conventional PDC cutter (right ). The polycrystalline diamond material has been engineered to provide a higher level of impact strength and resistance to abrasive wear (graph, center).

Rela

tive

scal

e

Wear resistance

Impact strength

Diamond thickness

Stinger cross section

Diamond

PDC cross section0

0.25

0.50

0.75

1.00

1.25Stinger elementPDC

Diamond

Substrate

Substrate

14. Azar M, White A, Segal S, Velvaluri S, Garcia G and Taylor M: “Pointing Towards Improved PDC Bit Performance: Innovative Conical Shaped Polycrystalline Diamond Element Achieves Higher ROP and Total Footage,” paper SPE/IADC 163521, presented at the SPE/IADC Drilling Conference and Exhibition, Amsterdam, March 5–7, 2013.

15. Azar M, White A, Velvaluri S, Beheiry K and Johny MM: “Middle East Hard/Abrasive Formation Challenge: Reducing PDC Cutter Volume at Bit Center Increases ROP/Drilling Efficiency,” paper SPE/IADC 166755, presented at the SPE/IADC Middle East Drilling Technology Conference and Exhibition, Dubai, October 7–9, 2013.

16. Azar et al, reference 15.

Summer 2014 55

PDC cutters and scale back the blades on which they were mounted. This resulted in a void at the center of the bit. As drilling proceeds, this void permits a small column of rock to develop at the bit’s center. As this rock column forms, it becomes less confined. Designers positioned the CDE at the center of the bit (above). As the uncut rock column builds in height, the Stinger element imposes a point load on the column to fracture and crush it while the bit drills ahead.

Engineers also used FEA software to investi-gate the stress field at the precise point that the Stinger element indents the formation. Their study confirmed that, compared with standard PDC cut-ters, the Stinger CDE required less force at the contact point to cause fracture generation within the unconfined rock. In addition, the Stinger ele-ment contributes to bit centralization, thus reduc-ing the potential for damaging vibrations.

Smith engineers also needed to modify the bit nozzle orientation to efficiently clean and cool the new cutting structure. Using fluid dynamics software, they performed a hydraulic analysis in which nozzle positions were adjusted to enhance cuttings removal and cleaning of the conical ele-ment and borehole.

Full-scale testing of a CDE-equipped bit was conducted in a pressurized drilling simulator to validate the results of a 4D modeling study. The testing corroborated earlier experimental con-clusions, whereby a modified bit created a stress relieved rock column that was crushed by the Stinger element at the center of the hole (above right). These tests also revealed that CDE-equipped PDC bits generate much larger drill cuttings than do standard PDC bits, thus facili-tating better rock characterization by geologists or mud logging personnel.

Modeling and testing results convinced engi-neers that a single CDE positioned at the center of the bit would improve ROP performance and enhance dynamic stability. They selected an 83/4-in. PDC bit for field testing using a Stinger conical element positioned at the bit’s center. Field tests were conducted in the US; PDC bits with Stinger CDEs were successfully run in the Williston basin of North Dakota, in the Cotton Valley Formation of East Texas and in the inter-bedded sand and shales of the Wasatch Formation in Utah. Each test demonstrated substantial gains in ROP and reductions in bit wear.

Following successful runs in North America, a PDC bit with a Stinger CDE was tested in Zubair field, Iraq. Vibration problems were causing low ROP and inconsistent drilling performance. Formations encountered in the 121/4-in. hole sec-tion included medium-to-hard carbonates and interbedded intervals that had caused stick-slip and lateral vibration problems in offset wells. These conditions forced the operator to slowly control drill through the interval, thus compro-mising ROP.

To solve the problem, a 121/4-in. 6-bladed PDC bit with 16-mm [0.63-in.] cutters was selected, and a CDE was installed at the center of the bit. The modified bit, run on a steerable BHA, drilled 595 m [1,950 ft] of hole from casing shoe to TD in one run, achieving an ROP 29% greater than that of the best offset run of 18.5 m/h [60.7 ft/h] and 56% better than the average ROP of 15.3 m/h [50.2 ft/h] previously attained while drilling three offset wells. The bit also exhibited more stable behavior, drilling with less stick-slip and lower vibration levels than those experienced in offset wells.

Cutter Revolution In many applications, PDC bits have demon-strated significant advantages over roller cone bits for extended footage capabilities and high ROP. However, hard and abrasive formations—which pose some of the toughest conditions that any bit might face—typically cause significant wear on fixed shearing elements. In such envi-ronments, the fixed cutter of the PDC bit can chip, causing drilling efficiency to decline. On a PDC bit, the cutter is fixed in place and most of it is shielded within the body of the blade itself—only a small portion of the diamond table con-tacts the formation (below). As the cutter shears the rock, the exposed portion of the cutter gradu-ally grows dull through abrasive action.

> Stinger CDE. After cutting structures are removed from the bit center (left ), space is created for placement of a CDE (right ). This space also enables development of a small rock column that is easily crushed by the CDE.

> Results of full-scale testing. Inefficient cutters at the center of a standard PDC bit allow development of a central mound of rock (left ). However, because the central cutters have been removed, a small column of stress-relieved rock develops (right ). When a Stinger CDE engages this rock column, it fractures and crushes the rock.

> PDC cutting surface. More than 60% of the circumferential edge of a fixed PDC cutter is retained within the body of the bit and goes unused. The actual amount used will vary with cutter size, back rake and depth of cut.

Bit body

Rock face

PDC cutter

Unused cutter

PDC cutter

Cutting edge

Used Unused

56 Oilfield Review

An FEA evaluation of fixed cutter wear showed that frictional heat concentrates where the cutter edge contacts the rock (above). The combination of high temperature, abrasion and WOB eventually causes the cutting edge to wear flat. The resulting surface, known as a “wear flat,” is subjected to an even higher degree of frictional heat, which induces more wear as the bit contin-ues to drill.

Mechanical loading, in combination with con-centrated heat buildup at the cutting edge, can weaken the diamond bonding and damage the cutter edge. As wear progresses through the syn-thetic diamond table, it eventually reaches the tungsten carbide substrate, causing a noticeable reduction in shearing efficiency and ROP.

Because heat buildup causes accelerated cut-ter wear, and more wear generates even more heat, bit designers sought to break this cycle by keeping the cutting edge both cool and sharp. To manage heating by reducing friction at the cut-ting edge, Smith engineers created the industry’s first cutter that enables the diamond table to fully rotate while drilling. The 360° rotation reduces localized wear by keeping the cutter edge significantly cooler and sharper than fixed cutters. Bit designers mounted the cutter onto a shaft and fitted the assembly into a tungsten car-bide sleeve that enables the cutter to rotate freely.17 The sleeve is installed in a housing that is brazed into the bit blade to secure the cutter while it rotates (below left). The cutter’s orienta-tion in the bit blade, relative to its contact angle with the formation, creates a rotational force that causes the cutter to spin on its shaft as the bit rotates.

The ONYX 360 cutter is designed for high-wear areas on the cutting structure and is mounted only at certain locations along a bit. Using the IDEAS integrated drillbit design plat-form, Smith engineers mapped the areas of high-est wear on a PDC bit and positioned the ONYX 360 rolling cutter there. For example, wear flats often occur on cutters along the shoulder area of a bit, where high velocity at the outer edge of the bit and the relatively large volume of rock removed by these cutters cause accelerated cutter degradation. By placing these low-wear, sharp-edged cutters at high-wear locations, bit designers have seen improved durability and sus-tained ROP across longer intervals. Placement of the ONYX 360 cutters varies for each bit, depend-ing on such factors as bit size, blade count and type of lithology drilled.

The ONYX 360 cutter has been tested on a granite cylinder using a VTL. Vertical, tangential and radial forces were imposed on the cutter and recorded during the test. The test provided the basis for comparing the performance of the ONYX 360 cutter with that of a premium fixed cutter (next page, top left). During the test, the fixed cutter required increased force—from 200 to 1,200 lbf [900 to 5,300 N]—to maintain a con-

stant depth of cut as it wore. The rolling cutter required a relatively low and gradual weight increase, starting at 200 lbf, with gradual increases to 600 lbf [2,670 N], while depth of cut remained constant.18 This test demonstrated that the ONYX 360 cutter required less force to main-tain a consistent depth of cut while removing more rock volume than the fixed cutter removed. Visual inspection of the cutters showed that the edge of the rolling cutter was slightly rounded, while the fixed cutter had a 0.12-in. [3-mm] wear flat.19 Because the rolling cutter remained sharper longer and dissipated heat better than the fixed cutter, it was able to shear more rock with less wear than the fixed cutter could.

Put to the TestDuring the course of drilling an extended lateral section, an operator developing a field in Hemphill County, Texas, had to contend with the highly abrasive Granite Wash formation. This for-mation, a mix of disaggregated granite and its constituent feldspar and quartz grains, is formed from the remnants of intrusive igneous rock that was eroded and deposited downstream from its source. It is hard, abrasive and tough on bits. While drilling the 61/8-in. diameter horizontal sec-tion, the operator was plagued by poor PDC bit performance. Cutter damage and abrasive wear reduced ROP to unacceptable levels and forced the driller to make frequent trips for new bits, thus adversely affecting project economics.

To solve this problem, the operator elected to mount a 61/8-in. PDC bit equipped with seven ONYX 360 rolling cutters as part of a steerable BHA. This bit successfully drilled out the casing shoe then made 1,562 ft [476 m] of horizontal

17. Zhang Y, Burhan Y, Chen C, Tammineni S, Durairajan B, Mathanagopalan S and Ford R: “Fully Rotating PDC Cutter Gaining Momentum: Conquering Frictional Heat in Hard/Abrasive Formations Improves Drilling Efficiency,” paper SPE 166465, presented at the SPE Annual Technical Conference and Exhibition, New Orleans, September 30–October 2, 2013.

18. Zhang et al, reference 17.19. Zhang Y, Baker R, Burhan Y, Shi J, Chen C, Tammineni S,

Durairajan B, Self J and Segal S: “Innovative Rolling PDC Cutter Increases Drilling Efficiency Improving Bit Performance in Challenging Applications,” paper SPE/IADC 163536, presented at the SPE/IADC Drilling Conference and Exhibition, Amsterdam, March 5–7, 2013.

20. Zhang et al, reference 19.21. Bruton G, Smith M, Mueller L and Ford R: “Constructing

Difficult Colony Wash Lateral with Innovative Rolling Cutter Technology Improves Drilling Performance,” paper IADC/SPE 167956, presented at the IADC/SPE Drilling Conference and Exhibition, Fort Worth, Texas, March 4–6, 2014.

22. The Smith Bits DRS drilling records system is an extensive library of bit run information. Initiated in 1985, this database contains more than 3 million records from oil and gas fields around the world.

> Heat-induced cutter degradation. Finite element analysis shows how frictional heat concentrates along the cutter edge, where it contacts the rock; friction and heat aid in creating a wear flat.

Temperature, °C400

200

300

Fixed PDC cutter

Wear flat

> ONYX 360 rolling cutter. The ONYX 360 cutter shaft is fully contained within an integrated housing to ensure continuous rotation and cutter retention during drilling. The bit’s drilling force, coupled with the cutter’s orientation relative to the rock face, causes the cutter to rotate. As a result of rotation, the entire edge of the cutter is used, thus spreading wear more evenly along the cutting edge.

Cutter

Substrate

Rotating shaft

Housing

Summer 2014 57

hole through the Granite Wash at an ROP of 24.79 ft/h [7.6 m/h], or 44% faster than that of the best fixed cutter bit previously used in that field. This bit was just one of more than 70 bits equipped with the rolling cutter to drill the Granite Wash formation. A statistical review of bit performance showed an average of 56% more footage drilled compared with results in 450 off-set runs drilled with fixed cutter bits.20

In another Granite Wash well, Chesapeake Operating Inc. ran PDC bits equipped with ONYX 360 cutters to drill a 61/8-in. hole section.21 A review of the DRS drilling records system data-base shows that of 42 wells drilled within a 2-mi

[3-km] radius of the Chesapeake well, only two had a 61/8-in. lateral section drilled through the Granite Wash formation using only PDC bits.22 The bits used in those two wells employed fixed PDC cutters. A performance analysis showed that bits equipped with the ONYX 360 rolling cutter demonstrated significant increases in durability and ROP compared with the fixed cutter bits. PDC bits with rolling cutters drilled an average 30% more footage than fixed cutter bits in the first well and 75% more than in the second well. The rolling cutter bits drilled the 61/8-in. section with four fewer bit trips, saving five days of rig time through that interval.

Raising the BarPDC bits equipped with Stinger conical diamond elements are helping improve ROP in many of today’s challenging wells worldwide. Bits fitted with ONYX 360 rolling cutters are delivering unmatched performance in abrasive lateral applications such as those in wells from the cen-tral US. Operators who have tried these specially equipped bits are requesting them for use in upcoming wells.

Successes in the field are speeding the evolu-tion of diamond cutting elements for PDC bits. Already, a new generation of PDC bits is being tested with multiple Stinger CDEs, spread across the entire profile of the bit. This second genera-tion of Stinger bits has been subjected to harsh conditions, particularly in hard, interbedded for-mations that can damage the cutting structures of conventional PDC bits. During the first 100 runs, these bits have proved their reliability downhole, drilling more than 90% farther than standard bits and resulting in fewer trips to change bits (below left).

With additional development, innovations in polycrystalline diamond bit technology will con-tinue to change the way the industry drills and expand the range of PDC applications. The result-ing gains in ROP and bit life are expected to fur-ther reduce drilling costs and positively impact financial viability of difficult drilling prospects. —MV

> Results of vertical turret lathe tests. The ONYX 360 rolling cutter (blue) requires substantially less vertical force to drill a longer distance, enumerated as the number of cutting passes, compared with that required for a premium fixed cutter (red).

Aver

age

verti

cal f

orce

, lbf

Number of cutting passes

1,400

1,200

1,000

800

600

400

200

00 100 200 300 400 500 600 700

Premium fixed cuttersONYX 360 rolling cutters

> Comparison of drilling efficiency relative to the conventional PDC bit performance. The second generation of Stinger bits showed marked improvements in footage drilled and ROP.

Perc

ent

Footage drilled Average rate of penetration

200

150

100

50

0

Conventional PDC bitSecond-generation Stinger bit

Farrukh Akram is a Schlumberger Heavy Oil Product Champion and a Technical Advisor and Product Manager for unconventional and heavy oil software; he is based in Abingdon, England. He has more than 10 years of petroleum industry engineer-ing experience, primarily in reservoir characteriza-tion, field development planning and production optimization. He started his career with Shell Canada as a reservoir engineer focusing on heavy oil and low-permeability shale rock before joining Schlumberger in 2006. He has authored multiple technical papers and published articles on heavy oil reservoir characterization and recovery optimiza-tion. He was previously a director of the SPE Calgary Section. Farrukh earned an MS degree in petroleum engineering from Dalhousie University, Halifax, Nova Scotia, Canada, and a BS degree in mechanical engineering from NED University of Engineering and Technology, Karachi, Pakistan.

William J. Bailey is a Principal Scientist at the Schlumberger-Doll Research Center in Cambridge, Massachusetts, USA. He has 25 years of industry experi-ence in reservoir engineering, modeling of unconven-tional assets, multiphase flow in conduits, equipment failure analysis and computationally extensive optimiza-tion problems, including coupled full-field assets. He has published 50 articles, holds eight patents and has contributed to five books. He was chair of the SPE New York and New England Section and review chair for the SPE Production & Operations journal and received the inaugural SPE A Peer Apart Award. He reviews for various SPE journals and serves on the Journal of Petroleum Technology editorial and SPE books commit-tees. Bill has an MS degree (Hons) from Imperial College London and a PhD degree from the Norwegian Technical University in Trondheim, both in petroleum engineering. He also has an MBA degree from the University of Warwick, Coventry, England.

Greg Bruton is the Manager of Drilling Technology for Chesapeake Operating, Inc. in Oklahoma City, Oklahoma, USA. He joined the company in 2006 as a senior drilling engineer following a 25-year career with Gulf Oil and Chevron. Greg’s drilling assignments with Chevron took him to several locations around the world. He received a BS degree in chemical engineering from New Mexico State University, Las Cruces, USA.

Gabriele Busanello is a Senior Geophysicist in the Schlumberger PetroTechnical Services Geosolutions Center in Abu Dhabi, UAE. Before joining Schlumberger in 2007, he had experience in the field and with onboard seismic processing. He began his Schlumberger career with the data processing team dedicated to multicomponent streamer development in Gatwick, England. Since then, he has been assigned to the marine division first commercial DISCover* project in Mumbai, to the ocean-bottom cable and transition zone teams in Jakarta, and then to the UniQ* land seismic team in Perth, Australia. Gabriele earned a master’s degree in geology and applied geo-physics from Università degli Studi di Trieste, Italy.

Zhifeng Chen is Account Manager for WesternGeco for the China Area. Based in Beijing, he is responsible for marine survey bidding and UniQ system sales. Prior to his current position, which began in 2002, he worked for 10 years mainly in the Schlumberger Wireline Segment. Zhifeng holds a master’s degree in geophys-ics from the China University of Petroleum, Huadong, People’s Republic of China.

Ron Crockett is the Diamond Research and Development Manager for Novatek in Provo, Utah, USA. Prior to joining Novatek, he worked for Smith International and ReedHycalog. Ron has held a variety of positions in his 27 years in the diamond industry, including product engineer and production manager.

Dave DenBoer based in Provo, Utah, is a Research Engineer with MegaDiamond, a Schlumberger company. He has 30 years of experience in the industry, working in configuration management and product engineering. Dave obtained his BS degree in design engineering from Brigham Young University, Provo.

Mark Egan is Chief Area Geophysicist for Geosolutions, part of the PetroTechnical Services Segment of Schlumberger. He has nearly 40 years of experience in seismic acquisition and data processing techniques working in the US, UK and Middle East. He is the author of 40 conference papers and technical articles, which focus mainly on innovative seismic imaging techniques to meet specific exploration and development challenges. Mark, who is based in Houston, is a member of the European Association of Geoscientists and Engineers, the SEG and the SPE. He holds a PhD degree in geophysics from the University of Houston.

Nobuhisa Eguchi is the Manager of Science Operations at the Center for Deep Earth Exploration (CDEX) with the Japan Agency for Marine-Earth Science and Technology in Yokohama, Japan. He man-ages science support personnel such as lab techni-cians and curators and processes such as preexpedition site surveys of drilling sites and logging services for scientific operations onboard the drilling vessel Chikyu. On occasion, he also serves as the Chikyu Expedition Project Manager. Before starting his career with CDEX, he was a science coordinator at IODP Management International Inc., the central management organization of the Integrated Ocean Drilling Program (now the International Ocean Discovery Program). Nobuhisa has a PhD degree in marine geology from the University of Tokyo.

Craig Fleming is Senior Marketing Services Manager for Smith Bits, a Schlumberger company, in Houston. For the past seven years, he has served as the compa-ny’s technical editor and marketing writer. His respon-sibilities include managing the division’s publishing effort producing articles for presentation at industry meetings and for inclusion in trade journals. He also serves as a marketing analyst responsible for tracking industry trends. Prior to his current position, he was employed as a technical writer in the marketing com-munications department of a service company and as staff writer for IHS in Houston. He also worked for five

years as a petroleum geologist generating drilling pros-pects for several independent oil companies. Craig, who has been an active member in the SPE for 17 years, received a BS degree in geology from Oklahoma State University, Stillwater.

Euan Forbes, based in Calgary, is Segment Product Champion for Pathfinder, a Schlumberger company. He has more than 25 years of experience in the petro-leum industry, primarily in surveying, MWD, LWD and passive ranging. Euan earned a BS degree (Hons) in petroleum geology from the University of Aberdeen.

Robert Ford is the Product Champion for the ONYX 360* rolling cutter for Smith Bits. He started his oilfield career with Halliburton in 1990 as a surface data logger in the North Sea oil fields. After joining Smith International in 1991, he has worked in various engineering and management positions focused on roller cone and PDC bits. Robert holds a BSc degree in geology from the University of Sheffield, South Yorkshire, England.

Michael A. Freeman is a Scientific Advisor, Drilling Fluid Chemistry at M-I SWACO, a Schlumberger com-pany, in Houston; he has held the position since 2007. He joined M-I Drilling Fluids in 1993 as a fluids research chemist and has since led numerous R&D efforts in brine technology, corrosion inhibitors and other specialized dense brines and drilling fluid prod-ucts. At SWACO, he was team leader of waste manage-ment chemistry. He established the M-I SWACO completion fluids laboratory and served as M-I SWACO senior technical advisor from 2003 to 2008. Mike began his oil industry career as a research chemist with Exxon R&D Laboratories in 1980 in Baton Rouge, Louisiana, USA. He has a BA degree in chemistry from Wabash College, Crawfordsville, Indiana, USA, and a PhD degree in analytical chemistry from the University of North Carolina at Chapel Hill, USA.

Masafumi Fukuhara is Physics Metier Manager and Scientific Advisor at the Schlumberger Kabushiki Kaisha Center (SKK) in Sagamahira, Japan. He began his career with Schlumberger in 1984 at SKK working on a fabrication process for pressure sensors; a year later, he moved to the Schlumberger-Doll Research Center in Ridgefield, Connecticut, USA, where he spent the next three years as an experimental physi-cist working on nuclear magnetic resonance. He returned to SKK in 1988 and held various positions, including project manager for acoustic logging tools and methane hydrate R&D program manager. From 2009 to 2012, he was program manager for borehole seismic and acoustic imaging at the Schlumberger Moscow Research Center before returning to SKK. Masafumi has served as a member of the engineering development panel of the Integrated Ocean Drilling Program (IODP) Science Advisory Structure and a member of the engineering task force of IODP Management International, Inc. He earned an MS degree in physics from Tokyo Metropolitan University and a postgraduate diploma in physics from the University of Tokyo.

Contributors

58 Oilfield Review

Gary Garcia works in marketing and is the New Technology Trainer for Smith Bits in Houston. Previously, he was product champion and project man-ager for the central Stinger* conical diamond element. Gary has more than 30 years of experience in the industry; he joined Smith Bits in 1991, starting in design engineering. Gary has also served as director of career development and training for Smith Bits and has held positions in manufacturing engineering, design engineering and sales. He received a BS degree in technology from the University of Houston.

Thomas Heesom is Sales and Marketing Manager for WesternGeco UniQ System Sales in Dubai. He has had positions in the land seismic industry since 2000, when he joined WesternGeco as a field geophysicist working on land seismic field crews in the Middle East and North Africa. After six years of field positions, he served as training instructor, product champion and marketing analyst. Thomas obtained a master’s degree in geophysics from the University of Leeds, England.

Koji Kusaka, based in Tokyo, is the Schlumberger Oilfield Services Manager for the Japan, Korea and Taiwan Area. After joining Schlumberger in 1983, he worked as a wireline logging engineer for seven years in Libya, Italy, Taiwan and Malaysia. He has also held marketing and management positions in France and Japan. From 2002 to 2005, he was seconded as a proj-ect consultant to the Japan Oil, Gas and Metals National Corporation methane hydrate research proj-ect. Koji holds a BS degree in geology from Hiroshima University, Japan, and a diploma in reservoir manage-ment jointly awarded by Institut Français du Pétrole, Rueil-Malmaison, France; Delft Technical University, the Netherlands; and Imperial College London.

David H.-S. Law is the Heavy Oil Technical Director for Schlumberger in Edmonton, Alberta, Canada. Before joining Schlumberger in 2005, David was the thermal gravity strategic area coleader and the reser-voir simulation group leader for the Alberta Research Council, now Alberta Innovates. He has more than 25 years of experience in technology development in the areas of heavy oil–bitumen recovery, thermal reser-voir simulation and greenhouse gas storage. David holds two patents and has authored more than 100 technical publications. He received a BS degree from the National Taiwan University and MS and PhD degrees from the University of Alberta, all in chemi-cal engineering. David is registered with the Association of Professional Engineers and Geoscientists of Alberta and is a member of the SPE Canada. He has served on various technical program committees for SPE heavy oil applied technology workshops, forums and conferences, Canadian International Petroleum conferences and World Heavy Oil Congresses. He was a 2010/2011 SPE Distinguished Lecturer.

Xue Lei is a Senior Geophysicist for Geosolutions, part of the PetroTechnical Services Segment of Schlumberger and the Technical Leader for geology and seismic interpretation and inversion in Beijing. Before joining Schlumberger in 2007, Xue was a geo-physicist at the China National Petroleum Corporation (CNPC) Sichuan Geophysical Company (SCGC) for 17 years. She has more than 24 years of experience, including 80 projects that encompassed seismic acqui-sition, data processing and seismic inversion and

interpretation. She served as the director of data pro-cessing and interpretation and then senior geophysi-cist for CNPC SCGC. Xue, who has authored or coauthored 14 publications, has a BS degree in geo-physics from the Petroleum University of China, Beijing, and an MS degree in geophysics with geologi-cal engineering from Chengdu University of Technology, People’s Republic of China.

Rong Li is an Area Geophysicist for Geosolutions, part of the PetroTechnical Services Segment of Schlumberger; she is based in Beijing. She has more than ten years of experience working with land and marine seismic data processing in the principal basin areas of China. Rong obtained BS and MS degrees in geophysics from Peking University, Beijing.

Bo Liang is Vice President for the China National Petroleum Corporation Sichuan Geophysical Company (SCGC) in Chengdu, Sichuan, People’s Republic of China. He is responsible for the SCGC international market and the company’s development strategy. Before this assignment, which he began in 1992, he worked in various positions focused on geophysics projects. Bo earned a bachelor’s degree in geophysics from the Southwest Petroleum University, Nanchong (now in Chengdu).

Jeff Lund, based in Provo, Utah, is a Manager of the PDC cutter engineering group for MegaDiamond. He has worked for more than 30 years in the oil and gas industry, with more than 20 years devoted to drill bits and cutters. Jeff, who holds numerous patents related to drilling, holds a BS degree in mechanical engineer-ing from the University of Utah, Salt Lake City.

Heloise Bloxsom Lynn is the President of Lynn Incorporated. She has 40 years of experience in the oil and gas industry. She began her career with Texaco in Houston and worked four years for Amoco-British Petroleum in Houston performing seismic acquisition, processing and interpretation. Heloise cofounded Lynn Incorporated in 1984, and since then her consulting work has focused on using azimuthal anisotropy for the characterization of naturally fractured reservoirs and the in situ horizontal stress field. She was a 2004 SEG/AAPG Distinguished Lecturer on the topic “The Winds of Change—Anisotropic Rocks, Their Preferred Direction of Fluid Flow and Their Associated Seismic Signatures.” Heloise earned a BA degree in geology and math from Bowdoin College, Brunswick, Maine, USA, and an MS degree in exploration geophysics and PhD degree in geophysics, both from Stanford University, California, USA.

Alberto Malinverno is Senior Research Scientist at the Lamont-Doherty Earth Observatory of Columbia University in Palisades, New York, USA, where he has worked since 2005. His interests include microbial methanogenesis, gas hydrates, carbon cycling in conti-nental margin sediments, geophysical inverse prob-lems and the Neogene tectonic evolution of the Mediterranean region. He has sailed on four Integrated Ocean Drilling Program expeditions, focus-ing on downhole geophysical measurements. Previously, he was a research scientist and manager at the Schlumberger-Doll Research Center, Ridgefield, Connecticut. Before joining Schlumberger in 1992, Alberto spent three years as a postdoctoral and an associate research scientist at the Lamont-Doherty Earth Observatory. He obtained an undergraduate

degree in geological sciences from Università degli Studi di Milano, Italy, and MS and PhD degrees, also in geological sciences, from Columbia University, New York City.

Kyaw Moe is the Deputy Director of the Research and Development Center for Ocean Drilling Science at the Japan Agency for Marine-Earth Science and Technology (JAMSTEC), based in Yokohama, Japan. He leads ultradeepwater drilling research using the JAMSTEC drillship Chikyu in partnership with indus-try. He has supervised various science services at the Center for Deep Earth Exploration (CDEX) since the-Chikyu began operations in 2005. Previously, he served as logging staff scientist and expedition project man-ager for nine riser and riserless expeditions onboard the research vessel Joides Resolution and the Chikyu.His research interests focus on data integration, and he has been involved in the Nankai Trough Seismogenic Zone Experiment (NanTroSEIZE), Costa Rica Seismogenesis Project (CRISP) and the Japan Trench Fast Drilling Project (JFAST) projects at all stages, from preexpedition site survey to data acquisi-tion during expeditions and postexpedition data analy-ses. Before his graduate studies at Ocean Research Institute, University of Tokyo, Moe taught at the University of Yangon, Myanmar, and worked in mud logging in Southeast Asia offshore explorations.

Anastasia Poole is a Staff Geophysicist for WesternGeco; she is based in Gatwick, England. Anastasia has 10 years of experience in land seismic acquisition and data processing methods and has worked in Egypt, UAE, Oman, Yemen and Australia. In her current position, she tests new data management and reservoir characterization algorithms and work-flows specifically for the UniQ land seismic system. Anastasia received BS and MS degrees in geology from the Lomonosov Moscow State University, Russia.

Terry Stone is a Schlumberger Advisor based in the Abingdon Technology Center, England, where he develops ECLIPSE* and INTERSECT* simulators. Prior to joining Schlumberger in 1995, Terry worked as a group leader of a mathematical modeling group for the Alberta Research Council, Canada, a senior research engineer for Mobil Research and Development Corporation and an E&P standalone products manager for Scientific Software-Intercomp and Intera. He holds a BS degree in mathematics from the University of Windsor, Ontario, Canada, and MS and PhD degrees in nuclear engineering from McMaster University, Hamilton, Ontario.

Malcolm Taylor is the Business Development Manager for Novatek in Provo, Utah. His primary responsibility is managing an oilfield services project portfolio; he also has global responsibility for the technical and manufacturing strategy for drill bits and a PDC drillbit product line, including manufacture of PDC cutters. Prior to his current position, Malcolm was vice presi-dent of manufacturing for ReedHycalog. He also worked in marketing and product development for Schlumberger. Most of his 30-year career has been involved with drill bits, though he has also worked on other drilling equipment and automotive engines. He holds several patents and is a Fellow of the Institute of Mechanical Engineers and a Chartered Engineer. He is a graduate of the University of Cambridge.

Summer 2014 59

development. He has designed sand control and ICD completions for locations worldwide, both for conven-tional and thermal wells. Before joining ACT in 2007, he consulted for a variety of companies throughout the world, primarily with Adams Pearson Associates. Glenn received a BSc degree in engineering science, with a nuclear engineering option, from the University of Toronto, Ontario, and an MSc degree in mechanical engineering from the University of Calgary.

Fusen Xiao is Deputy Manager for PetroChina Southwest Oil and Gas Company Research Institute in Chengdu, Sichuan, People’s Republic of China. He has worked at the institute for 30 years and is responsible for guiding geophysics research. Before this assign-ment, he worked for Sichuan Geophysical Company on a variety of geophysics projects. Fusen earned a bache-lor’s degree in geophysics from Southwest Petroleum University, Chengdu, and a doctorate degree from Chengdu University of Technology.

K.C. Yeung is Director, Technology Development and Innovation at Brion Energy in Calgary, where he evalu-ates and facilitates new technologies related to oil sands and other developments. He joined Brion Energy in 2012 as director, oil sands technology. Prior to his current position, he held management and supervisory positions with Husky Energy and Suncor Energy. He began his oil industry career in 1977 as a reservoir engineer in Calgary with Texaco Canada. K.C. has a BS degree (with distinction) and an MS degree, both in mechanical engineering, from the University of Hawaii, Honolulu, USA.

joined Delft Geophysical, which was later acquired by Schlumberger. After five years in the field working on various land and transition zone (TZ) seismic crews in Europe, North America and South America, he worked as operations geophysicist with the WesternGeco, then known as Geco-Prakla, TZ team, involved primarily in bidding and survey design for both cable and nodal recording systems. Following a position as crew supervisor in Australia, he was a member of the Asia Area survey design team based in both Perth and Kuala Lumpur; he then moved to Dubai in 2003 as division and area geophysicist for the WesternGeco land and TZ acquisition group. Peter, who is a Schlumberger Eureka Technical Career Principal, has an MSc degree in applied geo-physics from the mining engineering faculty of the Delft University of Technology, the Netherlands.

Allen White is a Product Champion for novel cutting structures for Smith Bits in Houston. During his nine years in the industry, he has also worked as a product champion for PDC cutters, as an engineering manager for design engineering and as a design engineer. Allen obtained bachelor’s and master of science degrees in mechanical engineering from Texas Tech University, Lubbock, USA.

Glenn Woiceshyn is Vice President of Engineering at Absolute Completion Technologies (ACT) in Calgary, where he specializes in completions design for sand control and flow control. He has more than 23 years of engineering experience in the petroleum industry, pri-marily in production, completions and research and

Harold Tobin, Professor of Geoscience at the University of Wisconsin (UW)–Madison, USA, special-izes in the physical properties and geology of fault zones. A veteran of numerous scientific ocean drilling expeditions, he has conducted research on tectonic plate boundaries offshore Japan, Costa Rica, the Pacific Northwest of the US, the Pacific coast of Canada and in the Caribbean, as well as on land in California and New Zealand. He is the chief project scientist of the NanTroSEIZE Project for the International Ocean Discovery Program, and he oper-ates the Halliburton Geoscience Visualization Center at UW–Madison, where his research group analyzes 3D seismic reflection imaging, well logs and core petro-physical data. He earned a BS degree in geology and geophysics from Yale University, New Haven, Connecticut, and a PhD degree from the University of California, Santa Cruz. Following his studies, Harold was a postdoctoral researcher with the Stanford University Rock and Borehole Geophysics program, California, then held a faculty position for eight years at New Mexico Institute of Technology and Mining, Socorro, before joining the UW–Madison faculty.

Peter van Baaren, based in Sneek, the Netherlands, is UniQ System Sales Geophysics Manager with WesternGeco; he joined the group in 2013. He is responsible for geophysical support for the UniQ System Sales organization; such support includes sur-vey design, infield quality control and UniQ specific data processing. Peter, who has more than 25 years of experience in the industry, began his career when he

60 Oilfield Review

An asterisk (*) denotes a mark of Schlumberger.

Lost Circulation Solutions. Since the beginning of the petroleum industry, lost circulation has presented challenges to operators. Today, as drilling operations are becoming increasingly challenging, and operators are drilling deeper and into depleted reservoirs, lost circulation situations are becoming more complex. Knowledge and understanding of the respective reser-voirs are key to identifying the type and cause of lost circulation. Actively acquiring this information facili-tates selection of the appropriate lost circulation solu-tions. This article describes lost circulation solutions and case histories of their successful deployment.

Coming in Oilfield Review

Boosting Production at Shushufindi. The mature Shushufindi-Aguarico oil field in northeast Ecuador was discovered in 1969. The giant field encompasses an area of 400 km2 [150 mi2]. In early 2012, Ecuador’s state-run oil company Petroecuador signed a 15-year contract with Consorcio Shushufindi S.A., a consor-tium led by Schlumberger to manage and increase production, discover new reserves and evaluate secondary and tertiary recovery opportunities in the field. This article will describe how production is being revitalized in the Shushufindi-Aguarico oil field.

Formation Testing. Well test results weigh heavily in operator decisions for asset development. Using bidirectional, real-time communication and control, a new compact formation test system is helping opera-tors make choices based on high-quality data that were not previously available using traditional forma-tion test technology.

Perforating Update. Engineers at the Schlumberger Reservoir Completions Center tested hundreds of perforation charges at simulated downhole condi-tions and discovered that the results of standard API tests are not entirely representative of downhole performance. The engineers incorporated these results into a stressed-rock perforating analysis program that simulates charge performance and perforation effectiveness. Ongoing shaped-charge research has also led to the development of charges engineered for specific rock types because charges designed for conventional reservoir rocks do not per-form the same in shale and coal.

Summer 2014 61

BOOKS OF NOTE

PaleoclimateMichael L. BenderPrinceton University Press41 William StreetPrinceton, New Jersey 08540 USA2013. 320 pages. US$ 80.00ISBN: 978-0-691-14554-9

The author, a professor of geosciences and atmospheric and ocean sciences, describes the major periods in Earth history, the physical controls on climate and how the record of past climate is determined. He explores the history of climate change over millions of years and describes the Holocene—the past 10,000 years—including manmade climate change in the context of the paleoclimate.

Contents:

• Earth’s Climate System

• The Faint Young Sun

• Precambrian Glaciations

• Regulation of the Earth System and Global Temperature

• The Late Paleozoic Ice Ages

• Equable Climates of the Mesozoic and Paleogene

• The Paleocene-Eocene Thermal Maximum

• The Long Cooling of the Cenozoic

• The Origin of Northern Hemisphere Glaciation and the Pleistocene Ice Ages

• Rapid Climate Change During the Last Glacial Period

• The Holocene

• Anthropogenic Global Warming in the Context of Paleoclimate

• Glossary, Index

The author has succeeded admira-bly in producing a clear, concise, yet detailed summary of a very important topic. The text is supplemented by an excellent selection of diagrams and data displays, five “boxes” with detailed discussion of key technical issues, and more than 300 references to the primary research literature. I found it easy to read yet thought provoking, consistently interesting

and, perhaps best of all, not at all intimidating in bulk or style. Highly recommended!

Green W: “Reviews,” The Leading Edge 33, no. 5

(May 2014): 566.

Oxygen: A Four Billion Year HistoryDonald E. CanfieldPrinceton University Press41 William StreetPrinceton, New Jersey 08540 USA2014. 224 pages. US$ 27.95ISBN: 978-0-691-14502-0

Donald Canfield, a professor of ecology, explores the relationship of the history of atmospheric oxygen to the evolution of life and the changing chemistry of the Earth. The author draws from fields such as geology, paleontology, geochem-istry, biochemistry, animal physiology and microbiology to explain why an oxygenated Earth became an ideal place for life.

Contents:

• What Is It About Planet Earth?

• Life Before Oxygen

• Evolution of Oxygenic Photosynthesis

• Cyanobacteria: The Great Liberators

• What Controls Atmospheric Oxygen Concentrations?

• The Early History of Atmospheric Oxygen: Biological Evidence

• The Early History of Atmospheric Oxygen: Geological Evidence

• The Great Oxidation

• Earth’s Middle Ages: What Came After the GOE

• Neoproterozoic Oxygen and the Rise of Animals

• Phanerozoic Oxygen

• Epilogue

• Notes, References, Index

Concise and easily read, Oxygen provides an ideal starting block for those interested in learning about Earth’s O2 history and, more broadly,

the function and history of biogeo-chemical cycles. . . . The endnotes provide valuable entries for readers who wish to explore particular points in greater depth. . . . And the detailed bibliography captures a vast swath of the relevant primary literature. I highly recommend Canfield’s book for anyone with even a remote interest in Earth history, as O2 singularly encompasses much of what makes our planet special.

Fischer WW: “Breathing Life into Oxygen,”

Science 343, no. 6173 (February 21, 2014): 840.

This is the sort of science writing we would all do well to read more of. . . . Engage[s] with the ambiguity of a world where evidence is imper-fect, knowledge evolves, and mistakes can be made in interpreting the data.

Scheffler I: “Perfectly Natural: Science and

Sustainability,” Los Angeles Review of Books

(February 22, 2014), https://lareviewofbooks.

org/review/sustainable-planet (accessed

March 12, 2014).

. . . his excellent descriptions of the scientific process show how competing hypotheses, and the scientists who present them, vie for supremacy. Canfield also offers a philosophical perspective: scientific understanding provides true insight into the structure of the natural world. . . .

“Book Review,” Publishers Weekly (October 14,

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

14502-0 (accessed March 12, 2014).

Radical Abundance: How a Revolution in Nanotechnology Will Change CivilizationK. Eric DrexlerPublic Affairs, a member ofThe Perseus Books Group250 West 57th Street, 15th FloorNew York, New York 10107 USA2013. 368 pages. US$ 28.99ISBN: 978-1-610-39113-9

K. Eric Drexler—the founding father of nanotechnology—discusses how rapid scientific progress may change our world. Because nanotechnology allows

atomically precise manufacturing, the author posits that we will soon have the power to produce radically more of what people want and at a lower cost. This paradigm shift, he suggests, will drastically alter the foundation of our economy and environment.

Contents:

• An Unexpected World: Atoms, Bits, and Radical Abundance; An Early Journey of Ideas; From Molecules to Nanosystems

• The Revolution in Context: Three Revolutions, and a Fourth; The Look and Feel of the Nanoscale World; The Ways We Make Things

• Exploring Deep Technology: Science and the Timeless Landscape of Technology; The Clashing Concerns of Engineering and Science; Exploring the Potential of Technology

• The Technology of Radical Abundance: The Machinery of Radical Abundance; The Products of Radical Abundance

• The Trajectory of Technology: Today’s Technologies of Atomic Precision; A Funny Thing Happened on the Way to the Future . . . ; How to Accelerate Progress

• Bending the Arc of the Future: Transforming the Material Basis of Civilization; Managing a Catastrophic Success; Security for an Unconventional Future; Changing Our Conversation About the Future

• Appendix I: The Molecular-Level Physical Principles of Atomically Precise Manufacturing

• Appendix II: Incremental Paths to APM

• Notes, Index

The technical and political chal-lenges of unleashing ‘atomically precise manufacturing’ are substan-tial, but Drexler cuts deftly through the complexities.

Kiser B: “Books in Brief,” Nature 497, no. 7449

(May 16, 2013): 315.

As a primer into the science and engineering behind APM [atomically precise manufacturing] and ‘nanosci-ence,’ Drexler offers an engaging way to enter that conversation.

“Book Review,” Publishers Weekly (July 8, 2013),

http://www.publishersweekly.com/978-1-61039-

113-9 (accessed February 28, 2014).

Oilfield Review62

• Conifers (Phylum Coniferophyta)

• Non-Eudicot Flowering Plants (Phylum Angiospermophyta; Subclasses Magnoliids, Monocotyledons, and Ceratophyliids)

• Eudicot Flowering Plants (Phylum Angiospermophyta; Subclass Tricolpates)

• Trace Fossils

• Reading the Pages of Deep History

• Concluding Remarks

• Postscript

• Appendix A: Key to the Major FBM Localities

• Appendix B: Summary List of FBM “Fish” Species

• Appendix C: Summary List of FBM Bird Species

• Appendix D: FBM Fossils That Have Been Enhanced, Restored, Inset, or Faked

• Appendix E: Using This Book and Comments on Bulletin 63

• Appendix F: Sources of Phylogenies Used in This Book

The Lost World of Fossil Lake is a splendidly illustrated compendium on these c.52 million-year-old fossils, written with grace and authority. . . . Grande’s enthusiasm shines through, as does his wish to communicate his great passion for fossils in general and this Eocene lake in particular. He has set out his intention to inspire generations of students for palaeon-tology, and he has accomplished it very well.

Clarkson E: “Book Review,” Times Higher

Education (May 16, 2013), http://www.timeshigh-

ereducation.co.uk/books/the-lost-world-of-fossil-

lake-snapshots-from-deep-time-by-lance-

grande/2003788.article (accessed April 22, 2014).

. . . lush, in-depth guide to the area’s fossilized fauna. . . . Beautifully illustrated in colour, including a field guide and atlas.

“Books in Brief,” Nature 498, no. 7455

(June 27, 2013): 431.

On the Frontier of Science: An American Rhetoric of Exploration and ExploitationLeah CeccarelliMichigan State University PressSuite 25, Manly Miles Building1405 South Harrison RoadEast Lansing, Michigan 48823 USA2013. 250 pages. US$ 44.95ISBN: 978-1-611-86100-6

Author Leah Ceccarelli argues that the metaphor “the frontier of science,” which has become ubiquitous in American rhetoric, guides scientific research in particular ways, sometimes blocking scientists from attaining the very goals they set out to achieve. Ceccarelli, who studied public address by scientists and politicians and the reception of their audiences, explores what happens when this metaphor is deployed, its effects on those who use it and what rhetorical devices are used by those who try to counter its appeal.

Contents:

• History of the Frontier of Science Metaphor

• The Frontier Metaphor in Public Speeches by American Scientists

• The Dangers of Bioprospecting on the Frontier: The Rhetoric of Edward O. Wilson’s Biodiversity Appeals

• Biocolonialism and Human Genomics Research: The Frontier Mapping Expedition of Francis Collins

• Reframing the Frontier of Science: George W. Bush’s Stem Cell Rhetoric

• Conclusion

• Notes, Bibliography, Index

Ceccarelli . . . explores how the metaphor of the frontier (with its close cousin, the pioneer) has become rooted in American scientific rhetoric. . . . One of the book’s most powerful lessons is the possibility that the frontier metaphor creates a culture of science that ‘sees science as a contest for territory’ rather than a global and collaborative effort. . . . While Ceccarelli limits the scope of her inquiry to the biological sciences, On the Frontiers of Science prompts questions of broader relevance.

Newell CL: “The Significance of ‘Frontier’,”

Science 343, no. 6173 (February 21, 2014): 841.

Novel Science: Fiction and the Invention of Nineteenth Century GeologyAdelene BucklandUniversity of Chicago Press1427 East 60th StreetChicago, Illinois 60637 USA2013. 400 pages. US$ 45.00ISBN: 978-0-226-07968-4

The birth of the science of geology and its intersection with literature are the focus of this book. Buckland argues that the scientists of the time were also literary men, and it was their ability to describe their new science in literary terms that helped the public learn about and begin to understand the geologic history of Earth.

Contents:

• Stories in Science: Fictions of a Former World; The Story Undone; Lyell’s Mock Epic; Maps and Legends

• Science in Stories: Kingsley’s Cataclysmic Method; Eliot’s Whispering Stones; Dickens and the Geological City

• Conclusion: Losing the Plot

• Appendix: “Lines on Staffa,” by Charles Lyell

• Notes, Bibliography, Index

Buckland tries to get inside the heads of the Britons who were writing into existence a scientific geology while developing a great literary form: the nineteenth century novel. She succeeds triumphantly. . . . Buckland’s book is the story of how [novelists and geologists] helped one another to write the past into exis-tence. . . . Buckland will send you scouring the second-hand bookshops for long-forgotten works.

Nield T: “Written in Stone,” Nature 496, no. 7446

(April 25, 2013): 428–429.

Although I found it more sharply focused in some parts than others . . . this fascinating book introduced me to perspectives that neither I nor most geologists have ever really thought about. But as a result of reading this mind-expanding book, I am thinking about them now.

Clarkson E: “Book Review,” Times Higher

Education (July 25, 2013), http://www.timeshigh-

ereducation.co.uk/books/novel-science-fiction-

and-the-invention-of-nineteenth-century-geology-

by-adelene-buckland/2005868.article (accessed

April 22, 2014).

The Lost World of Fossil Lake: Snapshots from Deep TimeLance GrandeUniversity of Chicago Press1427 East 60th StreetChicago, Illinois 60637 USA2013. 432 pages. US$ 45.00ISBN: 978-0-222-692296-6

The flora and fauna of the Early Eocene are well preserved in a fossil record from a lake that lasted about two million years and is now the dry lands of southwest Wyoming, USA. The author covers the formation’s origin and geologic history, describes the work of early collectors and explains the clas-sification of fossils. He goes on to offer a comprehensive and detailed tribute to the biodiversity of life that thrived during that era, from bacteria to birds and from fishes to flowering plants.

Contents:

• In the Beginning

• Fossils from the FBM: History, Controversy, and Quarry Life

• Exposing the Record of Past Life: Fossil Preparation

• Classification of Fossils and Their Place in the Web of Life

• Bacteria

• Arthropods (Phylum Arthropoda)

• Mollusks (Phylum Mollusca)

• Vertebrates (Phylum Chordata, Subphylum Vertebrata)

• Cartilaginous Fishes (Superclass Chondrichthii)

• Ray-Finned Fishes (Superclass Actinopterygii)

• Abundance and Distribution of Fish Species

• Tetrapods (Superclass Sarcopterygii)

• Amphibians (Class Amphibia)

• Non-Avian Reptiles (Class Reptilia; Superorders †Paracryptodira, Cryptodira, Squamata, and Crocodylomorpha)

• Birds (Class Reptilia; Superorder Aves)

• Mammals (Class Mammalia)

• Plants

• Green Algae (Phylum Chlorophyceae)

• Ferns and Horsetails (Phylum Filicopsida and Phylum Equisetopsida)

Drilling or workover rigs, iconic symbols of the oil field, are not alwaysrequired for drilling, completions or maintenance operations. Increasingly, the coiled tubing unit is used for many well intervention operations and certain drilling applications. Coiled tubing (CT) refers to a continuous length of small-diameter steel pipe and related surface equipment as well as associated drilling, completion and workover, or remediation, tech-niques. Coiled tubing oilfield technology was initially developed for working on live, producing wells. More recently, this technology has gained wider acceptance among operators for an expanding range of workover and drill-ing applications and for its ability to reduce overall costs. The trend toward extended-reach wells favors CT for its capability to drill or to convey tools and equipment in high angle wellbores.

At the center of any CT surface operation is a coiled tubing unit (CTU), the most prominent feature being a reel from which a continuous length of flexible steel pipe is spooled. To deploy tubing downhole, the CT operator spools the tubing off the reel and leads it through a gooseneck, which directs the CT downward to an injector head, where it is straightened just before it enters the borehole. At the end of the operation, the flexible tubing is pulled out of the well and spooled back onto the reel. On the hub of the storage reel, a high-pressure swivel joint enables pumping of fluids through the tubing while the reel rotates to spool pipe on or off the reel.

From the CTU control cabin, the CT operator controls the hydraulically driven injector head to regulate the movement and depth of the CT string. A stripper assembly beneath the injector head provides a dynamic seal around the tubing string, which is essential for running the CT in and out of live wells. A blowout preventer assembly between the stripper and wellhead supplies secondary and contingency pressure-control functions. The entire process is monitored and coordinated from the CTU control cabin.

Coiled tubing is available in diameters of 0.75 to 4.5 in.—2 in. is the most common size. It may range in length from 2,000 to more than 30,000 ft [600 to 9,000 m]. The tubing is coiled in a single continuous length, thus preclud-ing any need for making or breaking connections between joints. This per-mits continuous circulation while running in or out of the hole.

A Wide Range of ApplicationsCoiled tubing technology is frequently used to deploy tools and materials through production tubing or casing while remedial work is performed on producing wells. Coiled tubing fulfills three key requirements for downhole operations on live wells by providing a dynamic seal between the formation

pressure and the surface, a continuous conduit for fluid conveyance and a method for running this conduit in and out of a pressurized well.

Coiled tubing strength and rigidity, combined with its capability to cir-culate treatment fluids, offer distinct advantages over wireline techniques in workover operations. In addition to drilling and completion operations, oil and gas companies are using CT to help fish for lost equipment and for conveying well logging tools. It has been used to push or pull equipment through highly deviated or horizontal wellbores and past restrictions or to push obstructions beyond a zone of interest. Well logging is typically per-formed with tools that store data in memory; however, some logging opera-tions use an optional cable to provide surface power and readouts when running tools downhole on CT. Operators also employ coiled tubing to con-vey and place bridge plugs and mechanical, hydraulic or inflatable packers to establish zonal isolation.

One of the most common applications for CT is the cleanout and removal of fill materials that restrict flow through tubing or casing (below). Fill material can impede production by blocking the flow of oil or gas. It also may prevent the opening or closing of downhole control devices such as sleeves and valves. Common sources of fill are sand or fine material pro-duced from the reservoir, proppant materials used during hydraulic fractur-ing operations, debris from workovers and organic scale. Fill removal typically involves circulating a cleanout fluid, such as water or brine, through a jet nozzle run on the end of the CT. The circulating fluids carry the debris back to the surface through the annulus between the CT string and the completion tubing.

Summer 2014 63

DEFINING COILED TUBING

Big Reels at the Wellsite

Oilfield Review Summer 2014: 26, no. 2.

Copyright © 2014 Schlumberger.

For help in preparation of this article, thanks to Rich Christie, Sugar Land, Texas, USA.

Matt VarhaugSenior Editor

>Mechanical scale removal. A jetting tool can be used to remove scale from a producing well. The tool consists of a rotating head with opposing tangentially offset nozzles and a drift ring. Jetting action from the nozzles removes scale from tubular walls while the drift ring allows the tool to advance only after the internal tubular diameter is clean. Nonabrasive fluids are pumped through the nozzle for removal of soft scales; abrasive beads are used to remove hard scales. When tubulars are completely plugged, abrasive jetting is used in conjunction with a powered milling head.

Tubing wall

ScaleJet nozzle

Rotating head

Drift ring

Fluid flow

Oilfield Review64

DEFINING COILED TUBING

Coiled tubing technology also extends to well perforating operations—shooting holes in casing to initiate production in a well. In many wells, per-forating guns are run downhole on wireline; however, because wireline tools depend on gravity to reach the target zone, they may not reach target depth in horizontal or highly deviated wells. One alternative is to convey the guns downhole at the end of the CT, which allows for substantially longer gun strings and higher-angle deployments than are possible on wireline. These operations can even be performed with tubing in place.

Its capacity to circulate or inject fluids makes CT especially suited to initiating production in a well. When drilling or workover fluids exert hydro-static pressures that exceed formation pressure, reservoir fluids are pre-vented from entering the wellbore. Pumping nitrogen gas through the CT string and into the fluid column is a common method for reducing hydro-static pressure within the wellbore to initiate production. The CT string is run to its target depth, and the nitrogen is pumped through the string to reduce the density of the hydrostatic column. When the hydrostatic pres-sure of the fluid column drops below reservoir pressure, the well can flow.

Operators frequently utilize coiled tubing as a conduit for accurate place-ment of cement downhole. Cement is used for sealing perforations or casing leaks, for primary or secondary zonal isolation and for plugs used in kickoff or abandonment operations. A cement squeeze enables the operator to plug casing leaks or existing perforations by pumping cement slurry under pressure into these openings. The cement fills openings between the formation and the cas-ing, forming a seal. Setting a cement plug involves circulating a cement slurry into position using CT then withdrawing the CT string to a point above the top of cement. A slight squeeze pressure is applied if necessary, any cement remaining in the tubing is displaced by a tail slurry then the CT is pulled out of the hole.

Well treatment programs may use CT to convey stimulation fluids that boost production by restoring or improving reservoir permeability. In a matrix treatment, fluids are pumped into a reservoir at a pressure that is greater than reservoir pressure but less than the formation fracture threshold. This tech-nique pushes the fluids through the formation pore spaces without initiating fractures. A similar operation, fracture acidizing, pumps fluids at a pressure that intentionally initiates fractures.

CT can facilitate the installation of production tubing and associated com-pletion equipment. In certain wells, a string or section of CT may be left in the borehole as a permanent part of the completion. CT completions often provide a low-cost approach for prolonging the life of old wells. Typical installations include velocity strings, tubing patches and through-tubing gravel packs.

For example, in some wells, operators choose to install CT permanently as a velocity string inside existing production tubing. In this application, the CT reduces the cross-sectional flow area of the production tubing, thus yielding higher flow velocity for a given production rate and allowing fluids to be carried out of the well more efficiently.

Coiled tubing can serve both as a conveyance and a medium for patching production tubulars. A CT tubing patch can be positioned in a completion to cover mechanical damage or erosion in tubing, to permanently shut off a sliding sleeve or to isolate perforations. Packers set at the top and bottom of the patch hold it in position and provide the seal between the existing com-pletion and CT string.

Coiled tubing is also used in completion programs to convey downhole hardware, fluids and materials. Frequently, wells drilled in unconsolidated sands require the wire mesh screen of a gravel pack (GP) to prevent sand production. Common GP installations involve a washdown procedure. First,

the CT string is run to the GP depth. Gravel is then pumped through the coiled tubing. The CT string is retrieved to the surface, and a GP screen assembly is attached. As the cylindrical screen is run to the top of the gravel, fluid is pumped through the CT to agitate the gravel and settle the screen into place across from the perforations (above). The CT string is then retrieved to the surface. The GP keeps the sand in place while allowing formation fluids to flow through it. Should sanding begin later in the life of a well that does not have a GP, coiled tubing offers a means of installing a through-tubing GP completion, in which GP screens are installed through the existing produc-tion tubing without removing the original completion hardware.

CT technology has expanded into openhole operations, to include drilling and associated activities. Coiled tubing drilling (CTD) can accommodate a variety of applications, including directional or nondirectional wells. CTD is carried out with a downhole motor and, compared with conventional drilling applications, uses higher bit speeds and lower weight on bit. In directional wells, a steering assembly is required to direct the well trajectory. CTD is used in both overbalance and underbalance drilling applications.

Significant AdvantagesCT equipment and techniques present several advantages over those used in conventional drilling and workover operations. These advantages include rapid mobilization and rig-up, fewer personnel, smaller environmental foot-print and reductions in time associated with pipe handling while running in and out of the hole. Such capabilities are especially important in deep or high-angle wellbores. Coiled tubing can help the operator avoid the risk of formation damage inherent in killing a well by allowing continuous circula-tion during well intervention operations. These advantages may yield sig-nificant cost savings over conventional drilling or workover techniques.

> Gravel pack washdown. As the gravel pack screen is lowered toward the top of the gravel, surface pumps are activated. The pump rate is sufficient to fluidize the gravel without causing it to circulate back into the tubing. While the pumps are active, the CT is slowly lowered into the gravel until the screen reaches its setting depth. A ball pumped through the CT string releases the screen before the CT string is pulled back to the surface.

Screen

Gravel

Coiled tubingFluid flow

Fluid flow

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