SCHLUMBERGER OILFIELD REVIEW

SPRING 2006

VOLUME 18 N

UMBER 1

Spring 2006

Oilfield Review

Oilfield Technologies in Space

Sonic Advances

Acoustic Waves

High-Resolution Core Visualization

06_OR_002_0

58171schD01R1.qxp 5/31/06 8:34 PM Page 2

To say Chevron shared in Schlumberger sonic-logging tooldevelopment would be an understatement. Through theyears, I recognized that tool development, like most sci-ence, requires collaboration and that success is criticallydependent on human relationships like those we forgedwith Schlumberger R&D people. A couple of decades ago,as a relatively new Chevron employee, I got my first taste of collaboration with an engineering prototype of the Array-Sonic* tool. This was Chevron’s first experience with anarray tool, and it seemed we were entering a new world of borehole acoustics.

Some years later, we organized a test of dipole sonic toolsin the Sacramento Valley, California, USA. Steve Chang,then at Schlumberger-Doll Research, brought an experi-mental prototype of the DSI* Dipole Shear Sonic Imager,and Mobil brought their groundbreaking proprietary sonictools. Such tests let us build our in-house expertise. Forexample, we showed the original DSI development teamthat waveforms contained large dispersion, therefore justi-fying filtering at frequencies as low as 1.5 kHz, although we could not explain the cause for the dispersion in thoseearly days.

Our involvement with MSIP (Modular Sonic Imaging Plat-form), the engineering name for the Sonic Scanner*acoustic scanning platform, began in 2001, when Schlumberger Kabushiki Kaisha (SKK) requested a fieldtest of an experimental prototype in diatomite formations inCalifornia’s San Joaquin Valley. Diatomite poses challengesfor sound waves, as illustrated in “Sonic Investigations Inand Around the Borehole,” page 14. Hitoshi Tashiro, theMSIP project manager at the time, visited us from SKK andbravely showed results that were worse than those from theolder DSI tool. This is not surprising for an experimentalprototype tool, and Hitoshi said, “We have work to do.”Three more experimental prototype tests followed, with sus-tained field support from Chevron’s Dale Julander. On thefollowing visit, Vivian Pistre, the next project manager,demonstrated truly impressive results with the prototype.We at Chevron sensed the new tool’s potential and its com-plexity. We requested a written client guide, which DavidScheibner wrote and let us review at several stages. Weawaited the transition of this sonic tool to engineering-pro-totype status so we could have the acoustic waveforms tounderstand and use this technology in Chevron.

The first time we got these waveforms in 2004 was inter-esting. David oversaw the field acquisition in California andhe used The Sunday Los Angeles Times as padding materialfor protecting the data DVDs in shipment. For the packagegoing to SKK, he wrapped a DVD in the sports section, pre-sumably for the sports lovers there, and for me, he used thedelightfully colorful food section. When I saw the data

A Tool Journeys from Research to Commercialization: A Client’s Perspective

file—a 1-GB file seemed huge then—his thoughtful choiceallowed me to enjoy a huge photograph of a cantaloupebefore attacking the data. Today, we are undaunted by 6-GB waveform files.

Upon recovering from the initial jolt of handling wave-forms from three dimensions—well azimuth, well axis and time—and a long dipole chirp signal, we began ourown journey of learning about this new tool. This path hadits highs (as when we discovered a polarity error duringacquisition, offered a solution and Vivian readily agreed to implement it) and lows (as we waded through hundredsof mysteriously named mnemonics). Vivian kept a growinglist of our suggestions, many of which Schlumberger imple-mented. Throughout this time the development team wasunwaveringly patient in answering our multitudinous ques-tions. In addition to Vivian and David, we turned to TakeshiEndo with questions. John Walsh, Tom Plona and JeffAlford also helped us. In all, the tool had at least eight fieldtests in Chevron wells, the most extensive one being in aGulf of Mexico well, with strong support from Chevron’sLarry O’Mahoney.

Now that MSIP technology has become the Sonic Scannertool, we are entering another new world of boreholeacoustics. This tool can probe the formation around thewellbore in new ways, opening the door to better decisionsin drilling, completions and petrophysical predictions.Other new applications will emerge. History shows sonictechnologies advanced faster when both Schlumberger andclients had easy access to raw waveforms. We must remem-ber this history lesson of collaboration and sharing rawdata; otherwise we risk stagnation.

Gopa S. DeChevron Energy Technology CompanySan Ramon, California

Gopa S. De is a Research Consultant with Chevron Energy Technology Com-pany in San Ramon, California, USA. She began her career with Chevron OilField Research Company in 1982. Her major research interests are sonic loggingand rock physics. She has a PhD degree in condensed matter physics from theUniversity of California, San Diego. Gopa is a member of the American PhysicalSociety, the Society of Exploration Geophysicists (SEG), the SEG ResearchCommittee and the SPE Reservoir Evaluation & Engineering (SPEREE) Review Board. Gopa rendered her signature in both Roman and Bengali scripts.

1

* Mark of Schlumberger.

58171schD02R1.qxp 7/6/06 12:00 PM Page 1

Schlumberger

Oilfield Review4 A Closer Look at Pore Geometry

Geoscientists are applying advanced medical technology toimage internal structures of rock and core samples at high resolution. X-ray computed tomography provides a digital alternative to conventional core analysis. With the help of innovative visualization technology, digital core volumes can be virtually sliced, manipulated and viewed from any angle,revealing highly detailed information about porosity, permeability and rock composition.

14 Sonic Investigations In and Around the Borehole

Advances in sonic tool design and data quality are improvingour ability to characterize mechanical and fluid propertiesaround the borehole and deep into the formation. This articlehighlights sonic-logging applications including measurements inultraslow formations, radial profiling to identify near-wellboredamage, anisotropy analysis for designing completion opera-tions, enhanced permeability estimation and high-resolutionimaging far from the borehole.

Executive EditorMark A. Andersen

Advisory EditorLisa Stewart

Senior EditorsMark E. TeelMatt Garber

EditorsDon WilliamsonRoopa GirMatt Varhaug

Contributing EditorsRana RottenbergJoan Mead

Design/ProductionHerring DesignSteve Freeman

IllustrationTom McNeffMike MessingerGeorge Stewart

PrintingWetmore & CompanyCurtis Weeks

Address editorial correspondence to:Oilfield Review1325 S. Dairy Ashford Houston, Texas 77077 USA(1) 281-285-7847Fax: (1) 281-285-1537E-mail: editorOilfieldReview@slb.com

Address distribution inquiries to:Matt GarberSchlumberger Cambridge ResearchHigh Cross, Madingley RoadCambridge, England CB3 0EL(44) 1223 325 377Fax: (44) 1223 361 473E-mail: DistributionOR@slb.com

Useful links:

Schlumbergerwww.slb.com

Oilfield Review Archivewww.slb.com/oilfieldreview

Oilfield Glossarywww.glossary.oilfield.slb.com

On the cover:

The Cassini space probe approachesSaturn after its 61⁄2 year journeyacross the solar system. Some of theinstruments it carries are based onoilfield technologies and are designedto help scientists perform close-upstudies of Saturn, its rings, moonsand magnetic environment.

2

58171schD03R1.qxp 7/6/06 12:14 PM Page 2

Spring 2006Volume 18Number 1

63 Contributors

64 Coming in Oilfield Review

3

34 Borehole Acoustic Waves

This article describes simple and complex acoustic waves in the borehole. We examine propagation of compressional,shear, Stoneley and flexural waves in isotropic, anisotropic andinhomogeneous formations from monopole and dipole sources.

44 From Inner Earth to Outer Space

Sensitive oilfield detectors are helping scientists investigatethe fundamental nature and origin of objects in outer space. In 1996, the Near Earth Asteroid Rendezvous spacecraft, orNEAR, equipped with oilfield sensors, left Earth for the asteroid 433 Eros, some 160 million miles away. In this article,we discuss the NEAR mission along with other examples thatshow how oilfield technologies are being used in the quest forknowledge and understanding of outer space.

Syed A. AliChevron Energy Technology Co.Houston, Texas, USA

Abdulla I. Al-KubaisySaudi AramcoRas Tanura, Saudi Arabia

Roland HampWoodside Energy, Ltd.Perth, Australia

George KingBPHouston, Texas

Eteng A. SalamPERTAMINAJakarta, Indonesia

Y.B. SinhaIndependent consultantNew Delhi, India

Richard WoodhouseIndependent consultantSurrey, England

Advisory Panel

Oilfield Review subscriptions are available from:Oilfield Review ServicesBarbour Square, High StreetTattenhall, Chester CH3 9RF England(44) 1829-770569Fax: (44) 1829-771354E-mail: orservices@t-e-s.co.ukwww.oilfieldreview.comAnnual subscriptions, including postage,are 180.00 US dollars, subject toexchange-rate fluctuations.

Oilfield Review is published quarterly bySchlumberger to communicate technicaladvances in finding and producing hydro-carbons to oilfield professionals. OilfieldReview is distributed by Schlumberger toits employees and clients. Oilfield Reviewis printed in the USA.

Contributors listed with only geographiclocation are employees of Schlumbergeror its affiliates.

© 2006 Schlumberger. All rights reserved.No part of this publication may be repro-duced, stored in a retrieval system ortransmitted in any form or by any means,electronic, mechanical, photocopying,recording or otherwise without the priorwritten permission of the publisher.

58171schD03R1.qxp 5/25/06 10:46 PM Page 3

4 Oilfield Review

A Closer Look at Pore Geometry

Andreas Kayser Cambridge, England

Mark KnackstedtThe Australian National UniversityCanberra, Australia

Murtaza ZiauddinSugar Land, Texas, USA

For help in preparation of this article, thanks to Veronique Barlet-Gouédard, Gabriel Marquette, Olivier Porcherie and Gaetan Rimmelé, Clamart, France;Bruno Goffé, Ecole Normale Supérieure, Paris; and Rachel Wood, The University of Edinburgh, Scotland.Inside Reality and iCenter are marks of Schlumberger.

X-ray computed tomography has advanced the field of medicine for more than 30 years.

For nearly as long, it has also been a valuable tool for geoscientists. Improvements in

the technology are helping geoscientists uncover greater detail in the internal pore

structure of reservoir rock and achieve a better understanding of conditions that

affect production.

Information gained through core analysis isinvaluable for predicting the producibility of areservoir pay zone. While other methods enablepetrophysicists to estimate grain size, bulkvolume, saturation, porosity and permeability offormations, core samples often serve as thebenchmark against which other methods arecalibrated. However, notwithstanding severalhundred thousand feet of whole or slabbed coreresiding in core libraries around the world, mostwells are not cored.

The wealth of information obtained fromcores comes at a price. Coring often increases rigtime, lowers penetration rates and increases therisk of sticking the bottomhole assembly. At somewells, hostile downhole or surface conditionsmake coring too risky. In other cases, correla-tions are not sufficient to allow geologists toaccurately and confidently pick coring points.Instead, many operators rely on sidewall coresobtained through prospective pay zones, and maycompensate for lack of whole core data bysupplementing their usual logging program witha wider range of measurements.

58171schD04R1.qxp 5/22/06 10:01 AM Page 4

Spring 2006 5

As oil companies try to drain aging reservoirsmore efficiently, engineers and geoscientists maycome to regret earlier decisions to forgo coring.Once a well has been drilled through a pay zone,it is too late to go back to obtain whole cores,unless the well is sidetracked. However,mineralogy, grain size, saturation, permeability,porosity and other measures of rock fabric cansometimes be determined without coring.

With improvements on the early medical CAT-scan technique developed in 1972, geoscientistscan take a series of fine, closely spaced X-rayscans through a rock sample to obtain importantinformation about a reservoir.1 Using anondestructive technique called microcomputedtomography, a focused X-ray beam creates“virtual slices” that can be resolved to a scale ofmicrons, not just millimeters.2 These refinementsalso allow the option of examining smallersamples of rock; instead of depending on wholecores for porosity and permeability measure-ments, geoscientists can now use formationcuttings to estimate these properties.3 Althoughmany companies do not core their wells, theyusually employ the services of a mudlogger tocatch formation cuttings as they come over theshale shaker. When they don’t have core,geoscientists are finding that a sliver of rock canbe highly revealing.

This article reviews the development of X-raycomputed tomography (CT) and the ensuingtechnology transfer from medical to oilfieldapplications. We describe how the data can beevaluated using immersive visualization tech-niques and discuss a range of oilfield applicationsthat may benefit from this technology. Finally, wewill see how this technology served researchers intheir evaluation of casing cement and wellstimulation treatments.

CT Scan TechnologyOriginally developed for medical use by GodfreyNewbold Hounsfield in 1972, computed tomog-raphy uses X-ray scans to investigate internalstructures within a body, such as those of softtissue and bone.4 CT overcomes the problem ofsuperimposition exhibited in conventional X-rayradiography when three-dimensional features of internal organs are obscured by overlyingorgans and tissues imaged on two-dimensional X-ray film.

Rather than projecting X-rays through apatient and onto a film plate, as withconventional X-rays, the CT process takes adifferent approach. The CT scanner uses arotating gantry to which an X-ray tube ismounted opposite a detector array. The patient is

placed in the center of the gantry, while theopposing X-ray source and detectors rotatearound the patient. With the patient positionedroughly in the middle of the source-receiverplane, the rotating gantry allows a series ofclosely spaced radiographic scans to be obtainedfrom multiple angles. These scans, orradiographic projections, can then be processedto obtain a 3D representation of the patient(below).

CT radiographic projections are based on thedifferential attenuation of X-rays caused bydensity contrasts within a patient’s body. This

patient from this equation, attenuation is afunction of the energy of the X-ray as well as thedensity and atomic number of the elementsthrough which the X-ray passes. The correlationis fairly straightforward: lower-energy X-rays,higher densities and higher atomic numbersgenerally result in greater attenuation.5

Digital projection data are converted into acomputer-generated image using tomographic-reconstruction algorithms to map the distribu-tion of attenuation coefficients.6 This distributioncan be displayed in 2D slices, composed of pointsthat are shaded according to their attenuation

1. In the medical field, the computerized axial tomography (CAT) scan is sometimes also calledcomputer-assisted tomography, and is synonymous with computed tomography.

2. A micron, or micrometer, is equal to one millionth of ameter, or more commonly, one thousandth of a millimeter.It is abbreviated as µ, µm or mc. In English measure, amicron equals 3.937 x 10-5 in.

3. Siddiqui S, Grader AS, Touati M, Loermans AM andFunk JJ: “Techniques for Extracting Reliable Density and Porosity Data from Cuttings,” paper SPE 96918,presented at the SPE Annual Technical Conference andExhibition, Dallas, October 9–12, 2005.Bauget F, Arns CH, Saadatfar M, Sheppard AP, Sok RM,Turner ML, Pinczewski WV and Knackstedt MA: “What

is the Characteristic Length Scale for Permeability?Direct Analysis from Microtomographic Data,” paperSPE 95950, presented at the SPE Annual TechnicalConference and Exhibition, Dallas, October 9–12, 2005.

4. Hounsfield GN: “A Method of and Apparatus forExamination of a Body by Radiation such as X- orGamma Radiation,” British Patent No. 1,283,915(August 2, 1972).

5. For more on X-ray CT: Publication Services Department of the ODP Science Operator. http://www-odp.tamu.edu/publications/185_SR/005/005_5.htm(accessed January 27, 2006).

6. Feldkamp LA, Davis LC and Kress JW: “Practical Cone-Beam Algorithm,” Journal of the Optical Society of America A1, no. 6 (June 1984): 612–619.

> Thoracic CAT scan. Manipulating color and opacity values of differenttissues provides physicians with an unobstructed view of a patient’s lungsand skeletal system. (Image courtesy of Ajay Limaye, VizLab, The AustralianNational University.)

attenuation represents a decrease in energy as X-rays pass through various parts of the body.Some tissues scatter or absorb X-rays better thanothers: thick tissue absorbs more X-rays thanthin; bone absorbs more X-rays than soft tissue,while fat, muscle or organs allow more X-rays topass through to the detectors. Removing the

values (see “Moving from 2D Points to 3DVolumes,” page 6). Thus, in hospital scans, bonewould typically be assigned a light color tocorrespond with its comparatively highattenuation value, while air-filled lung tissuemight be assigned a darker color correspondingto low attenuation values.

58171schD04R1.qxp 5/22/06 10:01 AM Page 5

6 Oilfield Review

In the mid-1880s, Neo-Impressionist artistGeorges Seurat perfected a revolutionarytechnique of painting with tiny dots of color.Like Michel Chevrul before him, Seurat recog-nized that from a distance, the eye wouldnaturally blend together tiny dots of primarycolors to produce secondary shades. Using tinybrush strokes, Seurat and his contemporariescaptured scenes of cityscapes, harbors andpeople at work and leisure. This techniquecame to be known as pointillism.

Computers use a similar technique to displaytext and images; however, they work at a muchfiner scale. Every image portrayed on a com-puter monitor or video screen is composed ofmany, almost imperceptibly tiny dots, spacedat extremely close intervals. In a 2D picturescreen, each dot, or pixel (a word formedfrom the contraction of picture element) canbe defined by its horizontal (x) and vertical(y) screen coordinates. It is also defined by itscolor value. In color images, each pixel is alsoassigned its own brightness.

The number of shades that a pixel can take on depends on the computer and the numberof bits per pixel (bpp) it is capable of process-ing. Common values range from 8 bpp (28 bits,which translates to 256 colors) to 24 bpp (224 bits, or 16,777,216 colors). On an eight-bitgray-scale image, for instance, each pixelwould be assigned a value corresponding to ashade of gray, ranging from 0 to 255, where 0represents black and 255 represents white.

The number of pixels used to create animage controls its resolution (above right). As more pixels are used, the image can be por-trayed in greater detail, or higher resolution.Resolution is thus initially impacted by theimage acquisition system and later, by theimage display system.

Resolution in digital image acquisition systems is largely governed by the number oflight-sensitive photoreceptor cells, known asphotosites, which are used to record an image.These photosites (more commonly referred to

as pixels) accumulate charges correspondingto the amount of light passing through thelens and onto each cell.1 As more light fallsonto a photosite, the charge grows. Light isshut off to the lens once the shutter closes, at

which point the charge in each cell isrecorded by a processing chip and convertedto a digital value that determines the colorand intensity of individual pixels used to dis-play the image on screen. Resolution in these

Moving from 2D Points to 3D Volumes

> Pixel resolution. The sharpness and clarity of an image are affected by pixel count and the sizeof the pixels. To increase the number of pixels within a fixed space, pixel size must be reduced.As pixel size (in white) progressively decreases (left to right), more pixels can be used to providegreater detail in the image.

> Pixel to voxel. A flat pixel (left) takes on a new dimension when the slice on which it resides isstacked with other slices to form a volume (right). Adding the z-coordinate of the slice numberessentially assigns a depth-value to the pixel, thus creating a voxel within the stack of slices.

0

0 Color bar 256

0

200

400

600

800

1,000

Verti

cal c

oord

inat

es, y

200 400 600Horizontal coordinates, x

800 1,000

color

x

y

Pixel

0

200

400

600

800

1,000

Verti

cal c

oord

inat

es, y

0 200 400 600Horizontal coordinates, x

Slice number, z

800 1,000

Voxelx

y

z

58171schD04R1.qxp 5/22/06 10:02 AM Page 6

Spring 2006 7

Evolving to Industrial StrengthDensity contrasts within a rock body can beimaged just as they can within a human body(below). By the mid 1980s, CT technology wasmaking significant inroads into geoscienceapplications. In addition to quantitativedetermination of bulk density of rock samples,CT scanning was adapted to visualize microbialdesulfurization of coal, displacement of heavy oil,and oil flow through carbonate cores.7

It didn’t take long for those outside themedical field to recognize the potential of CTtechnology for nondestructive evaluation ofmaterials. Geoscientists soon joined the ranks ofother researchers, particularly those in the fieldof materials testing, who sought increasinglyfiner detail for imaging internal structures. Thiscapability has largely been realized throughdevelopment of industrial-strength CT systems,which can employ more powerful X-rays, a tighterfocal point and longer exposure times than thoseused in the medical field.8

7. Kayser A, Kellner A, Holzapfel H-W, van der Bilt G,Warner S and Gras R: “3D Visualization of a RockSample,” in Doré AG and Vining BA (eds): PetroleumGeology: North-West Europe and Global Perspectives –Proceedings of the 6th Petroleum Geology Conference.London: The Geological Society (2005): 1613–1620. Vinegar HJ: “X-ray CT and NMR Imaging of Rocks,”Journal of Petroleum Technology 38, no. 3 (March 1986):257–259.

8. For more on high-resolution X-ray CT: University of TexasHigh-Resolution X-ray Computed Tomography Facility.http://www.ctlab.geo.utexas.edu/overview/index.php#anchor1-1 (accessed January 30, 2006).

devices is often expressed not in terms of pho-tosites, but rather in megapixels. A 1.2-megapixel device, for instance, might havean area of 1,280 x 960 (1,228,800 pixels), whilehigher resolution would be attained by a 3.1-megapixel device measuring 2,048 x 1,536(3,145,728 pixels).

Image resolution can then be affected bythe medium on which it is displayed. A rela-tively low-resolution computer monitor mightbe described as a 640 x 480 display. Thismeans that the monitor has a width of 640 pix-els, spread across a height of 480 lines,totaling 307,200 pixels. If those pixels werespread across a 15-inch monitor, then anyimage displayed on that monitor would beallotted 50 dots per inch. To increase resolu-tion, either the screen size must be reducedor more pixels must be packed into thescreen. Modern applications generally takeboth approaches, squeezing a huge number of pixels into a smaller area.

To image a 3D object, the pixel is expandedinto another dimension. A third coordinate(z) is added to the x-y location to preciselydefine the pixel’s position within the volumeof a 3D object, thereby creating a voxel—short for volume pixel. In CT images, thez-coordinate often denotes depth, and is dic-tated merely by the position that atomographic slice holds within a volumeformed by stacking together numerous closelyspaced slices (previous page, bottom). Inaddition to x, y and z coordinates, a voxel candefine a point by a given attribute value. Inthe case of CT scans, that value is density,which is a function of the sample’s trans-parency to X-rays. Density values can be tiedto a color spectrum, while a range of intensi-ties can control the opacity of a voxel on acomputer screen. With this information and3D rendering software, a two-dimensionalimage of a 3D object can be generated forviewing at various angles on a computer screen.

> Density values of various minerals commonly found in sedimentary rock. X-rays used to visualize rock structures are affected, in part, by differencesin density and mineralogy within a sample.

Quartz

Calcite

Anhydrite

Barite

Celestite

Mineral Density, g/cm3 Mineral Density, g/cm3

2.64

2.71

2.98

4.09

3.79

Gypsum

Dolomite

Illite

Chlorite

Hematite

2.35

2.85

2.52

2.76

5.18

> A different kind of patient. A section of whole core is placed on a sliding gurney prior to imaging ata hospital CAT-scan facility.

1. Although experts may correctly assert that photositesare not actually pixels, the terms are becomingincreasingly interchangeable in the popular vernacu-lar, thanks largely to the broad appeal of digitalphotography, in which manufacturers of digital cam-eras describe resolution in terms of megapixels.

In the early days of CT rock scans, it was notunusual for geoscientists to work out agreementswith the only institution in town that couldprovide access to such sophisticated technology.Often in the dark of night, with as little attentionas possible, core samples from the oilpatch wouldbe wheeled into the pristine and sterile setting ofa hospital CAT-scanning facility for imaging andanalysis (below).

58171schD04R1.qxp.7.ps 7/5/06 5:02 PM Page 7

With the development of microCT (µCT),researchers are attaining much higherresolutions.9 Using µCT, researchers are some-times able to image their samples with voxelsizes as low as 2.5 µm. Depending on the size of asample and the number of pixels used to image it,voxel sizes of one-thousandth of the sample sizeare being attained. For example, a 1-megapixelcamera using 1,000 x 1,000 pixels couldconceivably resolve a 1-cubic centimeter sampleto about 10 µm. Similarly, a 16-megapixel camera(4,000 x 4,000 pixels) can resolve the samesample to 2.5 µm.

At such resolutions, geoscientists can distin-guish density or porosity contrasts inside a rocksample and can study pore space and poreconnectivity in great detail. This µCT technologypermits recognition of grains or cements withdifferent mineralogical compositions (right). Ithas even been used to differentiate grains of thesame type, such as those found in carbonates,where microporosity may vary between differentgrain types in the same rock.10

The Scanning ProcessThe scanning process to acquire µCT data is insome respects analogous to acquiring 3D seismicdata. A seismic crew shoots a series of regularlyspaced seismic lines. Coordinates of the startingand ending points of each line are surveyed,making it possible to infer the distance betweeneach line in the series. It is therefore possible todetermine the position of any point along anyline as well as the distance between pointswithin the series of lines. With this knowledge, aposition between any two points or lines can beinterpolated when the data are processed.

For µCT, a regular series of closely spacedscans are acquired to obtain high-resolutionvirtual slices of a sample. Each pixel in the slicerepresents a scanned point and has coordinatesthat correspond to an actual point in the sample.Because coordinates of each point are known,distances between each point and each slice canbe determined. And just like the seismic line,points or slices can be interpolated betweenexisting slices. By stacking the series of slicesclose together to make up a volume of data, eachpixel in a slice becomes part of the stack andtakes on a third dimension. In this way, eachpixel can be treated as a voxel.

The scanning process is carried out by highlyspecialized X-ray systems. Though severalcompanies offer research-grade systems, many X-ray microtomography devices are custom-built.Regardless of whether they are off-the-shelf orspecially designed, all rely on three primary

components: an X-ray source, a rotating stage onwhich the sample is placed and an X-ray camerato record the pattern of X-ray attenuation withina sample.

To scan a sample, it must be placed on therotating stage, positioned between the X-raysource and the camera. X-rays emitted from thesource are attenuated through scattering orabsorption before being recorded by thecamera.11 The camera then records a large seriesof radiographs as the sample rotates incre-mentally on its stage through 360°. A computerprogram stacks the digital projection data whilemaintaining true spacing between pixels andslices. CT algorithms are applied to these data toreconstruct the internal structure of the sampleand preserve its scale in three dimensions.

One such device was built in 2002 by TheAustralian National University in Canberra (next

page, top). Its source generates X-rays with a 2-to 5-µm focal spot. The X-ray beam expands fromthe focal point, creating a cone-beam geometry.12

Because magnification of the sample increaseswith proximity to the X-ray source, the rotatingstage and camera are designed to slideseparately on a rail, allowing researchers toadjust distances between source, sample andcamera. The sample stage can rotate the samplewith millidegree accuracy and can support up to 120 kg [265 lbm] of sample and associated test equipment.13

At this facility, the X-ray “camera” consists ofa scintillator that fluoresces green in response toX-rays, and a charge-coupled device (CCD) thatconverts this green light into electric signals.14

The camera has a 70-mm2 active area, containing4.1 megapixels (2,048 x 2,048 pixels). Thesystem’s large field of view allows researchers to

8 Oilfield Review

> Three-dimensional quantification and spatial distribution of sandstonecomponents. While most sandstones consist primarily of quartz grains andcement, X-ray imagery helps put other components into perspective.Differences in X-ray attenuation throughout the sample indicate changes indensity caused by porosity and various mineral constituents of the rock.Once mapped, these characteristics can be isolated for further scrutiny.

Sandstone grains and quartz cement: 78%

Barite cement: 1%

Pore space:16%

Calcite cement: 5%

58171schD04R1.qxp 5/22/06 10:03 AM Page 8

Spring 2006 9

image a 60-mm specimen with a 30-micron pixelsize. They can also zoom in for high-resolutionscanning to image a 4-mm specimen with 2-micron pixels.

Approximately 3,000 projections are neededto generate a 2,0483 voxel tomogram. Betweeneach projection, the sample stage is rotated0.12°. The entire process takes 12 to 24 hours,depending on the type of sample and the filteringsteps required to reduce sampling artifacts. Theresulting 24 gigabytes of projection data are

processed by supercomputer, and it takes 128central processing units about 2 hours togenerate the tomogram.

Visualization Technology Once individual radiographic projections havebeen compiled into a 3D data volume file, thedata can be loaded into an immersive visuali-zation environment for detailed examination.With Inside Reality virtual reality technology, thedata can be imaged and manipulated like anyother volume of 3D data. Originally developed tohelp visualize seismic volumes based on miles orkilometers of data, Inside Reality technology canalso handle data volumes based on much finer,submillimeter scales.

Geoscientists utilize this advanced visuali-zation technology to view a data volume from anydirection. This capability enables bedding planes

and fracture planes of rock samples to be viewedorthogonally, even when the physical sample hasbeen cut obliquely to these planes. Sedimentaryand structural features of the rock sample aretypically analyzed in the form of slices ortransparency views through a volume.

While the scanning process relies on densitydifferences to distinguish features within asample, the visualization process depends largelyon opacity differences. One way to exposefeatures deep within a volume comprisingmillions of voxels is to render surrounding voxelsinvisible. Opacity rendering is the key tovisualization. Each voxel is assigned a value alonga transparency-opacity spectrum, thus makingsome voxels stand out while others fade away.Without this capability, the opacity of outer voxelswould hide all features lying within the volume.

Voxel-based technology can be used todetermine the volume and geometry of rockgrains, cement, matrix and pore space within asample. Using Inside Reality opacity-renderingtools, geoscientists can assign different values ofthe opacity-transparency spectrum to variouscomponents within a volume. This techniqueallows geoscientists to distinguish betweenmaterials of different density values. Forexample, the distribution of cement betweenmineral grains shows up as a distinctive color,while setting pore space to zero-opacity makes ittransparent, thus showing the spaces betweengrains. This allows the viewer to separate rockgrains from cement, matrix and pore space toreveal internal sedimentary and structuralfeatures (below).

9. Abbreviations for microcomputerized tomography rangefrom µCT (where the Greek letter mu is a standardsymbol for the prefix “micro”) to uCT (where u is asubstitute for mu) to mCT (where the m stands for micro)to XMT for X-ray Microtomography.

10. Kayser A, Gras R, Curtis A and Wood R: “VisualizingInternal Rock Structures: New Approach Spans FiveScale-Orders,” Offshore 64, no. 8 (August 2004): 129–131.

11. Ketcham RA and Carlson WD: “Acquisition, Optimizationand Interpretation of X-Ray Computed TomographicImagery: Applications to the Geosciences,” Computers& Geosciences 27, no. 4 (May 2001): 381–400.

12. Sakellariou A, Sawkins TJ, Senden TJ and Limaye A: “X-Ray Tomography for Mesoscale PhysicsApplications,” Physica A 339, no. 1-2 (August 2004):152–158.Sakellariou A, Sawkins TJ, Senden TJ, Knackstedt MA,Turner ML, Jones AC, Saadatfar M, Roberts RJ,Limaye A, Arns CA, Sheppard AP and Sok RM: “An X-Ray Tomography Facility for Quantitative Prediction of Mechanical and Transport Properties in Geological,Biological and Synthetic Systems,” in Bonse U (ed):Developments in X-Ray Tomography IV, Proceedings ofSPIE—The International Society for Optical Engineering,Vol. 5535. Bellingham, Washington, USA: SPIE Press(2004): 473–474.

13. This test equipment includes pumps or other devicesused to study fluid flow or mechanical compaction.

14. Rather than exposing film to light, CCD technologycaptures images in a technique similar to commondigital photography. A CCD uses a thin silicon wafer torecord light pulses given off by a scintillator. The CCDsilicon wafer is divided into several thousand individuallight-sensitive cells. When a light pulse from thescintillator impinges on one of these cells, thephotoelectric effect converts the light to a tiny electricalcharge. The charge within a cell increases with everylight pulse that hits the cell. Each cell on the CCD siliconwafer corresponds in size and location to an imagepixel. The pixel’s intensity is determined by themagnitude of the charge within a corresponding cell.

> A high-resolution X-ray tomography device at The Australian National University. The rotatingsample stage and charge-coupled device (CCD) camera slide on a track, enabling adjustment of thedistance between the camera, sample and X-ray source. With this device, a sample can be magnifiedfrom 1.1 to more than 100 times its original size. The stage rotates with millidegree accuracy and canbe fitted with fluid pumps for imaging flow through porous media. (Figure courtesy of The AustralianNational University.)

Approximately 1.5 meters

Rotation stage X-ray sourceScintillator + CCD

> Sandstone pores. An opacity filter is used to render different features in volume windows usingInside Reality software. The left window above and behind the yellow arrow shows only quartz grains(light green) in this eolian sandstone from the Rotliegendes formation in Germany. A volume showingonly pore space (blue) is in the background on the right. The smaller volume in the foreground on theright shows late diagenetic barite cement (red). The slice making up the base image indicates quartz(gray), pore space (blue), barite (red) and carbonate cement (orange). The yellow arrow for scale is 1 mm long.

1.0 mm

58171schD04R1.qxp 5/22/06 10:04 AM Page 9

The ability to manipulate opacity values playsan important role in the seedpoint and volume-grower tools featured as part of the Inside Realitytoolbox. Using the seedpoint tool, the viewerselects a point within a slice or volume. This

point has a certain X-ray attenuation value. Oncea point is selected, the program can automat-ically pick all neighboring voxels of a similarvalue that are connected to that point. Thisfeature can help a geoscientist pick a point

within a volume known to represent porosity, forexample, and the volume-grower tool will display allinterconnected porosity within the volume (left).

Because each voxel is defined in part by itscoordinates, the distance between any two voxelscan be measured. To facilitate this process, theInside Reality system uses a ruler tool to providea visual scale. This tool can be used to measuregrain or pore size in three dimensions, helpinggeoscientists estimate pore-volume proportionsand connectivity.

Taking rock samples from the laboratory to animmersive visualization environment enables anasset team to share important information andconcepts about reservoir samples so they can makemore informed decisions. Inside Reality virtualreality technology lets geoscientists share 3Dvirtual core data with those in remote sites to helpasset teams collaborate with company experts andpartners around the world (below left).

Applications Rock fabric and textural data provide geologistswith key information used in analyzing facies andin determining depositional environments.Geologists and petrophysicists can now obtainimportant information about grain size, shapeand matrix from digital scans of core or corefragments. A single core-fragment image canyield thousands of individual grains. By digitallydisaggregating grains in a scanned sample,analysts can obtain coordinates of all voxelscomposing each grain, the number of neighboringgrains and grain-overlap information.15

From such a dataset, geologists can derive acomprehensive analysis of grain sizes anddistribution to obtain a full suite of statistical

10 Oilfield Review

15. Saadatfar M, Turner ML, Arns CH, Averdunk H,Senden TJ, Sheppard AP, Sok RM, Pinczewski WV,Kelly J and Knackstedt MA: “Rock Fabric and Texturefrom Digital Core Analysis,” Transactions of the SPWLA46th Annual Logging Symposium, New Orleans,June 26–29, 2005, paper ZZ.

16. Both the Udden-Wentworth and the Krumbein scales areused to classify rock samples according to diameter; theformer is a verbal classification while the latter isnumerical. According to the Udden-Wentworth scale,sedimentary particles larger than 64 mm in diameter areclassified as cobbles. Smaller particles are pebbles,granules, sand and silt. Those smaller than 0.0039 mmare designated as clay. Several other grain-size scalesare in use, but the Udden-Wentworth scale (commonlycalled the Wentworth scale) is the one that is mostfrequently used in geology. The Krumbein scale is alogarithmic scale, which assigns a value designated asphi to classify the size of the sediment. Phi is computedby the equation: ø = –log2 (grain size in mm).

17. Arns CH, Averdunk H, Bauget F, Sakellariou A,Senden TJ, Sheppard AP, Sok RM, Pinczewski WV andKnackstedt MA: “Digital Core Laboratory: Analysis ofReservoir Core Fragments from 3D Images,”Transactions of the SPWLA 45th Annual LoggingSymposium, Noordwijk, The Netherlands, June 6–9,2004, paper EEE.

18. Bennaceur K, Gupta N, Monea M, Ramakrishnan TS,Tanden T, Sakurai S and Whittaker S: “CO2 Capture andStorage—A Solution Within,” Oilfield Review 16, no. 3(Autumn 2004): 44–61.

> Sandstone tracking. An opacity filter has been used to highlight quartz grains in sandstone from aRotliegendes gas reservoir in Germany. In the volume (light gray), interconnected porosity (blue) isimaged using the volume-grower tool provided by Inside Reality software. Fringe (red) along the edgeof the porosity indicates possible connections to neighboring pores detected automatically by thesoftware. Carbonate cement (orange) is also shown in the volume. The horizontal slice shows quartzgrains (dark gray), pore space (black), carbonate cement (medium gray), and barite cement (white).

1.0 mm

> Visualization using Inside Reality technology. Bringing sample volumes intoan iCenter secure networked collaborative environment allows asset teamsto become immersed in their data. Stereo projection creates a perception ofdepth, providing a different perspective on the 3D nature of the rock and itsmicrostructure. Inside Reality visualization software provides a detailedimage of a foraminifera fossil measuring 1.5 x 1.0 mm (inset). This 3Dvisualization allows examination of the fossil from many different angles. Theanimated avatar mirrors the pointing motions and actions of another viewerwho is interacting with these data from a remote site.

58171schD04R1.qxp 5/22/06 10:04 AM Page 10

Spring 2006 11

measurements (above left). Grain volume ismeasured by counting the voxels in each distinctgrain, from which size is derived and then gradedagainst standard Udden-Wentworth or Krumbeinscales of grain sizes.16 Automated programs cantrack and classify individual grains according tograin shape characteristics of sphericity androundness or classify according to texturalcategories, such as sorting, grain contacts, and matrix or grain-support. Some programs can also measure anisotropy in grain orientationto help geoscientists ascertain sediment-transport direction.

More important than the detailed measure-ment of rock grains is the analysis of the spacebetween the grains and the contents therein.Opacity-rendering tools work particularly well inshowing what is not rock—that is, its porosity.Researchers can obtain a good picture of porosityby decreasing the opacity of dense voxelsrepresenting rock grains and cements, whilesimultaneously increasing the opacity of low-density voxels (right). This same opacity-rendering technique highlights the extent ofinterconnected porosity within the rock. Oncethe porosity is brought up on screen, geo-scientists can measure the size of pore spacesand pore throats using the ruler tool. Poreinterconnectivity can also be charted, using porenetwork models based on tomographic imaging(above right). Pore-throat and pore-size distri-bution, along with interconnectivity, figure

prominently in determining relative permeabilityand recovery estimates in reservoir samples—parameters that can be hard to quantify whendifferent fluids compete to flow through the same opening.

A variety of other measurements can be takenfrom tomographic images, from which importantinformation is derived. Analysts can directlycorrelate image data on pore structure andconnectivity to measures of formation factor,permeability and capillary drainage pressures.Comparisons of results obtained from µCTimages and conventional laboratory measure-ments on the same core material have generallyshown good agreement.17

Studying Effects of Carbon Dioxide on Casing CementIn an important application beyond the realm ofconventional petrophysics, µCT was used to studythe effects of carbon dioxide [CO2] on casingcement. Greenhouse gases, particularly CO2,have been linked to rising temperatures aroundthe world. Capturing CO2 emissions andsequestering them in the subsurface have beenproposed as a measure to reduce atmosphericgreenhouse-gas concentrations until low-emission energy sources become viable.18

However, CO2 becomes supercritical whentemperature and pressure conditions exceed

> Statistics obtained from a single slice of a sample. More than 4,100 grainswere virtually disaggregated from a single slice, allowing researchers tocompile detailed statistical data used to characterize rock fabric andtexture. When compared with other samples, these statistical measurescan help geologists sort out the depositional environment of the rock.(Adapted from Saadatfar et al, reference 15.)

Freq

uenc

y

0-1 0 1 2 3 4

10

20

30

40

50

= -log2 (diameter)

Medium

Grain Size

CoarseVery coarsesand Fine Silt

> A whole lot of nothing. By manipulating the opacity of a scanned sample image, it is easy to visuallyexamine either sand grains (green) or pore space (blue). In many evaluations, this detailed analysis ofpore space can reveal critical clues to future performance of a reservoir.

Grains and quartz cement

Opacity change

Pores and pore throats

> Pore-scale information derived from tomographic images. Pore centers(blue spheres), connected by pore throats (blue cylinders), are usedto model porosity within a sample of carbonate rock (yellow). The sizeand location of pore centers and pore throats in this network reflectactual conditions within the rock microstructure. The complexity andheterogeneity of carbonate pore networks are brought to theforefront as part of the rock matrix is rendered semitransparent whilepore space is rendered opaque. (Image courtesy of The AustralianNational University.)

58171schD04R1.qxp.11.ps 7/5/06 5:02 PM Page 11

31.1°C and 73.8 bar [87.9°F and 1,070 psi]—conditions that are easily exceeded in mostmedium to deep wells.19 Therefore, an importantaspect of any CO2 sequestration project is toknow how downhole materials will react tosupercritical CO2 (scCO2).

Scientists at Schlumberger CambridgeResearch in England have collaborated withtheir counterparts at Schlumberger RiboudProduct Center in Clamart, France, to investi-gate long-term effects of CO2 storage on wellboreintegrity. One such experiment sought todetermine how scCO2 would react with casingcement.20 Long used in oil and gas wells tohydraulically isolate pay zones from the surfaceand other permeable zones, portland-basedcements play a critical role in wellbore integrity.

This study focused on a sample of neatcement.21 The cylindrical cement sample wascured for three days at 90°C and 280 bar [194°Fand 4,061 psi]. Scientists obtained CT scans ofthe cement cylinder before exposing it to scCO2.The cement was then subjected to a wet scCO2

environment and kept at 90°C and 280 bar for30 days. Two sample plugs were cut from theoriginal cylinder and then scanned.

Using Inside Reality software, researcherswere able to manipulate the data volume tovisualize porosity and microfractures and arbi-trarily slice through zones of interest. Bycomparing scans acquired before and aftertreatment, researchers noted significant changes to the cement plug, resulting from scCO2

attack. Of particular interest were the formationand distribution of microfractures, along with azone of aragonite replacement and a zone ofmineral alteration characterized by highsecondary porosity.

The reaction between scCO2 and cementproduced an irregular carbonation front,extending 4 mm [0.16 in.] from the outer edge ofthe core toward its center. This lighter coloredcarbonation front was readily apparent in thegray-scale 3D volume, and in a color-coded slice(above right). Subsequent X-ray diffractionanalysis determined that the alteration front hada different composition than the original cement,which had been replaced by aragonite. Porositywas clearly enhanced in the regions around the microfractures and the aragonite front (right).

The tests suggested that exposure to scCO2

may cause conventional cement to lose morethan 65% of its strength after only six weeks.These important observations provided animpetus for creating new blends of cement.Schlumberger researchers have developed newscCO2-resistant cementing materials that display

good mechanical behavior after exposure toscCO2 gas. Laboratory tests on these newmaterials show only a slight decrease incompressive strength during the first two days,and essentially no loss for the subsequent three months.

Examining Wormholes Caused by Stimulation TreatmentsResearchers have also used CT imaging to studythe effects of heterogeneity on carbonate matrixstimulation. In one experiment, it was instru-mental in visualizing the effects of the originalporosity distribution on acid-dissolution patterns.

12 Oilfield Review

> A sample plug of neat cement. Only a few centimeters in length, this sample revealed importantinformation concerning the behavior of supercritical CO2 on portland cement. The tomographic gray-scale image of the cement sample (right), scanned with a resolution of 18.33 µm, shows a highconcentration of aragonite along the edge of a carbonation front, accompanied by an alteration front.An additional dissolution front of high porosity extends farther into the core. Circular holes with adiameter of 500 µm may represent air bubbles. Microfractures are filled with aragonite crystals.Lighter features represent higher CT values, signifying different mineralogy in the case of the filledmicrofracture, or different amounts of microporosity, in the case of the alteration front.

0 1cm

2

Alteration front

CT ImageSample Plug

Carbonation frontZone of very low porosity

Air bubble(Diameter 0.5 mm)

Dissolution front

Filled microfracture

Zone of very high porosity

> Highlighting the extent of supercritical CO2 alteration. Color-coding enhances features that may notbe readily apparent in gray-scale imaging. Microfractures formed during the supercritical CO2 attackserved as conduits for further aragonite alteration. The concentration of aragonite along the fracturesand the edge of the alteration front can be visually distinguished using color-coding provided byInside Reality software. Materials imaged are unaltered neat cement (green), an alteration front(yellow), and mineral-filled microfractures or carbonation front (red). Increased porosity (blue) marksthe extent of various dissolution patterns.

Neat cement

Aragonite front

S y s t e m M e n u – M a i n M e n uT o o l s

Restore Scene

Save SceneSnapshot

System Menu

Colormap

Faul t

Fence

Reser voi r

Ruler

Sketch

Sl ice

Sur face

ume Est imat ion

Volume Window

Wel l

Growing Stereo

Ins ide Real i ty

Vers ion 5.1 [90]

AUTOSAVE

SCR_040917_1736_1

SCR_040917_1847_1

58171schD04R1.qxp 5/22/06 10:05 AM Page 12

Spring 2006 13

Stimulation treatments are commonlyperformed in wells where poor permeabilitylimits production due to naturally tightformations or formation damage. A commonstimulation technique involves the injection ofacid into carbonate formations. Acid dissolvessome of the formation matrix material andcreates flow channels that increase thepermeability of the matrix.

The efficiency of this process depends on thetype of acid used, reaction rates, formationproperties and injection conditions. Whiledissolution increases formation permeability, therelative increase in permeability for a givenamount of acid is greatly influenced by injection

formation to facilitate the flow of oil. Better still,wormholes require only a small volume of acid toproduce significant increases in permeability.Researchers are therefore investigating factorsthat influence production of wormholes.

CT scanning has proved instrumental indetermining the effects that injection rate andspatial distribution of porosity have on dissolutionpatterns formed during stimulation experiments(below). Because it is nondestructive, thistechnique allows for characterization of the corebefore and after the treatment experiment so the development and shape of the wormhole canbe evaluated.

applications, it is easy to envision the potentialspread of new applications for µCT.

The technology will no doubt proveinstrumental in improving the interpretation andapplication of laboratory and log data. As anincreasingly important tool in nondestructivetesting, its application can be extended tolaboratory testing of unconsolidated or friableformation samples. The combination of µCTimaging with numerical calculations may lead tomore accurate predictions of a wide range of rockproperties crucial to exploration, reservoircharacterization and recovery calculations.

Further applications include development ofimproved cross-property correlations anddevelopment of libraries of 3D images that will

19. Above its critical point at 31.1°C and 73.8 bar, CO2becomes a supercritical fluid. In this compressed state,its properties lie between those of a gas and a liquid.With a lower surface tension than its liquid form,supercritical CO2 can easily penetrate cracks andcrevices. Unlike CO2 gas, however, it can dissolvesubstances that are soluble in liquid CO2.

20. Barlet-Gouédard V, Rimmelé G, Goffé B and Porcherie O:“Mitigation Strategies for the Risk of CO2 MigrationThrough Wellbores,” paper IADC/SPE 98924, presentedat the IADC/SPE Drilling Conference, Miami, Florida,USA, February 21–23, 2006.

21. Neat cement has no additives that would alter its settingtime or rheological properties.

> Visualizing wormhole development. A sample of Winterset limestone was scanned by CT before (bottom) and after (top) acid injection. This data volumeis displayed using Inside Reality visualization technology, in which pore space is rendered opaque, while surrounding voxels are rendered transparent.Initial distribution of pores (bottom) shows discrete clusters of pores (blue) along the long axis of the core. After acidizing (top), the core shows increasedporosity, with a dissolution pattern extending from right to left that further marks the flow of acid during injection.

conditions. At extremely low injection rates, acidis spent soon after it contacts the formation,resulting in relatively shallow dissolution alongthe face of the injection zone. High flow ratesproduce a uniform dissolution pattern becausethe acid reacts over a large region. In either case,the resulting gains in permeability requirerelatively large expenditures of acid.

However, at intermediate flow rates, longconductive channels known as wormholes areformed. These channels penetrate deep into the

Peering into the FutureTomography is not new to the oil industry. At theupstream end of the tomography spectrum liescrosswell seismic tomography; at the downstreamend is industrial process tomography forrefineries. As a research tool, µCT is used acrossa broad suite of industrial applications to monitorperformance of polymer-enhanced foams andpolyethylene resins or to view phase separationand pore-space characterization in formationsamples. Across this range of tomographic

allow a more rigorous and quantitative descrip-tion of rock type and texture. These quantitativedescriptions can be integrated with classicalsedimentological descriptions. The technologycan also make a significant contribution to thestudy of elastic behavior, porosity-permeabilitytrends and multiphase flow properties such ascapillary pressure, relative permeability andresidual saturations.

Future technological innovations will probablyinclude higher resolution to overcome problemsin predicting porosity when micropores fallbelow the detection capability of the presenttechnique. With the improving resolution of theirsamples, µCT technology is helping today’sgeoscientists to better see their world in a grainof sand. —MV

58171schD04R1.qxp 5/22/06 10:06 AM Page 13

14 Oilfield Review

Sonic Investigations In and Around the Borehole

J.L. Arroyo FrancoM.A. Mercado OrtizPemex Exploración y ProducciónReynosa, Mexico

Gopa S. DeChevron Energy Technology CompanySan Ramon, California, USA

Lasse RenlieStatoil ASAStjørdal, Norway

Stephen WilliamsNorsk Hydro ASABergen, Norway

For help in preparation of this article and in acknowledge-ment of their contributions to the development of the SonicScanner acoustic scanning platform and applications,thanks to Sandip Bose, Jahir Pabon and Ram Shenoy,Cambridge, Massachusetts, USA; Tom Bratton and Adam Donald, Denver, Colorado, USA; Chung Chang, TarekHabashy, Jakob Haldorsen, Chaur-Jian Hsu, Toru Ikegami,David Johnson, Tom Plona, Bikash Sinha and Henri-PierreValero, Ridgefield, Connecticut, USA; Steve Chang, Takeshi Endo, Hiroshi Hori, Hiroshi Inoue, Masaei Ito,Toshihiro Kinoshita, Koichi Naito, Motohiro Nakanouchi,Akira Otsuka, Vivian Pistre, Atsushi Saito, Anthony Smits, Hitoshi Sugiyama, Hitoshi Tashiro and Hiroaki Yamamoto,Sagamihara, Kanagawa, Japan; Rafael Guerra and Jean-Francois Mengual, Rio de Janeiro, Brazil; Dale Julander,Chevron, Bakersfield, California, USA; Larry O’Mahoney,Chevron, New Orleans, Louisiana, USA; Marcelo OsvaldoGennari, Reynosa, Mexico; Pablo Saldungaray, Veracruz,Mexico; Keith Schilling, Bangkok, Thailand; Kwasi Tagborand John Walsh, Houston, Texas; Badarinadh Vissapragada,Stavanger, Norway; Canyun Wang, Beijing, China; Erik Wielemaker, The Hague, The Netherlands; and Smaine Zeroug, Paris, France.Array-Sonic, CBT (Cement Bond Tool), DSI (Dipole ShearSonic Imager), ECS (Elemental Capture Spectroscopy), FMI (Fullbore Formation MicroImager), HRLA (High-Resolution Laterolog Array), LSS (Long-Spaced Sonic Tool),MDT (Modular Formation Dynamics Tester), OBMI (Oil-Base MicroImager), Platform Express, Sonic Scanner, TLC(Tough Logging Conditions) and Variable Density are marks of Schlumberger.

Sonic measurements have come a long way since their introduction 50 years ago.

The latest advancement in sonic technology delivers the highest quality data seen

to date, allowing acoustic measurements to characterize mechanical and fluid

properties around the borehole and tens of feet into the formation.

Finding and producing hydrocarbons efficientlyand effectively require understanding the rocksand fluids of a reservoir and of surroundingformations. Three basic oilfield measurements—electromagnetic, nuclear and acoustic—havebeen devised to achieve this end. With advancesin tool design and in data acquisition, processingand interpretation, each measurement type hasevolved to produce more and different information.None, perhaps, has evolved more than theacoustic, or sonic, measurement.

In their early days, sonic measurements wererelatively simple. They began as a way to matchseismic signals to rock layers.1 Today, sonicmeasurements reveal a multitude of reservoirand wellbore properties. They can be used toinfer primary and secondary porosity, permea-bility, lithology, mineralogy, pore pressure,invasion, anisotropy, fluid type, stress magnitudeand direction, the presence and alignment offractures and the quality of casing-cement bonds.

Improvements in sonic measurements areenhancing our ability to determine some of theseproperties. Accuracy is improving in the basicmeasurements, which consist of estimates ofcompressional- (P-) and shear- (S-) wave slow-nesses.2 Variations in slowness are also becomingmore fully characterized, leading to an improvedunderstanding of how formation propertieschange over distance and with direction.

Formation properties often vary directionally,so to be completely described, they must bemeasured in three dimensions. The borehole hasa natural, cylindrical 3D coordinate system:axial, or along the borehole; radial, or perpen-dicular to the borehole axis; and azimuthal, oraround the borehole. Variations around and awayfrom the borehole depend on many factors,including the angle the borehole makes withsedimentary layering. Axial variations are typicalof vertical boreholes in horizontal layers, and canindicate changes in lithology, fluid content,porosity and permeability. Radial rock- and fluid-property variations arise because of nonuniformstress distributions and mechanical or chemicalnear-wellbore alteration caused by the drillingprocess. Azimuthal variations can indicate aniso-tropy, which is caused by layering of mineralgrains, aligned fractures or differential stresses.

Improved characterization of compressionaland shear slownesses in terms of their radial,azimuthal and axial variations is now possiblewith a new sonic tool, the Sonic Scanner acousticscanning platform. High-quality waveforms andadvanced processing techniques lead to moreaccurate slowness estimates, even in unconsoli-dated sediments and large boreholes, and also toreliable through-casing slowness measurements.These improvements result in better characteri-zation of subsurface rock and fluid properties,meaning more stable wellbores, longer-lastingcompletions and enhanced production.

1. Léonardon EG: “Logging, Sampling, and Testing,” inCarter DV (ed): History of Petroleum Engineering. NewYork City: American Petroleum Institute (1961): 493–578.

2. Slowness, also called interval transit time, is thereciprocal of speed, or velocity. The common unit ofslowness is microseconds per foot (µs/ft).

58171schD05R1.qxp 5/22/06 10:43 AM Page 14

Spring 2006 15

This article describes the advances in tooldesign and resulting data quality of the SonicScanner tool. Examples from the USA, Norwayand Mexico highlight applications that includedetermining shear velocities in ultraslowformations, radial profiling for optimizingdrilling, completion and sampling operations,fluid-mobility logging, fracture characterizationand imaging away from the borehole.

Engineering SuccessMore so than electromagnetic and nuclearlogging tools, a sonic tool’s very presence in aborehole can introduce a bias to themeasurements it acquires. The steel tool housingis extremely efficient at propagating sonic waves.Sonic-logging tool designers have minimized thisunwanted effect by isolating the transmittersfrom the receivers with insulating materials or by milling slots and grooves into the steel sonde (see “History of Wireline Sonic Logging,”page 32). These efforts were aimed at delayingundesirable signals and making the tool astransparent to the measurement as possible.

The Sonic Scanner tool is completelydifferent from other tools. Its design, materialcomposition and components were engineered sothat the effects of its presence could be modeled.These effects can then be incorporated intopredicting the complete tool-borehole-formationresponse. These theoretical predictions havebeen verified by experimental results in a testwell having known formation properties. As aresult, tool effects can be predicted accurately inisotropic homogeneous formations, and real-timecorrections can be made at the wellsite.

The transmitter-receiver (TR) geometry andfunctionality of the new tool were carefullydesigned to provide P-, S-, Stoneley- and flexural-wave slowness measurements at varying radialdepths of investigation (for a review see “BoreholeAcoustic Waves,” page 34). These modes areacquired at a logging speed of 1,800 ft/h[549 m/h]. For the typical scenario with formationcompressional and shear speeds increasing withdistance from the borehole, this is achieved byincreasing TR spacing to probe deeper into theformation. The Sonic Scanner tool combines thislong-spaced approach with the close TR spacing ofa borehole-compensated arrangement, and alsoadds azimuthally distributed receivers. The toolfeatures 13 axial stations in a 6-ft [1.8-m] receiverarray. Each station has eight receivers placedevery 45° around the tool, for a total of 104 sensors

58171schD05R1.qxp 5/22/06 10:43 AM Page 15

on the tool.3 A monopole transmitter sits on each end of the receiver array, and anothermonopole transmitter and two orthogonallyoriented dipole transmitters are located fartherdown the tool (below).

Each of the three Sonic Scanner monopoletransmitters produces a stronger pressure pulsethan transmitters in previous sonic tools. With asharp “click,” they generate clear P- and S-waves,the low-frequency Stoneley mode and the high-frequency energy needed for cement evaluation.

Each of the two dipole transmitters is ashaking device consisting of an electromagneticmotor mounted in a cylinder suspended in the

tool. This mechanism generates a high-pressuredipole signal without inducing vibration in thetool housing. The shaking source can be driven intwo modes: the traditional dipole source in pulsemode produces a deep “click”; the new sourcealso produces a “chirp” with a frequency sweep(bottom left). The chirp mode sustains eachfrequency for a longer duration than narrow-band dipole sources, providing more dipoleenergy to the formation.

As in earlier sonic tools, such as the DSIDipole Shear Sonic Imager, the two dipolesources are oriented orthogonally. One vibratesin line with the tool reference axis, and the otherat 90° to the axis. These devices generate strongflexural modes—waves that gently shake theentire borehole the way a person might shake atree from its trunk. Flexural modes propagate upand down the borehole and also into theformation to different depths that depend ontheir frequencies. The frequency content—300 Hz to 8 kHz—of the new chirp dipole sourceexcites flexural modes in all borehole andformation conditions, including slow formations,and ensures maximum signal-to-noise ratio.

The new sonic tool delivers P, S, Stoneley andflexural-mode waveforms with unprecedentedquality. An example from a typical fast formationoffshore Norway shows waveforms acquired fromthe monopole and dipole transmitters (nextpage, top). At high frequencies, the monopolesource generates clear P-, S- and Stoneley waves,while at low frequencies, it generatespredominantly Stoneley waves. The X- and Y-dipole transmitters generate flexural waves. Thedispersion curves show slowness versus frequencyfor the nondispersive shear, slightly dispersiveStoneley and highly dispersive flexural arrivals.The low-frequency limit of the flexural-wavedispersion curve is in line with the slowness ofthe shear head wave and the true shear slownessof the formation. The two flexural curves match,indicating absence of azimuthal anisotropy.

Waveforms from the same sources in a slowformation in the USA display evident differencescompared with fast-formation results (next page,bottom). The high-frequency monopole source

16 Oilfield Review

3. Pistre V, Kinoshita T, Endo T, Schilling K, Pabon J,Sinha B, Plona T, Ikegami T and Johnson D: “A ModularWireline Sonic Tool for Measurements of 3D (Azimuthal,Radial, and Axial) Formation Acoustic Properties,”Transactions of the SPWLA 46th Annual LoggingSymposium, New Orleans, June 26–29, 2005, paper P.Pistre V, Plona T, Sinha B, Kinoshita T, Tashiro H,Ikegami T, Pabon J, Zeroug S, Shenoy R, Habashy T,Sugiyama H, Saito A, Chang C, Johnson D, Valero H-P,Hsu CJ, Bose S, Hori H, Wang C, Endo T, Yamamoto Hand Schilling K: “A New Modular Sonic Tool ProvidesComplete Acoustic Formation Characterization,”Expanded Abstracts, 75th SEG Annual Meeting, Houston (November 6–11, 2005): 368–371.

> The Sonic Scanner tool, with 13 axial stations in a6-ft receiver array. Each station has eight azimuthallydistributed receivers, giving the tool 104 sensors.Three monopole transmitters allow acquisition oflong-spaced and short-spaced data for boreholecompensation at varying depths of investigation.Two orthogonal dipole transmitters generateflexural waves for characterization of shear-waveslowness in slow and anisotropic formations.

10 ft

Isolator Far transmitter section

Uppermonopole

Electronics Receiver section

X and Ydipole

Lowermonopole

Farmonopole

R13 R1

200

Pres

sure

, Pa

400

0

–200

–400

Time, ms0 2 4 6 8 10 12 14 16 18 20

Mag

nitu

de, d

B

Frequency, Hz

0

20

40

60

80

100

103 104

The frequency sweep of the Sonic Scannerdipole transmitter. The strong chirp creates awide-band response (inset) that is flat fromabout 300 Hz to 8 kHz.

>

58171schD05R1.qxp 5/22/06 10:44 AM Page 16

Spring 2006 17

>Waveforms (left) from a fast formation offshore Norway. Monopole transmitters (top) at high frequencies (left) generate clearP-, S- and Stoneley waves, and low frequencies (right) generate mostly Stoneley waves. Flexural waveforms generated by thedipole transmitters (bottom) are recorded on the X (left) and Y (right) receivers. Dispersion analysis (right) shows slightly dispersiveStoneley data, highly dispersive flexural data and nondispersive shear data. The compressional wave is excited only at frequencieshigher than 8 kHz in this formation, and is not shown on the dispersion curve. [Modified from Pistre et al, reference 3 (SEG).]

Slow

ness

, µs/

ft

400

300

200

100

00 2 4 6 8

Frequency, kHz

Stoneley

Flexural

Shear

0 5 10 15 20Time, ms

1

3

5

7

9

11

13

Y Dipole

0 5 10 15 20Time, ms

1

3

5

7

9

11

13

Low Frequency

Rece

iver

num

ber

Rece

iver

num

ber

0 2Time, ms

4

1

3

5

7

9

11

13

High Frequency

1

3

5

7

9

11

13

0 5 10 15 20Time, ms

X Dipole

P

S

Stoneley

>Waveforms (left) from a slow formation in the USA. The high-frequency monopole source (top left) generates no shear waveand smaller Stoneley waves than in the fast-formation case. At low frequency, the monopole source (top right) generatespredominantly Stoneley waves. The X- and Y-dipole transmitters generate low-frequency flexural waves compared with the fastformation. Anisotropy in this formation causes flexural-wave splitting, creating a fast and slow flexural wave (bottom left andright, respectively). The low-frequency dispersion data (right) include the Stoneley mode and two flexural modes. Higherfrequency dispersion analysis of the P-wave data reveals dispersion—labeled leaky compressional—at higher frequencies.[Modified from Pistre et al, reference 3 (SEG).]

0 10Time, ms

20

1

3

5

7

9

11

Fast Flexural

30 0 10 20 30Time, ms

1

3

5

7

9

11

Slow Flexural

Rece

iver

num

ber

Rece

iver

num

ber

0 2 4 6Time, ms

1

3

5

7

9

11

13

High Frequency

1

3

5

7

9

11

13

Time, ms

Low Frequency

0 10 20 30

Slow

ness

, µs/

ft

900

800

700

600

0 1 2 3Frequency, kHz

500

Stoneley

Slow shear

Flexural Fast shear

Slow

ness

, µs/

ft

210

190

170

0 42 8Frequency, kHz

1506

Leakycompressional

Compressional

58171schD05R1.qxp 5/22/06 10:44 AM Page 17

generates no direct shear wave but does generateleaky-compressional waves. At low frequencies,the monopole source again generates Stoneleywaves, but, in addition, there is a strong leaky-compressional wave generated. The X- and Y-dipole transmitters generate flexural waves withthe characteristic low-frequency response of aslow formation. The dispersion data include theslightly dispersive Stoneley mode and the leaky-compressional wave, but no shear head wave, asexpected in a slow formation. In the absence of ashear head wave, the shear slowness is estimatedfrom the low-frequency limit of the flexural mode.

The flexural mode is not as dispersive as in afast formation, but more dispersive than thatexpected from a homogeneous, isotropic forma-tion. At low frequency, the two flexural-wavedispersion curves level off at different slow-nesses, indicating azimuthal anisotropy. Theflexural waveforms have been mathematicallyrotated into fast and slow shear-wave directions.4

Analysis of flexural-wave dispersion curvesfrom the Sonic Scanner tool classifies formationsaccording to anisotropy type by comparingobserved dispersion curves to those modeledassuming a homogeneous isotropic formation(below). In a homogeneous isotropic formation,shear waves do not split into fast and slowcomponents, so the two observed flexural-wavedispersion curves have identical slowness-versus-frequency signatures, and overlie the modeledcurve. In cases of intrinsic anisotropy, such asshales or fractured formations, the fast and slowshear-wave dispersion curves are separateeverywhere and tend to the true slowness at zerofrequency.5 In formations that have undergonedrilling-induced damage and are near failure butare otherwise homogeneous and isotropic, thetwo dispersion curves are identical but showmuch greater slowness at high frequencies thanthe modeled dispersion for a homogeneousisotropic formation. In formations with stress-induced anisotropy, the fast and slow shear-wavedispersion curves cross. This characteristicfeature is caused by near-wellbore stress

concentrations.6 These simplified relationshipsbetween dispersion curves are valid when onlyone physical mechanism controls wave behavior.When multiple mechanisms are involved, such asif both stress-induced and intrinsic anisotropyare present, the curves can be different.

In addition to acquiring openhole measure-ments in isotropic, anisotropic, homogeneous andinhomogeneous formations, the Sonic Scannertool provides high-quality results behind casing.The improved tool design records waveformsthrough casing with high signal-to-noise ratio.Powerful transmitters and large bandwidth allowacquisition of formation slowness data throughcasing and cement of varying thickness.

The ability to measure formation propertiesthrough casing allows companies to monitor themechanical effects of production on the produc-ing formation. Many formations undergo compac-tion, weakening or other changes with time as aresult of pressure depletion or water injection.

In an example from a Statoil well in the NorthSea, Sonic Scanner data were acquired in both8.5-in. open hole and behind 8-in. OD casingbefore any production (next page). The openholelogs in the zone of interest indicate a slower,softer formation between X,296 and X,305 m. The caliper log flags a washout in this interval.When compared with the openhole logs, thecased-hole compressional and shear slownessesare markedly similar, even through the washed-out zone. The dispersion curves in the two casesare also similar.

In the Middle East, the Sonic Scanner toolhas been used multiple times to acquire slownessthrough 133⁄8-in. casing in hole sizes larger than20 inches. In each case, despite poor cement,good shear-wave slowness data were acquiredover the entire interval, and adequatecompressional slowness was recorded over atleast half the interval.

The Sonic Scanner tool not only obtainsslowness results behind casing, but can alsosimultaneously evaluate the quality of thecement bond and the top of cement. Signalsrecorded by receivers 3 and 5 ft [0.9 and 1.5 m]from the two near monopole transmitters areprocessed to produce a discriminated attenua-tion measurement that is free of tool-normalization fluid effects and pressure andtemperature drifts. The results are comparableto those of the CBT Cement Bond Tool, but arealso corrected for casing and cement properties.Evaluation of well integrity and formationproperties in the same tool run can avoidseparate logging runs and reduce rig-time andmobilization costs.

18 Oilfield Review

> Flexural-wave dispersion curves for classifying formation anisotropy andinhomogeneity. In a homogeneous isotropic medium (top left), observeddispersion curves for flexural waves recorded on orthogonal dipole receivers(red and blue curves) match modeled flexural-wave dispersion (blackcircles). In an inhomogeneous isotropic formation (top right), both observeddispersion curves show greater slowness with increasing frequency than thehomogeneous isotropic model. Greater slowness with increasing frequencyindicates that the near-borehole region has become slower, a sign ofdamage all around the borehole. In a homogeneous anisotropic medium(bottom left), such as one with intrinsic anisotropy, the fast flexural-wavedispersion curve (red) matches the homogeneous isotropic model (to a firstapproximation), while the slow flexural-wave dispersion curve (blue) has thesame shape but is translated to higher slowness. In an inhomogeneousanisotropic medium (bottom right), the two observed flexural-wavedispersion curves cross. This phenomenon is a result of near-wellbore stressconcentration, and indicates stress-induced anisotropy.

Frequency

Slow

ness

Fast shear

Homogeneous Isotropic

Frequency

Slow

ness

Fast shear

Inhomogeneous Isotropic

Damaged formation,near failure

Frequency

Slow

ness

Fast shearSlow shear

Homogeneous Anisotropic

Intrinsic anisotropy:shales, layering, fractures

Frequency

Slow

ness

Fast shearSlow shear

Inhomogeneous Anisotropic

Stress-inducedanisotropy

58171schD05R1.qxp 5/22/06 10:45 AM Page 18

Spring 2006 19

Extreme SlownessSome formations are so slow that not only is theS-wave slowness greater than that of the mud,but the P-wave slowness approaches that of themud. In these circumstances, the P-wave losesenergy to the formation, in what is known as aleaky-P mode, and is dispersive. At the low-

frequency limit, the leaky-P dispersion curvetends toward the P-wave slowness, and at thehigh-frequency limit, it reaches the borehole-fluid slowness.7

The Antelope formation in the Cymric oil fieldin the San Joaquin Valley, California, is such acase, combining extreme slowness with other

complications that make sonic logging chal-lenging.8 The formation lithology is diatomite and cristobalite—forms of opalized silica.Permeability is low, averaging 2 mD. From earlierstudies, compressional-wave slowness in thisformation is known to approach 200 µs/ft, which isnear the slowness of the mud wave, and shear-

4. The X- and Y-dipole sources are separated by 1 ft. Whilethis avoids electrical cross-talk, it also means thatwaveforms must be shifted by 1 ft before Alford rotation.This reduces the number of collocated waveforms from13 to 11.Alford RM: “Shear Data in the Presence of AzimuthalAnisotropy: Dilley, Texas,” Expanded Abstracts, 56th SEGAnnual International Meeting, Houston (November 2–6,1986): 476–479.

5. For anisotropy to be identified in this way, the anisotropysymmetry axis must be perpendicular to the boreholeaxis. For example, crossed-dipole logging tools invertical wells can detect anisotropy caused by alignedvertical fractures, and in horizontal wells can detectanisotropy caused by horizontal laminations.

6. Sinha BK and Kostek S: “Stress-Induced AzimuthalAnisotropy in Borehole Flexural Waves,” Geophysics 61,no. 6 (November-December 1996): 1899–1907.Winkler KW, Sinha BK and Plona TJ, “Effects ofBorehole Stress Concentrations on Dipole AnisotropyMeasurements,” Geophysics 63, no. 1 (January-February1998): 11–17.

7. Valero H-P, Peng L, Yamamoto M, Plona T, Murray D andYamamoto H: “Processing of Monopole Compressional inSlow Formations,” Expanded Abstracts, 74th SEGInternational Meeting, Denver (October 10–15, 2004):318–321.

8. Walsh J, Tagbor K, Plona T, Yamamoto H and De G:“Acoustic Characterization of an Extremely SlowFormation in California,” Transactions of the SPWLA 46thAnnual Logging Symposium, New Orleans, June 26–29,2005, paper U.

> Openhole (left) and cased-hole (right) results in a Statoil North Sea well. The Sonic Scanner tool measures P-, S- and Stoneley-waveslownesses in open hole and behind casing, even where the caliper (Track 1) indicates a washed-out zone (between X,296 and X,305 m)in the openhole logs. Flexural-mode slowness displayed in Track 2 of each set is more sharply defined, with a narrower color band, inthe cased-hole example than in the openhole logs. In the dispersion curves (bottom), compressional-wave slowness is in dashed greenand shear-wave slowness is in dashed blue.

X,290

X,320

X,310

X,300

0 150gAPI

Gamma Ray6 16in.

Bit Size

Dept

h, m

80 540µs/ft

Shear Slowness

µs0 10,000

Arrival Timeµs0 10,000

STC CoherenceWaveform

Variable Density Log

X,290

X,320

X,310

X,300

Dept

h, m

0 150gAPI

Gamma Ray6 16in.

Caliper6 16in.

Bit Size

Washout

80 540µs/ft

Shear Slowness

STC Coherence

µs0 10,000

Arrival Timeµs0 10,000

WaveformVariable Density Log

300

250

200

150

100

50

00 2,000 4,000 6,000 8,000 10,000

Frequency, Hz

Slow

ness

, µs/

ft

300

250

200

150

100

50

0

Ampl

itude

, dB

300

250

200

150

100

50

00 2,000 4,000 6,000 8,000 10,000

Frequency, Hz

Slow

ness

, µs/

ft

300

250

200

150

100

50

0

Ampl

itude

, dB

58171schD05R1.qxp 5/22/06 10:45 AM Page 19

20 Oilfield Review

> Shear-wave slownesses computed from flexural-wave logging in the extremely slow Antelope formation, Cymric field, California. In the diatomite zone,down to 1,500 ft, shear slownesses in Track 3 average 700 µs/ft and approach 900 µs/ft in some intervals. Below that, shear slownesses decrease to about400 µs/ft. The large separation between minimum and maximum offline energy in the depth track indicates anisotropy. Track 1 shows gamma ray (green),hole diameter (yellow), hole azimuth (light blue) and azimuth of the continually rotating tool (dark blue). Azimuth of the fast shear wave, shown in Track 2(red), is relatively constant in the anisotropic zone above 1,500 ft, in spite of continual tool rotation. In addition to fast (red) and slow (blue) shear slownessesestimated from dispersion analysis, Track 3 shows Stoneley-wave slowness (black), P-wave slowness (green curve), and slowness-based (left edge oftrack) and time-based (right edge of track) anisotropies. Track 4 shows the waveforms and time windows used for flexural-wave analysis. Slowness-time-coherence projections for fast and slow shear are shown in Track 5 and Track 7, respectively. Slowness-frequency-analysis (SFA) projections for fast andslow shear are shown in Track 6 and Track 8, respectively. (Modified from Walsh et al, reference 8.)

0 gAPI 150

Gamma Ray5 in. 20

Hole Diameter0 deg 360

Total Azimuth0 deg 360

Hole Azimuth-10 deg 90

Sonde Deviation

0 100

0 100

1,250

1,300

1,350

1,400

1,450

1,500

1,550

1,600

Dept

h, ft

-90 deg 90

Fast Shear Azimuth

Azimuth Uncertainty

200 µs/ft 1,200

∆T-Based Fast Shear200 µs/ft 1,200

∆T-Based Slow Shear

0 %

∆T-Based Anisotropy100

100 %

Time-Based Anisotropy0

200 µs/ft 1,200

50 µs/ft 360

0 µs 30,000

Window Start0 µs 30,000

Window Stop0 µs 30,000

Fast Waveform0 µs 30,000

Slow Waveform

200 µs/ft 1,200

Fast Shear

Coherence

200 µs/ft 1,200

Fast Shear

Energy

200 µs/ft 1,200

Slow Shear

Coherence

200 µs/ft 1,200

Slow Shear

Energy2 4 6 160

Anisotropy Flag, %

Compressional ∆T

Stoneley ∆T

MaximumEnergy

MinimumEnergy

OfflineEnergy

Slowness-Frequency Analysis

Slowness-Frequency Analysis

58171schD05R1.qxp 5/22/06 10:45 AM Page 20

Spring 2006 21

wave slowness exceeds 800 µs/ft in some sections.9

Nine-component vertical seismic profiles andcrossed-dipole sonic logs have detected changinganisotropy magnitude and direction with depthand around the field.10 Knowledge of acousticvelocities and of anisotropy can be important fordesigning fracture stimulations and enhanced oil-recovery operations.

Measurements with the Sonic Scanner toolprovide new insight into the acoustic behavior ofthese complex rocks. Waveforms were recorded inan interval from 972 to 1,650 ft [296 to 503 m] in awell near the crest of the Cymric structure. In thediatomite zone, down to 1,500 ft [457 m], shearslowness estimated from flexural-mode dispersionprocessing is at least as great as that found inearlier logging programs, averaging 700 µs/ft andapproaching 900 µs/ft in some intervals (previouspage). Below that, shear slowness decreases toabout 400 µs/ft in the cristobalite zone.

Much of the logged interval exhibitsazimuthal anisotropy, as indicated by the largeseparation between minimum and maximumoffline energy, and also between the fast and slowshear-wave slownesses. Anisotropy magnituderanges from 4 to 8%, consistent with results ofprevious studies.11 Slowness anisotropy iscalculated by dividing the difference betweenfast and slow shear slownesses by their average.The azimuth of the fast shear direction isbetween N35W and N15W, in general agreementwith previous studies.12

Along with the typical fast and slow shear-slowness curves and slowness-time-coherence(STC) projections seen in many sonic-log plots,displays of Sonic Scanner data include newquality-control tracks showing slowness-frequency analysis (SFA). To create SFA plots, adispersion curve is generated at each depthusing the recorded dipole flexural waveforms(above right).13 The dispersion curve is projectedonto the slowness axis, and this projection is

plotted in a log versus depth presentation,similar to the way an STC projection isconstructed. The estimated slowness log fromdispersive STC processing is overlaid on the SFAprojection, and if the estimated slownessmatches the low-frequency limit of the SFAprojection, the quality of the slowness estimate ishigh. In azimuthally anisotropic formations, SFAprojections may be plotted for both the fast andslow shear directions.

In this extremely slow formation, themonopole source does not excite a compressionalhead wave, but rather a strong leaky-P mode.Compressional slowness must therefore beestimated from dispersive STC processing,analogous to the technique for determiningshear slowness from flexural modes.Compressional slowness is estimated at 192 µs/ft

9. Hatchell PJ, De GS, Winterstein DF and DeMartini DC:“Quantitative Comparison Between a Dipole Log andVSP in Anisotropic Rocks from Cymric Oil Field,California,” Expanded Abstracts, 65th SEG AnnualInternational Meeting, Houston (October 8–13, 1995):13–16.

10. De GS, Winterstein DF, Johnson SJ, Higgs WG andXiao H: “Predicting Natural or Induced FractureAzimuths from Shear-Wave Anisotropy,” paperSPE 50993-PA, SPE Reservoir Evaluation & Engineering 1,no. 4 (August 1998): 311–318.

11. De et al, reference 10.12. Hatchell et al, reference 9.13. Plona T, Kane M, Alford J, Endo T, Walsh J and Murray D:

“Slowness-Frequency Projection Logs: A New QCMethod for Accurate Sonic Slowness Evaluation,”Transactions of the SPWLA 46th Annual LoggingSymposium, New Orleans, June 26–29, 2005, paper T.

> Construction of a slowness-frequency-analysis (SFA) log for controlling the quality of shear-slownessestimation from flexural waves. Dipole flexural waveforms at each depth (top left) are analyzed fortheir slowness at varying frequencies. Resulting data are plotted on a slowness-frequency plot(bottom left), with circle size indicating amount of energy. Energies are color-coded and projectedonto the slowness axis. The color strip is plotted at the appropriate depth to create a log (right). The slowness estimate from dispersive STC processing is plotted as a black curve. If this matches thezero-frequency limit of the SFA projection, the slowness estimate is good.

13

11

9

7

54

1

12

10

23

6

8

0 2,500 5,000 7,500 10,000Time, µs

400

350

300

250

200

150

100

50

00 2,000 4,000 6,000 8,000

Frequency, Hz

Slow

ness

,

Wav

efor

m n

umbe

r

µs/

ft

0

50

100

300

250

200

150

Ampl

itude

, dB

µs/ft100 400

Shear Slowness

SFA Energy

15,000

15,050

15,100

Waveforms from 15,061 ft

58171schD05R1.qxp 5/24/06 2:25 PM Page 21

in the shallow diatomite section and at 175 µs/ftin the cristobalite section (top).

Following on the initial success of the SonicScanner tool, Chevron is planning to run the toolin more wells in this field in 2006. Sonicvelocities will support microseismic fracture-mapping techniques.14

Radial Profiles of Slowness VariationVariations in formation properties may benatural or induced by the drilling process, andmay be beneficial or detrimental to the E&Pactivity at hand. By fully characterizing P- and S-wave slownesses in a large volume around theborehole, the cause of the variation can be

understood, and decisions can be maderegarding how to take advantage of or mitigatethe situation.

In a recent exploration well in the SouthTimbalier area of the Gulf of Mexico, Chevronsuccessfully penetrated a target sandstone. Inother wells, the same formation had presentedcompletion challenges, so the logging program inthis well included measurements to assess itsmechanical properties.

Radial profiles of compressional and shearslownesses can reveal important informationabout the state of the formation near theborehole. Radial variation in compressionalslowness is revealed by examining the differencein P slowness detected by the receiver array fromthe near and far monopole transmitters. Raysfrom the near transmitter sample the alteredzone near the borehole, while rays from the fartransmitter sample the unaltered zone, alsocalled the far field.

A clear picture of radial variation emergeswhen the P-wave data from all three transmittersand 13 receivers undergo tomographic recon-struction.15 This inversion technique uses ray

22 Oilfield Review

> Dispersion curves for compressional arrivals in the upper, diatomite zone (left) and the lower,cristobalite zone (right). Compressional-wave slowness is estimated by the slowness of the leaky-P mode at low frequency. [Modified from Walsh et al, reference 8].

220

210

200

190

180

170

160

150

Slow

ness

, µs/

ft

Ampl

itude

, dB

300

250

200

150

100

50

00 2,000 4,000 6,000 8,000 10,000

Frequency, Hz

Depth = 1,470 ft

Compressional slownessfrom monopole leaky P

220

210

200

190

180

170

160

150

Slow

ness

, µs/

ft

Ampl

itude

, dB

300

250

200

150

100

50

00 2,000 4,000 6,000 8,000 10,000

Frequency, Hz

Depth = 1,510 ft

Compressional slownessfrom monopole leaky P

> Estimating compressional slowness by processing leaky-P dispersion data in the slow Antelope formation (left). Traditional monopole processing inTrack 2 does not give slowness estimates as reliably as does dispersive STC processing (Track 3). STC plots (right) from two different depths show theimproved coherence delivered by dispersive STC processing (right) compared with traditional STC processing (left). Track 4 shows slowness-frequencyanalysis (SFA) using leaky-P dispersion data, such as those shown in the dispersion curves (below). (Modified from Walsh et al, reference 8.)

225

200

175

150

Slow

ness

, µs/

ft

1,000 2,000 3,000 4,000Time, µs

5,000 6,000

Dispersive STC Processing3.5 kHz to 6.5 kHz

Depth = 1,470 ft

225

200

175

150

Slow

ness

, µs/

ft

1,000 2,000 3,000 4,000Time, µs

5,000 6,000

Depth = 1,510 ft225

200

175

150

Slow

ness

, µs/

ft

1,000 2,000 3,000 4,000Time, µs

5,000

Depth = 1,510 ft

225

200

175

150

1,000 2,000 3,000 4,000Time, µs

Slow

ness

, µs/

ft

Traditional STC Processing

Depth = 1,470 ft

1,470

1,480

1,490

1,500

1,510

4 14in.

Coherence Coherence EnergyWaveformAmplitude

140 240µs/ft140 240µs/ft 140 240µs/ft

CompressionalSlownessMonopoleProcessing

Caliper4 14in.

Bit Size0 150gAPI

Gamma Ray

Washout

Mea

sure

d de

pth,

ft

CompressionalSlownessLeaky-P

Processing

Slowness-Frequency Analysis

2,000 µs 4,000

Time

58171schD05R1.qxp 5/22/06 10:46 AM Page 22

Spring 2006 23

tracing to calculate signal arrival times at all thesensors, and updates an initially homogeneousformation model to create a final model thatsatisfies the observed data. To visualize theresulting compressional-slowness radial profile,the differential percentage between observedslowness and far-field slowness is color-codedand plotted against radial distance from theborehole wall (right).

Data from this Chevron well showed that thesandstones of interest exhibited radial variationsin compressional slowness approaching 15% atthe borehole wall and extending up 1 ft [30 cm]into the formation. However, quantifying theradial P-wave slowness variation alone does notidentify its cause. Compressional-slownessvariations can be caused by fluid changes, suchas invasion of drilling fluid, or by radial changesin stress or formation strength. Additionalinformation from the shear-slowness radialprofile could help distinguish between these.

Shear-slowness radial profiles are constructedin a multistep procedure.16 Semblance processingof flexural waveforms at low frequencies providesan initial estimate of formation elastic param-eters. These parameters are used to model ahomogeneous isotropic formation. Differencesbetween measured and modeled slownesses at alarge selection of frequencies form the input toan inversion procedure that yields the actualflexural-slowness radial profile. The results areplotted with colors that represent the amount ofdifference between observed slowness and theslowness of the unaltered, far-field formation.

In the South Timbalier case, the shear-slowness radial profile shows a large differencein near-wellbore slowness compared with far-field slowness. Flexural-wave dispersion curves

14. Bennett L, Le Calvez J, Sarver DR, Tanner K, Birk WS,Waters G, Drew J, Primiero P, Eisner L, Jones R, Leslie D,Williams MJ, Govenlock J, Klem RC and Tezuka K: “The Source for Hydraulic Fracture Characterization,”Oilfield Review 17, no. 4 (Winter 2005/2006): 42–57.

15. Zeroug S, Valero H-P and Bose S: “Monopole RadialProfiling of Compressional Slowness,” prepared forpresentation at the 76th SEG Annual InternationalMeeting, New Orleans, October 1–3, 2006.Hornby BE: “Tomographic Reconstruction of Near-Borehole Slowness Using Refracted Sonic Arrivals,”Geophysics 58, no. 12 (December 1993): 1726–1738.

16. Sinha BK: “Near-Wellbore Characterization Using RadialProfiles of Shear Slowness,” Expanded Abstracts,74th SEG Annual International Meeting, Denver (October 10–13, 2004): 326–331.

> Compressional and shear radial profiling in a Chevron Gulf of Mexico well. P-wave data from allthree transmitters and 13 receivers are input to tomographic reconstruction based on tracing raysthrough a modeled formation with properties that vary gradually away from the borehole. Thepercentage difference between observed compressional slowness and slowness of the unaltered,far-field formation is plotted on color and distance scales to indicate the extent of difference awayfrom the borehole (Track 6). In these sandstones, compressional slowness near the borehole variesby up to 15% from far-field slowness, and the variation extends to 1 ft from the borehole wall. Shear-wave radial profiles appear in Tracks 3 and 5 for the fast and slow shear differences from far-fieldslowness, respectively. Large differences, attributed to plastic yielding in the near-wellbore region,are shown in red, and extend out to about 10 in. from the borehole wall. These differences occur onlyin the sandstone intervals, identifiable from the gamma ray log in Track 4.

X,480

X,490

X,500

X,510

X,520

X,530

X,540

X,550

1.7 2.7g/cm3

Density9 19in.

Hole Diameter

300 100µs/ft

Fast Shear

Slow Shear

300 100

Compressional

180 80

µs/ft

µs/ft

Mea

sure

d de

pth,

ft

2 0ft

Distance fromBorehole Center

0 2ft

Distance fromBorehole Center

0 2ft

Distance fromBorehole Center

10 110gAPI

GammaRay

Fast ShearDifferential

0 25%

Slow ShearDifferential

0 25%

CompressionalDifferential

0 25%

58171schD05R1.qxp 5/22/06 10:46 AM Page 23

also indicate a high degree of near-wellborealteration (above). The analysis is complicatedsomewhat by the addition of anisotropy; the fastshear and slow shear waves exhibit distinctdifferences relative to the unaltered, far-fieldslowness. In the sandstones, both fast and slowshear slownesses are up to 20% greater than thefar-field slowness in a zone roughly 10 in. [25 cm]from the borehole wall.

The radial heterogeneity in shear slownessrules out invasion or other fluid-related causes ofnear-wellbore alteration, because shear wavesare almost insensitive to changes in pore fluid.Fluid-related changes would cause onlycompressional-slowness radial variation. Themeasurable radial variation in shear slownessindicates that the formation has undergonemechanical damage in the form of plastic

yielding of grain contacts. The caliper shows nowellbore enlargement through this zone, so thedamaged material has not yet fallen into theborehole, but the increase in shear slowness nearthe borehole wall indicates that it is near failure.The Sonic Scanner data indicate a wide zone ofdamage that will require extra care when thetime comes to design a well completion.

Compressional and shear radial profiles bringnew information not previously available fromany logging tool. Borehole imaging tools andcalipers have been able to deliver images orevidence of drilling-induced borehole irregular-ities such as breakouts and fractures, but areuseful only after the borehole shape haschanged. The Sonic Scanner tool probes deepinto the formation to reveal mechanical damagebeyond the borehole wall.

Radial profiling may also help to fine-tuneprograms for acquisition of fluid samples. In anexample from the North Sea, Sonic Scannercompressional radial profiles were computed fortwo intervals from which samples weresubsequently acquired using the MDT ModularFormation Dynamics Tester. Zone A showed littledifference between near-wellbore and far-fieldslowness (next page). Two fluid samples weretaken from this interval after pumping times of75 and 80 minutes and no sampling problems. InZone B, the radial profile indicated formationdamage out to 12 in. from the borehole wall.During the attempt to obtain a fluid sample, theprobe on the sampling tool became plugged, andno sample was obtained.

Formation damage does not necessarily meanthat samples cannot be acquired, but sampling inthese zones may have an increased risk of toolplugging or sticking. To minimize these risks,sampling from potentially damaged zones shouldbe delayed and attempted later in the samplingprogram, so that samples from other intervalscan be collected first with less risk.

Characterizing Permeable Zones and FracturesPetrophysicists and reservoir engineers have longsought a continuous measurement of permeabilityto optimize well completions and productionscenarios, but continuous permeability is one ofthe most difficult properties to measure in an oilwell. Using empirical relationships calibrated tocore measurements, permeability or mobility—the ratio of permeability to viscosity—can beinferred from other measurements such asporosity or nuclear magnetic resonance logs.

24 Oilfield Review

17. Brie A, Endo T, Johnson DL and Pampuri F: “QuantitativeFormation Permeability Evaluation from StoneleyWaves,” paper SPE 49131, presented at the SPE AnnualTechnical Conference and Exhibition, New Orleans,September 27–30, 1998.

18. Kimball CV and Endo T: “Quantitative Stoneley MobilityInversion,” Expanded Abstracts, 68th SEG AnnualInternational Meeting and Exhibition, New Orleans(September 13–15, 1998): 252–255.Liu H-L and Johnson DL: “Effects of an ElasticMembrane on Tube Waves in Permeable Formations,”Journal of the Acoustic Society of America 101, no. 6(June 1997): 3322–3329.

> Comparison of flexural-wave dispersion seen in a South Timbalier wellwith modeled results (top). Observed flexural-wave slownesses (red andblue circles) show much larger dispersion than the model for a homogeneousisotropic formation (blue curve). The large difference at higher frequenciesindicates near-wellbore damage. Stoneley-wave slownesses appear asgreen circles. In the bottom figure, the difference between observed andmodeled flexural slowness is plotted against distance, in borehole-radius ratiounits. The difference between observed and modeled flexural slownessamounts to 20% out to a distance equivalent to about two borehole radii.

Alteration radius/borehole radius

400

500

600

700

800

~20% shear alteration

Slow

ness

, µs

/ft

Slow

ness

, µs

/ft

Frequency, Hz

1,000 2,0000

1 2 3 4 5 6 7

3,000 4,000 5,000

400

500

600

700

800

58171schD05R1.qxp.24.ps 7/5/06 5:07 PM Page 24

Spring 2006 25

Direct measurements can be made with wirelineformation testers at isolated points along the well,or on core, but these require additional loggingruns and coring costs. Stoneley-wave analysis is apowerful technique that delivers a direct,continuous measurement of mobility along the well.17

The idea of measuring mobility from theStoneley wave was first expressed in the 1970s,but proved difficult in practice. Many attemptshave been made to develop empirical correla-tions between permeability and Stoneleyattenuation, but these methods requiredcalibration with other information and neglectedseveral important factors, such as mudcakepermeability and the presence of the tool itself.Approaches that simplified the complex behaviorof Stoneley waves were seldom successful, but aninversion method that uses a model derived fromfull Biot poroelastic theory reliably determinespore-fluid mobility from Stoneley waveforms.18

For application with Sonic Scanner data, the fullBiot inversion technique was extended toincorporate tool response.

The full Biot inversion scheme requiresseveral borehole, mudcake and formationparameters to evaluate fluid mobility usingStoneley-wave data. The list includes: holediameter; mud slowness, attenuation anddensity; formation P and S slowness, density andporosity; grain modulus; pore-fluid modulus anddensity; and mudcake density, bulk modulus,shear modulus, thickness and membranestiffness. The computation outputs fluid mobilityand associated error ranges.

This inversion technique has been availablefor several years, but application has not alwaysbeen successful because the inversion requiresextremely low-frequency Stoneley waves—downto 300 Hz. Data with this frequency content havenot been available in the past, because earliersonic tools interacted negatively with low-frequency signals and required filtering toremove frequencies below 1,500 Hz. Now, thewideband sources of the Sonic Scanner toolgenerate strong Stoneley waves with reliable low-frequency content for mobility calculations.

An example from a Statoil well in theHaltenbanken area of the Norwegian Sea showsgood correlation between mobilities calculatedfrom Stoneley waves and those measured by MDT pretests. Input values of formation and fluid properties of a zone near the oil/watercontact were determined with logs from thePlatform Express integrated wireline logging

> A compressional-wave radial profile indicating intervals of successful andrisky fluid sampling. In interval A, the compressional-wave radial profile(Track 3) shows a small differential between near-wellbore slowness and far-field slowness. There is little near-wellbore alteration in the zone wherethe MDT Modular Dynamics Formation Tester successfully collected twoformation-fluid samples. In Track 3, the amount of slowness differencebetween near and far field is indicated by gold and brown color intensity,while depth of alteration is indicated by the horizontal length of the coloredarea. In interval B, the compressional-wave radial profile shows darkercolors, indicating a higher degree of near-wellbore alteration extendingfarther away from the borehole. In this zone, the MDT probe becameplugged and was not able to collect any formation-fluid samples. Track 2displays the slowness gradient obtained from the tomographic reconstruction.The gradient indicates the difference in slowness from one slowness-model cell to the next, moving away from the borehole in small increments.

Gamma Ray0 100gAPI

Caliper6 16in.

Positionof MDT Tool

A

B

CompressionalSlowness Gradient

0 % 10

Slowness Differential

0 % 10

Distance fromBorehole Center

0 ft 3

Distance fromBorehole Center

0 ft 3

58171schD05R1.qxp 5/22/06 10:47 AM Page 25

tool. The MDT results from eight drawdownpretests and one tight pretest correlate closelywith mobilities extracted from Stoneley-waveanalysis (right).

The continuous mobility log exhibits highmobility inside sand packages and low mobilitynear shale streaks and at the depth of the tightMDT pretest. Because the Sonic Scannermobility results are somewhat sensitive to a fewparameters that are not well constrained bylogging measurements, such as mud slowness,mud attenuation and mudcake stiffness, testswere conducted to study the effect of uncertaintyin these parameters on the mobility error bars.The continuous mobility log shown is the onewith the least uncertainty.

When the borehole is in good condition,continuous mobility logs from Stoneley wavescan be used to obtain a quick permeabilityestimate for selecting sampling points andperforation intervals, and may function as asupplement to core or formation-tester permea-bility points over an extended interval.

Stoneley waves can also be used tocharacterize permeability associated with openfractures. In the US Rocky Mountains, forexample, hard-rock reservoirs depend onhydraulically induced fractures for economicproduction. However, the highly unequal in-situstresses in the region give rise to natural fracturestoo. If natural fractures are encountered in a well,cementing and stimulation designs must beadjusted to prevent cement from entering thenatural-fracture system. For example, fiber-basedtreatments for both cementing and stimulationcan be used to reduce fluid losses.19 Stimulationprograms need to take into account themagnitude and direction of the principal stresses.Optimizing the completion design requiresknowledge of the fracture and stresscharacteristics around the wellbore and in the formation.

An open fracture intersecting a boreholecauses Stoneley waves to reflect and attenuate.20

Analysis of Stoneley waveforms quantifies thesechanges, which are input to an inversion forfracture aperture.21 However, washouts, boreholerugosity and abrupt changes in lithology also cancause Stoneley reflections, and should beconsidered in the analysis.22

An example of successful application of thismethod comes from Colorado, USA.23 In this gasreservoir, porosity ranges from 3 to 7% andpermeability is in the microdarcies. Stoneley-wave analysis quantified the aperture andpermeability of fractures that were also seen on

26 Oilfield Review

> Comparing fluid-mobility values from MDT pretests with those fromStoneley-wave processing in a Statoil well in the Haltenbanken area of theNorwegian Sea. In Track 3, continuous fluid-mobility values (blue curve)and uncertainties (gray shading) estimated from Stoneley-wave analysiscorrelate well with discrete mobility values obtained from MDT drawdownpretests (red dots). The two measures of mobility match even at the tightMDT pretest at X,X42.15 m, where the Stoneley mobility also shows anextremely low value. Porosity, gamma ray, density, caliper and shale volumeare plotted in Track 1. Track 2 shows acoustic slownesses. Track 4 displaysrelative volumes of lithology and fluids.

Mobility Error Bar

Coal

Oil

Bound Water

Sand

1.96 2.96

Density

X,X00

X,X50

g/cm3

0 1

Shale Volume6 16

Caliperin.

0 150

Gamma RaygAPI

1 0

Porositym3/m3

240 40

Mud Slownessµs/ft

300 200

Reconstructed Stoneley

300 200

Stoneley Slowness300 100

Shear Slowness300 0

Compressional ∆T

µs/ft

µs/ft

µs/ft

µs/ft

1 10,000

1 10,000

1 10,000

Stoneley Mobility

Mobility Error

MDT Mobility

mD/cP

Water

Shale

m3/m3 mD/cP

mD/cP

Dept

h, m

58171schD05R1.qxp 5/22/06 10:47 AM Page 26

Spring 2006 27

FMI Fullbore Formation MicroImager images(above). With the broad range of Stoneley-modefrequencies acquired by the Sonic Scanner tool, these open natural fractures can be reliably characterized.

Shear-Wave Directions in MexicoSmall directional variations in formationproperties can have a major impact on drillingand completion strategies, but these may bedifficult to measure. For example, sonic velocities

may be different in one horizontal directioncompared with the orthogonal horizontaldirection. This phenomenon, called elasticanisotropy, occurs in most sedimentary rocks andis caused by layering, aligned fractures or stress

19. Bivins CH, Boney C, Fredd C, Lassek J, Sullivan P,Engels J, Fielder EO, Gorham T, Judd T,Sanchez Mogollon AE, Tabor L, Valenzuela Muñoz A and Willberg D: “New Fibers for Hydraulic Fracturing,”Oilfield Review 17, no. 2 (Summer 2005): 34–43.Abbas R, Jaroug H, Dole S, Effendhy, Junaidi H, El-Hassan H, Francis L, Hornsby L, McCraith S,Shuttleworth N, van der Plas K, Messier E, Munk T,Nødland N, Svendsen RK, Therond E and Taoutaou S: “A Safety Net for Controlling Lost Circulation,” OilfieldReview 15, no. 4 (Winter 2003/2004): 20–27.

20. Hornby BE, Johnson DL, Winkler KH and Plumb RA:“Fracture Evaluation Using Reflected Stoneley WaveArrivals,” Geophysics 54, no. 10 (October 1989):1274–1288.Brie A, Hsu K and Eckersley C: “Using the StoneleyNormalized Differential Energies for Fractured ReservoirEvaluation,” Transactions of the SPWLA 29th AnnualLogging Symposium, San Antonio, Texas, June 5–8, 1988,paper XX.

21. Endo T, Tezuka K, Fukushima T, Brie A, Mikada H andMiyairi M: “Fracture Evaluation from Inversion ofStoneley Transmission and Reflections,” Proceedings of the 4th SEGJ International Symposium, Tokyo(December 10–12, 1998): 389–394.

22. Tezuka K, Cheng CH and Tang XM: “Modeling of Low-Frequency Stoneley-Wave Propagation in an IrregularBorehole,” Geophysics 62, no. 4 (July-August 1997):1047–1058.

23. Donald A and Bratton T: “Advancements in AcousticTechniques for Evaluating Open Natural Fractures,”prepared for presentation at the SPWLA 47th AnnualLogging Symposium, Veracruz, Mexico, June 4–7, 2006.

> Identifying permeable fractures in Colorado using Stoneley waves. The fracture aperture, or amountof opening, computed from Stoneley-wave reflection and transmission is displayed in Track 2. Track 3shows fracture permeability computed from the Track 2 apertures. Zones containing permeablefractures correlate with zones in which the FMI logs (Track 6) show fractures. The same zones appearas anisotropic, with large offline energy differences (depth track), and also show large differencesbetween measured Stoneley slowness and slowness modeled for an elastic, impermeable formation(orange shading, Track 1). Track 4 shows measured Stoneley waveforms, with amplitude reduction inthe fractured zones. Track 5 shows waveforms generated by the Tezuka model of reference 22.(Modified from Donald and Bratton, reference 23.)

X,100

X,200

X,300

X,400

0 100

0 100

MaximumEnergy

MinimumEnergy

Dept

h, ft

Modeled Stoneleyµs/ft250 150

S-Se

Washout

Caliperin.4 14

Bit Sizein.4 14

∆T Stoneley

µs/ft250 150

Stoneley Aperture

Fracture Widthin.0 0.5

Stoneley Permeability

Fracture Trace Lengthµs/ft10 0

Fracture Porosityft3/ft30.1 0

Fracture PermeabilitymD100,000 1010

µs0 20,440 µs0 20,440

Stoneley VariableDensity Log Tezuka Model

0 120 240 360

Resistive ConductiveFMI Image

OfflineEnergy

Orientation North

58171schD05R1.qxp 5/22/06 10:47 AM Page 27

imbalance.24 Until now, wireline sonic tools havebeen able to quantify the magnitude andorientation of elastic anisotropy only if thedifference in velocities was at least 5%. The highquality of data provided by the Sonic Scannertool allows reliable measurement of anisotropyas small as 1%, and also helps interpretersdetermine the cause of the anisotropy.

Pemex Exploración y Producción wanted toevaluate the amount and direction of anisotropyin tight gas-producing sandstone formations inthe Burgos basin of northern Mexico. Theseformations have low permeabilities and must bestimulated to produce gas in commercialquantities. Optimal development depends oncorrectly orienting hydraulic fractures in thelocal stress field so that each vertical well drains

its designated volume. Knowledge of elasticanisotropy orientation and magnitude would helpin the design and application of oriented-perforating techniques prior to fracturetreatments and would also improve the success ofinfill-drilling campaigns.25

28 Oilfield Review

> A crossed-dipole log (left) from the Pemex Cuitlahuac-832 well, showing zones with isotropy and with differing amountsof anisotropy. Zone A, an isotropic zone, has low offline energy (depth track) and equal fast and slow shear-waveslownesses (Track 3). Anisotropic Zones B and C have nonzero offline energies and different fast and slow shear-waveslownesses. Anisotropy magnitude, either slowness-based or time-based (edges of Track 3), is about 8% in Zone B andabout 2% in Zone C. The azimuth of the fast shear wave (Track 2) remains constant through the anisotropic intervals,even though the tool is rotating (Track 1), giving confidence in the anisotropy values. Dispersion curves from the threeintervals (right) show characteristic relationships. In Zone A (top), as in other isotropic formations, the dispersion curvesfor flexural waves recorded in the two dipole directions (red and blue circles) overlie each other. At the bottom of ZoneB (bottom), the dispersion curves cross each other. The flexural wave that is fast near the borehole, at low frequencies(red dots), becomes the slower wave with distance from the borehole (blue dots). This indicates that stress-inducedanisotropy is the dominant mechanism of anisotropy in this section. Shallower in Zone B (middle), the dispersion curveslook as though they could cross, but the high-frequency components of the fast shear wave are lost. At this depth, open,induced fractures were visible in OBMI Oil-Base MicroImager logs. (Modified from Wielemaker et al, reference 25.)

Slow

ness

, µs

/ft

4,0002,0000 6,000 8,000

150

200

250

300

350

50

100

150

200

250

300

Ampl

itude

, dB

Frequency, Hz

Depth = 1,593.04 m

Depth = 1,658.87 m

Depth = 1,665.27 m

Slow

ness

, µs

/ft

4,0002,0000 6,000 8,000

150

200

250

300

350

50

100

150

200

250

300

Ampl

itude

, dB

Frequency, Hz

Slow

ness

, µs

/ft

4,0002,0000 6,000 8,000

150

200

250

300

350

50

100

150

200

250

300

Ampl

itude

, dB

Frequency, Hz

1,600

Depth,m

1,650

0 gAPI 150

Gamma Ray

Hole Diameter

0 deg 360

5 in. 20

Total Azimuth

0 deg 360

Hole Azimuth

-10 deg 90

Sonde Deviation

0 100

0 100

-90 deg 90

Fast Shear Azimuth

Azimuth Uncertainty 350 µs/ft 1,200

350

0

0 20%

2 4 6 16

µs/ft 1,200

Fast Shear ∆T

Slow Shear ∆T

∆T-Based Anisotropy

20 %

Time-Based Anisotropy

0

Anisotropy Flag, %

MaximumEnergy

MinimumEnergy

OfflineEnergy

A

B

C

58171schD05R1.qxp 5/22/06 10:48 AM Page 28

Spring 2006 29

When the vertical stress is the maximumstress, hydraulic fractures propagate in thedirection of the maximum horizontal stress andthey open in the direction of minimum horizontalstress. Shear waves travel fastest when polarizedin the direction of maximum horizontal stress(SH) and slowest when polarized in the directionof minimum horizontal stress (Sh). This isbecause additional stress stiffens the formation,increasing velocity, and reduced stress converselydecreases velocity. Measuring the direction ofthe fast shear waves yields the preferreddirection of fracture propagation.

The directions, or azimuths, of fast and slowshear waves can be seen in a crossed-dipole log.A crossed-dipole log from the Cuitlahuac-832well shows both isotropic and anisotropic zones(previous page). Zone A, an isotropic zone, isidentified by near-zero offline energies and equal fast and slow shear-wave slownesses.26

Anisotropic Zones B and C are identified bynonzero offline energies and diverging fast andslow shear-wave slownesses.

The two anisotropic zones have differentamounts of anisotropy. In Zone B, anisotropymagnitude is about 8%. In Zone C, the amount ofanisotropy is about 2%. Although 2% is lower thanhas been reliably detected by other tools,interpreters have confidence in the valuebecause the waveforms are so clear and becausethe fast shear azimuth remains constant,between 30° and 40° over the interval, even withthe tool continually rotating.

Knowing the magnitude and azimuth ofanisotropy is vital, but this does not identify thecause. The anisotropy may be intrinsic to the rockor may be stress-induced; identifying the cause isimportant for understanding how stable thedrilling process will be and how a borehole willrespond to stress. Usually, additional information,such as borehole images or core analysis, isrequired to pinpoint the cause of anisotropy.

Analysis of flexural-wave dispersion curvesprovided by the Sonic Scanner tool helps toidentify anisotropy mechanisms at three depthsin the Cuitlahuac-832 well using only sonicmeasurements. Dispersion curves at 1,593.04 m,within Zone A, overlie each other closely andmatch the model for a homogeneous isotropicformation. Curves from 1,665.27 m, one of themost anisotropic intervals near the bottom ofZone B, show the crossover characteristic ofstress-induced anisotropy. Slightly shallower, at1,658.87 m, the fast and slow shear dispersioncurves are separated at low frequencies, but the

high-frequency data are missing, so it isimpossible to determine the curve trend or theanisotropy type. OBMI Oil-Base MicroImagerimages at this depth indicate the presence ofopen, induced fractures, which are the likelycause of the loss of high-frequency data and also strongly suggest stress-induced anisotropy.The 45° azimuth of fractures seen in OBMIimages correlates well with the 40° azimuth ofmaximum horizontal stress inferred from the fastshear direction.

In the Burgos basin, maximum horizontalstress has traditionally been taken to be parallelto the strike of the nearest faults. The resultsfrom Sonic Scanner logging in five wells in thisbasin indicate that local stress direction can vary significantly—up to 20° from the strike ofnearby faults—accentuating the importance ofmaking localized sonic measurements beforedesigning perforation, stimulation and infill-drilling operations.

Imaging Well Beyond the WellboreThe superior quality of waveforms acquired withthe Sonic Scanner tool allows for improvedimaging away from the borehole. Sonic imaginguses reflected P-waves to detect reflectors that are subparallel or at low angle relative to the borehole.

Norsk Hydro has used the imaging capabilityof the Sonic Scanner tool in a highly deviatedwell in the Norwegian Sea (above). Followingacquisition of standard sonic waveforms in oneTLC Tough Logging Conditions wireline pass, aseparate TLC imaging pass recorded waveformsevery 0.5 ft [15 cm] from the three monopolesources firing sequentially to the 104 receivers

24. Elastic anisotropy is sometimes called acousticanisotropy or velocity anisotropy. It can be expressed interms of a difference of velocities, slownesses, stressesor elastic parameters.Armstrong P, Ireson D, Chmela B, Dodds K, Esmersoy C,Miller D, Hornby B, Sayers C, Schoenberg M, Leaney Sand Lynn H: “The Promise of Elastic Anisotropy,” OilfieldReview 6, no. 4 (October 1994): 36–56

25. Arroyo Franco JL, Gonzalez de la Torre H, Mercado Ortiz MA, Weilemaker E, Plona TJ,Saldungaray P and Mikhaltzeva I: “Using Shear-WaveAnisotropy to Optimize Reservoir Drainage and ImproveProduction in Low-Permeability Formations in the Northof Mexico,” paper SPE 96808, presented at the SPEAnnual Conference and Technical Exhibition, Dallas,October 9–12, 2005.Wielemaker E, Saldungaray P, Sanguinetti M, Plona T,Yamamoto H, Arroyo JL and Mercado Ortiz MA: “Shear-Wave Anisotropy Evaluation in Mexico’s Cuitlahuac FieldUsing a New Modular Sonic Tool,” Transactions of theSPWLA 46th Annual Logging Symposium, New Orleans,June 26–29, 2005, paper V.

26. The difference between slownesses is called slownessanisotropy, and the difference between arrival times iscalled time-based anisotropy.

> Geologic cross section with trajectory of a deviated well in which Norsk Hydro acquiredSonic Scanner imaging data. The high deviation angle required TLC Tough LoggingConditions wireline logging.

Wellbore

Interval logged

with Sonic Scanner tool

58171schD05R1.qxp 5/22/06 10:48 AM Page 29

over a distance of 330 m [1,100 ft]. Waveformsfrom each source were processed in a sequencethat started with separating reflected P-wavesfrom Stoneley and refracted P-waves. Theazimuthal distribution of sensors at eachreceiver station allows identification of thedirection to the reflector. Then, traces from eachreceiver station were depth-migrated usingformation velocities measured by the SonicScanner logs from the earlier logging pass.27 Toaccount for tool rotation and the azimuthaldistribution of sensors, the image at eachreceiver station was reconstructed by depth-shifting and stacking images from eachazimuthal channel. Finally, the depth-migratedimages were stacked. Images were obtainedwithin 48 hours.

The results show a 5-degree dipping eventthat extends at least 13 m [43 ft] into theformation (right). The dip of the event is inagreement with the expected geology at the welllocation. The high-resolution event can becorrelated with a 1-m [3.3-ft] coal bed at thesame depth position indicated by petrophysicallogs (next page). The identification of a 1-m coalbed indicates the potential to obtain high-resolution images from a sonic-imaging survey.The resolution is far better than can be obtainedfrom any surface or borehole seismic survey(below right).

Another potential application of sonicimaging is the detection of vertical fracturesnear but not intersecting vertical boreholes.Current techniques such as borehole imagelogging and fracture identification from Stoneleyreflections work only if a fracture intersects theborehole. In many cases, a vertical well will missvertical fractures. Deep imaging with the SonicScanner tool expands the volume of investigationto enable the identification of features that maydelineate reservoir extent or the state of stressaway from the borehole.

30 Oilfield Review

27. Migration is a data-processing step that aims to sharpen,shift and relocate reflectors to their true locations.

> A gently dipping reflector imaged far from the borehole. The boreholetrajectory is shown in red. The high-resolution event detected by sonicimaging can be seen above and to the right of the borehole, near the centerof the image. The reflector correlates with a coal bed at the same depthposition indicated by petrophysical logs.

Verti

cal d

epth

, ft

X,240

X,260

X,280

X,300

X,320

X,340

X,360

X,380

X,460 X,480 X,500 X,520 X,540 X,560 X,580

Source-receiver midpoint, ft

> Comparing high-resolution sonic-imaging data with a surface seismic survey. The 1-m coal bedresolved by Sonic Scanner imaging (inset) cannot be seen in the surface seismic survey.

Sonic-imaging results on seismic scale

58171schD05R1.qxp 5/22/06 10:49 AM Page 30

Spring 2006 31

Scanning the Horizon The Sonic Scanner tool is a new development,and engineers, geologists and petrophysicists arestill finding new ways to use its data. By addingthe radial dimension and multiple depths ofinvestigation to the well-known axial andazimuthal sonic measurements, the SonicScanner tool performs enhanced characteri-zation of acoustic properties in inhomogeneousand anisotropic formations. With this infor-mation, customers are able to predict howformations and fluids will behave during drilling,stimulation and production.

The innovative tool design with predictableacoustics delivers waveforms of excellent qualityand at a wide frequency range. These capabilitiesallow slowness estimation in extremely slowformations, measuring azimuthal anisotropy assmall as 1 to 2%, and reliable application of low-frequency Stoneley modes for fluid-mobilityestimation and evaluation of natural fractures.Advanced quality control with slowness-frequency analysis adds confidence to slownessestimates obtained by dispersion analysis.

Complete recording of all data frommonopole and dipole sources to 104 receiversdistributed azimuthally around the tool removesuncertainties about formation geometry andstructure and improves through-casing andcement evaluation. Current capabilities obtainonly monopole compressional and shear data incased hole. One area of future advancement willbe to extend current openhole applications to cased wells.

Additional applications will arise as morecompanies gain experience with the SonicScanner tool and the high-quality data itproduces. While it is difficult to predict how therest of the oil and gas industry will evolve, sonic-logging enthusiasts anticipate another 50 years ofinvestigations in and around the borehole. —LS

> Petrophysical logs from the Norsk Hydro well in the Haltenbanken area of the Norwegian Sea,showing the 1-m coal bed delineated by sonic imaging. Platform Express resistivity logs (Track 2) anddensity and porosity logs (Track 3) are input, along with ECS Elemental Capture Spectroscopy data to derive mineralogy (Track 4). Nuclear magnetic resonance data appear in Track 5.

X,900

X,925

Caving

ECS Sigma

cu60 0

Caliper

in .6 16

Depth,m ohm-m0.2 200

Porosity

m3/m30.45 -0.15

Density

g/cm31.95 2.95

Density

g/cm32.5 3.0

Coal

Siderite

T2 Distribution

Log T2 Mean

ms0.3 3,000

ms0.3 3,000

T2 Cutoff

ms0.3 3,000

1- mcoal bed

HRLA1 Resistivity

HRLA2 Resistivity

HRLA3 Resistivity

HRLA4 Resistivity

HRLA5 Resistivity

Quartz

Clay

Carbonate

Pyrite

Gamma Ray

gAPI0 200

58171schD05R1.qxp 5/22/06 10:49 AM Page 31

32 Oilfield Review

In a patent awarded in 1935, ConradSchlumberger specified how a transmitterand two receivers might be used to measurethe speed of sound in a short interval of rockpenetrated by a borehole (right).1 He claimedthat the speed and attenuation of soundwould characterize lithology. His inventionfailed because neither logging engineers nor the technology of the time was able todetect the short time difference—tens ofmicroseconds (µs)—between signals travelingat the speed of sound to receivers separatedby just inches.

During World War II, the necessaryelectronics emerged, making sonic loggingpossible.2 According to one account, the firstoilfield application of sonic logging was forcasing-collar location, in 1946.3 Most otherhistorical accounts state that the first sonicapplications appeared after the 1948experiments by Humble Oil Research,followed by Magnolia Petroleum Company andShell.4 These companies designed devices tocollect sonic-velocity information for time-depth conversion of surface seismic sectionsand for correlating seismic reflections tolithologic interfaces. The tools featured onetransmitter and one or two receiversseparated from the transmitter by isolatingmaterial. By the mid-1950s, service companiesand oil companies were acquiring sonic-logging data to generate synthetic seismogramsfor comparison with surface seismic sections.5

In 1957, having licensed the Humble patent,Schlumberger introduced its first sonic tool,the velocity logging tool (VLT), for improvingseismic interpretation.

The early Magnolia Petroleum paper hadhinted at the additional possibility of usingsonic velocities to determine porosity andlithology, but it was scientists in the researchdivision of Gulf Oil Corporation who firstpublished experimental observationsconfirming the link.6 Within a short time,demand for sonic porosity-logging applicationsovertook that for seismic applications.

In 1960, field crews testing the VLTresponse in cased holes in Venezuela noticedthat certain zones caused unreadable, low-amplitude signals. They correctly concludedthat the anomalous signals could be attributedonly to cement condition. Measuring andrecording signal amplitude in addition toarrival time gave birth to an unexpectedapplication, and CBT Cement Bond Tool logssoon replaced the temperature survey fordetecting top of cement.

By the early 1960s, the first sonic tools hadacquired tens of thousands of logs, andengineers set about designing a second-generation tool to address three problems: tooldurability, and poor signal in the presence of

borehole irregularities and near-wellborealteration. The tool-durability problem arosebecause early tools used rubber to isolatereceivers from transmitters, therebypreventing undesirable sound waves frompropagating within the tool and overwhelmingdesired signals. However, rubber tended toabsorb gas from gas-rich formations, causingthe tool to expand and break apart as it wasbrought to surface. The tool was strengthenedby replacing the rubber with steel, but thenthe tool housing had to be shaped so that thepath of sonic waves traveling through the steelwould be longer than the paths through theformation and back to the receivers. (nextpage). Many modern sonic tools continue to

History of Wireline Sonic Logging

> Illustration from the 1935 patenton acoustic logging by ConradSchlumberger. The field engineer(13) was supposed to slide a sleeve(17) until sound coming fromreceivers (3 and 4) appeared toarrive simultaneously at each ear.

58171schD05R1.qxp 5/25/06 11:02 PM Page 32

Spring 2006 33

feature slots and grooves to slow down thearrival of signals—known as tool arrivals—that travel purely through the tool.

A way around the second problem, poor logs in bad hole, came from the Shell engineerresponsible for that company’s first sonictool.7 His borehole-compensating arrangementof receivers and transmitters not onlyeliminated the problem of poor signal inwashed-out zones, but also removed theeffects of tool tilt and eccentering on logresponse. Solving two of the three problemsthat plagued the earlier tools, Schlumbergerincorporated this idea into the all-steel designof the borehole-compensated (BHC) sonic toolthat was introduced in 1964. The BHC toolcontained two transmitters and four receivers.

Along with BHC technology came theability to view registered waveforms on an

oscilloscope in the logging truck. Appearingon the screen were not only the primary (P-)arrivals, or compressional waves, but alsosecondary (S-), or shear waves and laterarrivals. Recognizing the importance of shearwaves made the mid-1960s a time of intenseactivity in expanding sonic applications.Specialists at Shell proposed using the ratioof P to S velocity as a lithology indicator, andalso used sonic logs to predict overpressuredzones.8 Schlumberger engineers andresearchers evaluated use of P and Samplitudes to locate fractures. Althoughthese and other shear-wave applications hadbeen proposed, the acquisition systems of thetime recorded only the arrival time of the P-wave. The waveform itself, including P, S andlater arrivals, was not recorded.

Another drawback of the BHC tool was itsinability to accurately measure the trueformation interval transit time in zones ofinvasion, shale alteration and drilling-induceddamage. The 3- to 5-ft [0.9- to 1.5-m]

transmitter-receiver (TR) spacing capturedonly waves that propagated in the alteredzone, leaving the unaltered zone away fromthe borehole unexplored. By increasing thespacing to 8 to 12 ft [2.4 to 3.7 m], the LSSLong-Spaced Sonic Tool improved logresponse in altered shales. Sonic velocities ofthe unaltered formation are morerepresentative of the reservoir in its naturalstate and yield synthetic seismograms thatbetter match surface seismic traces.

The long TR spacing also stretched thereceived wavetrain, separating the P-, S- andother waves into recognizable packets ofenergy. Efforts intensified to capture the fullwaveform, leading to the development of toolsthat recorded digital waveforms from an arrayof receivers. The first commercial Schlumbergerversion of this technology, introduced in the1980s, was called the Array-Sonic full-waveformsonic velocity tool. Full-waveform logging gaverise to a host of new processing techniques.

The late 1980s saw research experimentswith a second-generation digital sonic tool.The DSI Dipole Shear Sonic Imager tool hadeight sets of four monopole receivers thatcould function as orthogonal dipole receivers,and carried one monopole source and twoorthogonally oriented dipole sources. Thedipole sources generated flexural waves,allowing characterization of formationanisotropy and shear slowness in slow as wellas fast formations.

Also in the late 1980s, Schlumbergerresearchers tested a variety of multireceiveracoustic tools for their ability to acquire sonicimages—seismic-like images far from theborehole.9 The first commercial sonic-imagingservice was run in 1996, but processing wastime- and personnel-intensive.

In 2005, the Sonic Scanner acousticscanning platform combined manyinnovations of the past and added radialmeasurements to simultaneously probe theformation for near-wellbore and far-fieldslownesses.10 The tool itself is fullycharacterized, with predictable acoustics. The wide frequency range of the monopoleand dipole transmitters delivers excellentwaveform quality in all formation types.

1. Schlumberger C: “Procédé et Appareillage pour laReconnaissance de Terrains Traversés par unSondage.” République Française Brevet d’Inventionnuméro 786,863 (June 17, 1935). Also Doll L: “Methodof and Apparatus for Surveying the FormationsTraversed by a Borehole,” US Patent No. 2,191,119(February 20, 1940) (submitted by the estate of ConradSchlumberger).

2. The terms “sonic” and “acoustic” are usedinterchangeably.

3. Pike B and Duey R: “Logging History Rich withInnovation,” Hart’s E&P (September 2002): 52–55,http://www.spwla.org/about/Logging-history.pdf(accessed April 28, 2006).

4. From Humble Oil: Mounce WD: “Measurement ofAcoustical Properties of Materials,” US PatentNo. 2,200,476 (May 14, 1940).From Magnolia Petroleum Company: Summers GC and Broding RA: “Continuous Velocity Logging,”Geophysics 17, no. 3 (July 1952): 598–614.From Shell: Vogel CB: “A Seismic Velocity LoggingMethod,” Geophysics 17, no. 3 (July 1952): 586–597.Léonardon, reference 1, main text.

5. Breck HR, Schoellhorn SW and Baum RB: “VelocityLogging and Its Geological and GeophysicalApplications,” Bulletin of the American Associationof Petroleum Geologists 41, no. 8 (August 1957):1667–1682.

6. Wyllie MRJ, Gregory AR and Gardner LW: “ElasticWave Velocities in Heterogeneous and PorousMedia,” Geophysics 21, no. 1 (January 1956): 41–70.Tixier MP, Alger RP and Doh CA: “Sonic Logging,”Journal of Petroleum Technology 11, no. 5 (May 1959):106–114.

7. Vogel CB: “Well Logging,” US Patent No. 2,708,485(May 17, 1955).

8. Hottman CE and Johnson RK: “Estimation ofFormation Pressures from Log-Derived ShaleProperties,” Journal of Petroleum Technology 17, no.6 (June 1965): 717–722.

9. Hornby BE: “Imaging of Near-Borehole StructureUsing Full-Waveform Sonic Data,” Geophysics 54, no.6 (June 1989): 747–757.

10. Pistre et al, reference 3, main text.

> A sonic-logging sonde with slots to slowdown tool arrivals.

58171schD05R1.qxp 5/22/06 10:50 AM Page 33

34 Oilfield Review

Borehole Acoustic Waves

Jakob B.U. HaldorsenDavid Linton JohnsonTom PlonaBikash SinhaHenri-Pierre ValeroKenneth Winkler Ridgefield, Connecticut, USA

For help in preparation of this article, thanks to Jeff Alford,Houston, Texas; and Andy Hawthorn and Don Williamson,Sugar Land, Texas.

Borehole acoustic waves may be as simple or as complex as the formations in

which they propagate. An understanding of wave-propagation basics is essential

for appreciation of modern sonic-logging technology.

Everyday sounds come from many sources.Keyboards click, crickets chirp, telephones ringand people laugh. Understanding the informa-tion contained in these sounds is something that most people do without thinking. For most,deciphering the sounds they hear is much moreimportant than knowing what sound waves areand how they travel.

However, for geoscientists and others whomust understand the information contained insound waves traveling in the Earth, it is essentialto know what sound waves are and how theytravel. This article reviews the basic types ofacoustic sources and the sound waves that travelin rocks near a borehole. We also discuss theeffects that variations in rock properties have onacoustic waves.

The acoustic waves recorded by a sonic-logging tool depend on the energy source, thepath they take and the properties of theformation and the borehole. In wireline logging,there are two primary types of sources, monopoleand dipole. A monopole transmitter emits energyequally in every direction away from its center,while a dipole transmitter emits energy in apreferred direction.

From a monopole transmitter located in thecenter of the borehole, a spherical wavefronttravels a short distance through the boreholefluid until it meets the borehole wall. Part of theenergy is reflected back into the borehole, andpart of the energy causes waves to propagate inthe formation (next page, top). The direction ofwave propagation is always perpendicular to the

wavefront. This simple case also assumes theformation is homogeneous and isotropic, andthat the sonic tool itself has no other effect onwave propagation.1

The 3D cylindrical setting of the wellborecomplicates this explanation, which can besimplified by examining a vertical plane throughthe axis of a vertical borehole. In the resulting 2Dsystem, spherical wavefronts become circles andpropagate in one plane. In a 3D world, wave-fronts propagate everywhere outward from thesource and surround the borehole symmetrically.

In the 2D simplification, when the wavefrontin the borehole mud meets the borehole wall, itgenerates three new wavefronts. A reflectedwavefront returns toward the borehole center atspeed Vm. Compressional, P-, and shear, S-, wavesare transmitted, or refracted, through theinterface and travel in the formation at speeds Vp

and Vs, respectively. In this simplest case of ahard, or fast, formation, Vp > Vs > Vm.

Once the refracted P-wave becomes parallelto the borehole wall, it propagates along theborehole-formation interface at speed Vp, fasterthan the reflected borehole-fluid wave.According to Huygens principle, every point onan interface excited by a P-wave acts as asecondary source of P-waves in the borehole aswell as P- and S-waves in the formation. Thecombination of these secondary waves in theborehole creates a new linear wavefront called ahead wave.2 This first head wave in the mud isknown as the compressional head wave, and its

58171schD06R1.qxp.p34.ps 5/22/06 4:44 PM Page 34

Spring 2006 35

arrival at the receivers is recorded as the Parrival. The P-wave takes longer to arrive atreceivers that are farther from the source. Thetime difference between P arrivals divided by thedistance traveled is known as ∆ t , or slowness,and is the reciprocal of speed. This is the mostbasic sonic-logging measurement.3

The P-wave that continues into the formationis known as a body wave, and travels on deeperinto the formation unless a reflector sends itback toward the borehole, at which time it iscalled a reflected P-wave. Standard sonic loggingignores reflected P-waves, but special applica-tions, such as those described near the end ofthis article, take advantage of the extrainformation contained in reflected P-waves.

The behavior of refracted S-waves is similarto that of refracted P-waves. When the refractedS-wave becomes parallel to the borehole wall, itpropagates along the borehole-formationinterface as a shear disturbance at speed Vs andgenerates another head wave in the boreholefluid. Its arrival at the receivers is recorded asthe S-wave. In this way, shear slowness of a fastformation can be measured by a tool surroundedby borehole fluid, even though S-waves cannotpropagate through the fluid.

In cases when the shear-wave speed is lessthan the mud-wave speed—a situation known asa slow formation—the shear wavefront in theformation never forms a right angle with theborehole. No shear head wave develops in thefluid. In both fast and slow formations, an S bodywave continues into the formation.

Another way of visualizing how P and S headwaves and body waves travel near the borehole isthrough ray tracing. Strictly speaking, ray tracingis valid only when the wavelength is muchsmaller than the diameter of the borehole, orwhen the wavefronts can be approximated asplanes rather than spheres or cones. Mostborehole acoustic modes, especially those at lowfrequencies, do not meet these conditions, butray tracing can still be useful for visualization. Aray is simply a line perpendicular to a wavefront,showing the direction of travel. A raypathbetween two points indicates the fastest travelpath. Changes in raypath occur at interfaces andfollow Snell’s law, an equation that relates theangles at which rays travel on either side of aninterface to their wave speeds (right). Amongother things, Snell’s law explains the conditionsunder which head waves form and why none formin slow formations.

Ray tracing is useful for understanding wherewaves travel and for modeling basics of sonic-tooldesign, such as determining the transmitter-receiver (TR) spacing that is required to ensurethat the formation path is faster than the directmud path for typical borehole sizes andformation P and S velocities. This ensures thatthe tool will measure formation properties ratherthan borehole-mud properties. Ray tracing alsohelps describe the relationship between TR

> The first few moments of simplified wavefront propagation from a monopole transmitter in a fluid-filled borehole (blue) and a fast formation (tan). Bothmedia are assumed homogeneous and isotropic. Tool effects are neglected. Time progression is to the right. Numbers in the upper left corner correspondto time in µs after the source has fired. Wavefronts in the mud are black, compressional wavefronts in the formation are blue, and shear wavefronts in theformation are red. The compressional head wave can be seen at 90 µs, and the shear head wave can be seen at 170 µs.

40 70 80 110 17090

Compressionalhead wave

Shearhead wave

>Wavefront reflection and refraction at interfaces,and Snell’s law. θ1 is the angle of incident andreflected P-waves. θ2 is the angle of refracted P-waves. θs is the angle of refracted S-waves.Vm is mud-wave velocity. Vp is P-wave velocity inthe formation, and Vs is S-wave velocity in theformation. When the angle of refraction equals90°, a head wave is created.

Reflected P

Refracted P

Refracted S

Incident P-wave

Source

Mud velocity, Vm P velocity, Vp > Vm

Borehole Formation

S velocity, Vs

=Vm

Sin θ1

Vp

Sin θ2 =Vs

Sin θs

θs

θ1

θ1

θ2

1. A homogeneous formation is one with uniform velocity.In other words, the velocity is independent of location.An isotropic formation is one with velocity independentof direction of propagation.

2. The head wave has a conical wavefront in 3D.3. Slowness typically has units of µs/ft.

58171schD06R1.qxp.p35.ps 5/22/06 4:44 PM Page 35

spacing and near-wellbore altered-zone thick-ness and velocity contrast (above). In addition,ray tracing is used in inversion techniques suchas tomographic reconstruction, which solves forslowness models given arrival-time information.

After the P and S head waves, the next wavesto arrive at the receivers from a monopole sourceare the direct and reflected mud waves. Theseare followed by trapped modes and interfacewaves that owe their existence to the cylindricalnature of the borehole. Trapped modes arisefrom multiple internal reflections inside theborehole. Wavefronts of particular wavelengthsbouncing between the walls of the boreholeinterfere with each other constructively andproduce a series of resonances, or normal modes.Trapped modes are not always seen on logs andmay be affected by borehole condition. In slowformations, trapped modes lose part of theirenergy to the formation in the form of waves that

radiate into the formation. These are called leakymodes, and propagate at speeds between P and Svelocities. Leaky modes are dispersive, meaningtheir different frequency components travel atdifferent speeds.

Stoneley WavesThe last arrivals from a monopole source areinterface, or surface, waves. Surface waves werefirst suggested by Lord Rayleigh in 1885.4 Heinvestigated the response at the planar surface ofan elastic material in contact with a vacuum andfound that a wave propagated along the surfacewith particle motion that decreased in amplitudewith distance from the surface—a propertycalled evanescence. Rayleigh’s findingspredicted the existence of waves that propagatealong the Earth’s surface and give rise to thedevastating shaking caused by earthquakes. The

same effect at a much smaller scale leads to“ground roll” noise in surface seismic surveys.

In 1924, Stoneley looked at waves propa-gating at the interface between two solids andfound a similar type of surface wave.5 Theparticular case corresponding to a fluid-filledborehole, that is, the interface between a solidand a liquid, was described not by Stoneley, butby Scholte.6 The waves traveling at the fluid-borehole interface are nonetheless known asStoneley waves. In other areas of geophysics,such as marine seismic surveys, waves travelingat a fluid-solid interface are called Scholte orScholte-Stoneley waves.7

A Stoneley wave appears in nearly everymonopole sonic log. Its speed is slower than theshear- and mud-wave speeds, and it is slightlydispersive, so different frequencies propagate atdifferent speeds.

The decay of Stoneley-wave amplitude withdistance from the interface is also frequency-dependent; at high frequencies, the amplitudedecays rapidly with distance from the boreholewall. However, at low frequencies—or at wave-

36 Oilfield Review

> Ray tracing using Snell’s law to model raypaths. Here,rays are traced through a formation that has radiallyvarying velocity in a zone of alteration. Velocity is lowernear the borehole and grows larger with distance, asituation that arises when drilling induces near-wellboredamage. Rays traveling to the receivers nearest thetransmitter travel only through the altered zone (darkbrown), and rays traveling to distant receivers sense thevelocity of the unaltered formation (light brown).

Transmitter

Receiverarray

Zone ofalteration

Unalteredformation

> The Stoneley wave, traveling at the interfacebetween the borehole and the formation. TheStoneley wave is dispersive and its particlemotion is symmetric about the borehole axis. Atlow frequencies, the Stoneley wave is sensitiveto formation permeability. Waves traveling pastpermeable fractures and formations lose fluid,and viscous dissipation causes attenuation ofwave amplitude and an increase in waveslowness. At open fractures, Stoneley waves are both reflected and attenuated. Red arrows in the center of the borehole symbolizeStoneley-wave amplitude.

Fracture

Receiver

Permeableformation

Stoneleywave

Transmitter

Condition Effect

Attenuated

Attenuatedand sloweddown

Reflected

58171schD06R1.qxp.36.ps 7/5/06 5:15 PM Page 36

Spring 2006 37

lengths comparable to the borehole diameter—the Stoneley amplitude decays very little withdistance from the borehole wall. At sufficientlylow frequencies, the amplitude is nearly constantfrom one side of the borehole to the other,creating what is known as a tube wave. Anexample of a tube wave is the water-hammereffect that can sometimes be heard in plumbingpipes when flow is suddenly disrupted.

The low-frequency Stoneley wave is sensitiveto formation permeability. When the waveencounters permeable fractures or formations,the fluid vibrates relative to the solid, causingviscous dissipation in these zones, whichattenuates the wave and slows it down (previouspage, right). The reductions in Stoneley-waveenergy level and velocity vary with wavefrequency. Stoneley-wave dispersion data over awide bandwidth of frequencies can be inverted toestimate formation permeability.8 Open fracturescan also cause Stoneley waves to reflect backtoward the transmitter. The ratio of reflected toincident energy correlates with fractureaperture, or openness. This technique for thedetection of permeable fractures works well inhard formations.9

All of the above waves propagate symmetri-cally up and down the borehole, and can bedetected by monopole receivers—typicallyhydrophones. Hydrophones are sensitive topressure changes in the borehole fluid, and haveomnidirectional response, meaning that theyrespond equally to pressure changes from any direction.

Waveforms recorded at a given depth areinitially displayed as a time series from the arrayof receivers (above). In some recordings, the

P-, S- and Stoneley-wave arrival times can beseen clearly, but often, data-processingtechniques are used to pick times accurately. Thedifference in arrival times divided by thedistance between receivers yields the slownessfor each mode. However, in many recordings,high noise levels, bad hole conditions or otherfactors can cause these arrivals to be indistinctor mixed with each other. In such cases, visual orautomated picking of arrival times fails to yieldtrue slownesses.

4. Strutt JW, 3rd Baron Rayleigh: “On Waves PropagatedAlong the Plane Surface of an Elastic Solid,” Proceedingsof the London Mathematical Society 17 (1885): 4. Rayleigh waves on the Earth’s surface have vertical and horizontal components of motion. Other surfacewaves discovered by A.E.H. Love have two horizontalmotion components.

5. Stoneley R: “Elastic Waves at the Surface of Separationof Two Solids,” Proceedings of the Royal Society,Series A 106 (1924): 416–428.

6. Scholte JG: “On the Large Displacements CommonlyRegarded as Caused by Love Waves and SimilarDispersive Surface Waves,” Proceedings of theKoninklijke Nederlanse Akademie van Wetenschappen51 (1948): 533–543.

7. Bohlen T, Kugler S, Klein G and Theilen F: “Case History1.5D Inversion of Lateral Variation of Scholte-WaveDispersion,” Geophysics 69, no. 2 (March–April 2004):330–344.

8. Winkler KW, Liu HL and Johnson DJ: “Permeability andBorehole Stoneley Waves: Comparison BetweenExperiment and Theory,” Geophysics 54, no. 1(January 1989): 66–75.

9. Hornby BE, Johnson DL, Winkler KW and Plumb RA:“Fracture Evaluation Using Reflected Stoneley WaveArrivals,” Geophysics 54, no. 10 (October 1989):1274–1288.

> Typical waveforms from a monopole transmitter in a fast formation, showing compressional, shearand Stoneley waves. The pink dashed lines are arrival times. A sonic-logging tool receiver array isshown at left.

Rece

iver

num

ber

13

11

9

7

5

3

1

Time, µs

1,0000 3,000 5,000

Compressionalwave

Shearwave

Stoneleywave

2

4

6

8

10

12

58171schD06R1.qxp.p37.ps 5/22/06 4:44 PM Page 37

Slownesses can be estimated in a robust waywith minimal human intervention using a signal-processing technique that looks for similarity—known mathematically as semblance, orcoherence—in waveforms across the receiverarray.10 The method starts with an assumedarrival time and slowness for each wave type andsearches the set of waveforms for the time andslowness that maximize coherence. The graph ofcoherence for different values of slowness andtime is called a slowness-time-coherence (STC)plot, from which local maxima of the coherencecontours can be identified (above). Maxima

corresponding to compressional, shear andStoneley slownesses plotted for each depthcreate a slowness log. The two dimensions of anSTC plot are compressed into a single dimensionby projecting the coherence peaks onto theslowness axis. This vertical strip of color-codedcoherences, when plotted horizontally at theappropriate depth, forms an element of an STC-projection log, a standard sonic-logging output.The slowness of each mode is plotted on top ofthe STC projection.

Dipole SourcesSo far, the discussion has focused on wavesgenerated by monopole sources, but for someapplications, a different type of source isrequired. For example, in slow formations, wheremonopole sources cannot excite shear waves, adipole source can be effective. The dipole sourceprimarily excites flexural waves, along withcompressional and shear head waves. The motionof a flexural wave along the borehole can bethought of as similar to the disturbance thattravels up a tree when someone standing on theground shakes the tree trunk. The analogy worksbetter if the tree trunk is fixed at the top and hasconstant diameter.

Typically, a tool designed to generate flexuralwaves will contain two dipole sources orientedorthogonally along the tool X- and Y-axes. Thedipole transmitters are fired separately. First,the X-dipole is fired, and a flexural waveform isrecorded. Then, the Y-dipole is fired, and aseparate measurement is taken. The flexuralwave travels along the borehole in the plane ofthe dipole source that generated it. The particlemotion of the flexural wave is perpendicular tothe direction of wave propagation, similar to S-waves, and flexural-wave slowness is related toS-wave slowness. Extracting S-wave slownessfrom flexural-wave data is a multistep process.

Flexural waves are dispersive, meaning theirslowness varies with frequency (below). In manysets of flexural waveforms, it is possible to seethe wave shape change across the receiver arrayas different frequency components propagate atdifferent speeds. Because the wave shape

38 Oilfield Review

> Flexural-mode waveforms, showing a changein wave shape across the receiver array. In thiscase, the wave shape stretches out in time fromnear receiver (bottom) to far receiver (top). Thechange in wave shape is caused by dispersion.

Wav

efor

m n

umbe

r

1

13

11

9

7

5

3

Time, μs1,000 3,500 6,000

> Slowness-time-coherence (STC) processing for monopole arrivals. Waveformsat a given depth (top left) are scanned over time windows and over a range ofangles—called moveouts, which are related to slowness. When the signals onthe waveforms within the window are best correlated, coherence is maximum.An STC plot for that depth (bottom left) displays color-coded coherence in theslowness-time plane, with maximum coherence in red. The coherence valuesare projected onto a vertical strip along the slowness axis and then displayedas a thin horizontal strip at the appropriate depth on the STC projection log(right). A slowness log for each wave is generated by joining the coherencemaxima at all depths.

Compressionalwave

Shearwave

Stoneleywave

13

11

9

7

54

1

12

10

23

6

8

Wav

efor

m n

umbe

r

1,000 2,000 3,000 5,000Time, µs

300

200

100

2,000 3,000 4,000 5,000

Slow

ness

, µs/

ft

Waveforms from 3,764.89 ft

4,000

1,000

Time, µs

µs/ft40 340

Slowness

STC Coherence

3,760

3,770

58171schD06R1.qxp.p38.ps 5/22/06 4:44 PM Page 38

Spring 2006 39

changes across the receiver array, standardmethods for estimating slowness, such as STCprocessing, which relies on wave-shapesimilarity, must be adapted to handle dispersivewaves. Dispersive STC processing identifies theslowness of individual frequency components.11

A plot of flexural-wave slowness versusfrequency is called a dispersion curve (below).Dispersion-curve analysis compares modeledacoustic dispersion curves for homogeneousisotropic formations with curves measured byborehole sonic tools.12

The radial depth of investigation of flexuralwaves is approximately one wavelength. Low-frequency flexural waves probe deep into theformation, and high-frequency flexural waveshave shallower depths of investigation. Analysisof flexural-mode slowness as a function offrequency can therefore provide detailedinformation about the formation near and farfrom the borehole.

At zero frequency, flexural-wave slowness isthe true formation shear slowness. Plottingflexural-wave slowness versus frequency andidentifying the zero-frequency limit of the curveallow estimation of formation shear slowness. Inthis way, analysis of flexural-wave dispersionallows estimation of shear slowness in fast orslow formations.13

Up to now, this article has concentrated onthe simplest case of a homogeneous isotropicformation and monopole and dipole sources.Such a formation has one P-wave slowness, oneStoneley-wave slowness and one S-waveslowness. Most of the applications for usingsonic-logging results to infer formation porosity,permeability, fluid type, elastic moduli, lithologyor mineralogy have been developed forhomogeneous isotropic formations. Additionalcomplexities arise in inhomogeneous oranisotropic formations. The rest of this articleaddresses anisotropy first, then looks atinhomogeneous formations.

AnisotropyThe spatial alignment of mineral grains, layers,fractures or stress causes wave velocity to varywith direction, a property called anisotropy.14 Inseismic surveys, the anisotropy of the overburdenshales is known to cause imaging difficulties thatneed to be corrected to place reservoir targets atthe correct location. Information about aniso-tropy is also needed whenever an understandingof rock mechanics is required. Directionaldrilling, drilling in tectonically active areas,designing oriented-perforating jobs, planninghydraulic-fracturing operations and developingpressure-supported recovery plans all benefitfrom knowledge of elastic anisotropy.

The natural processes that cause anisotropyalso cause it to have one of two mainorientations: horizontal or vertical. To a firstapproximation, horizontal layers create ananisotropic medium that may be consideredisotropic in all horizontal directions, butanisotropic vertically. Such a medium is knownas transversely isotropic with a vertical axis ofsymmetry (TIV) (above right). Similarly, verticalfractures create a simplified anisotropic mediumthat may be considered isotropic in any directionaligned with fracture planes, and anisotropic inthe direction orthogonal to fracture planes. Thismedium is known as transversely isotropic with ahorizontal axis of symmetry (TIH).

Sonic waves are sensitive to these directionaldifferences in material properties. Waves travelfastest when the direction of particle motion,called polarization, is parallel to the direction ofgreatest stiffness. Compressional waves haveparticle motion in the direction of propagation,so P-waves travel fastest in directions parallel tolayering and fractures, and travel more slowlywhen perpendicular to layering and fractures.

10. Kimball CV and Marzetta TL: “Semblance Processing ofBorehole Acoustic Array Data,” Geophysics 49, no. 3(March 1984): 274–281.

11. Kimball CV: “Shear Slowness Measurement byDispersive Processing of the Borehole Flexural Mode,”Geophysics 63, no. 2 (March–April 1998): 337–344.

12. Murray D, Plona T and Valero H-P: “Case Study ofBorehole Sonic Dispersion Curve Analysis,”Transactions of the SPWLA 45th Annual LoggingSymposium, June 6–9, 2004, Noordwijk, TheNetherlands, paper BB.The key parameters required for dispersion-curvemodeling are formation slowness, formation density,borehole-fluid velocity, borehole-fluid density andborehole diameter.

13. Sinha BK and Zeroug S: “Geophysical Prospecting UsingSonics and Ultrasonics,” in Webster JG (ed): WileyEncyclopedia of Electrical and Electronic EngineersVol. 8. New York City: John Wiley and Sons, Inc. (1999):340–365.

14. This holds for alignments on scales that are smaller thanthe wavelength of the waves in question.Armstrong P, Ireson D, Chmela B, Dodds K, Esmersoy C,Miller D, Hornby B, Sayers C, Schoenberg M, Leaney Sand Lynn H: “The Promise of Elastic Anisotropy,” Oilfield Review 6, no. 4 (October 1994): 36–47.

> Dispersion curves characterizing slowness at different frequencies in an isotropicformation. Shear waves are not dispersive; alltheir frequency components travel at the sameslowness. Stoneley waves are only slightlydispersive. Flexural modes excited by a dipolesource exhibit large dispersion in this formation.At the zero-frequency limit, flexural-waveslowness tends to the shear-wave slowness(dotted line).

Slow

ness

, µs/

ft

400

300

200

100

00 2 4 6 8

Frequency, kHz

Stoneley

Dipoleflexural

Shear

> Simplified geometries in elastic anisotropy. Inhorizontal layers (top), elastic properties may beuniform horizontally, but vary vertically. Such amedium may be approximated as transverselyisotropic with a vertical axis of symmetry (TIV),meaning that the formation may be rotated aboutthe axis to produce a medium with the sameproperties. In formations with vertical fractures(bottom), elastic properties may be uniform invertical planes parallel to the fractures, but mayvary in the perpendicular direction. This mediummay be approximated as transversely isotropicwith a horizontal axis of symmetry (TIH).

Vertical axisof symmetry

TIV

x

z

y

Horizontal axisof symmetry

x

z

y

TIH

58171schD06R1.qxp.p39.ps 5/22/06 4:44 PM Page 39

Shear waves have particle motion perpen-dicular to the direction of propagation (above).In isotropic media, S-wave particle motion is contained in the plane containing the Pand S raypaths. In anisotropic media, an S-wave will split into two shear waves withdifferent polarizations and different velocities.The S-wave polarized parallel to the layering or

fractures is faster than the S-wave polarizedorthogonal to layering or fractures. Flexuralwaves behave like S-waves, and so they split inthe same ways. In the discussion that follows, S-waves and flexural waves are usedinterchangeably.

Sonic logging can be used to detect andquantify the direction and magnitude ofanisotropy if the tool geometry and theanisotropy axis are properly aligned. In a TIHmedium, such as a formation with alignedvertical fractures, S-waves propagating along avertical borehole split into two waves, and thefast wave is polarized in the plane of the

40 Oilfield Review

> Particle motion and direction of propagation in compressional and shear waves.Compressional waves (A) have particle motion in the direction of wave propagation. Shearwaves have particle motion orthogonal to the direction of propagation. In a TIH anisotropicmaterial (bottom), a shear wave propagating parallel to fractures splits. The S-wave withvertically polarized particle motion, parallel to the fractures (C) is faster than the S-wavewith particle motion polarized orthogonal to fractures (B).

Particlemotion

Particlemotion

Particlemotion

A

B

C

Horizontal axisof symmetry

Time

Time

Time

Compressional-wave amplitude

Slow shear-waveamplitude

Fast shear-waveamplitude

Wave propagation direction

58171schD06R1.qxp.p40.ps 5/22/06 4:44 PM Page 40

Spring 2006 41

fractures (above). Similarly, in a TIV medium,such as a shale or a finely layered interval, S-waves propagating in a horizontal borehole split,and the fast wave becomes polarized in thebedding plane.

The polarization of S-waves split byanisotropy cannot be detected by a singlemonopole receiver. Directional receivers arerequired. A suitable directional receiver can becreated by substituting a single monopolereceiver with two or more pairs of monopolereceivers. Each pair of monopole receivers actsas a dipole reciever. For adequate recording offlexural waves, at least one dipole receiver isaligned with each dipole transmitter. At eachfiring of the dipole source, signals are recordedby the dipole receiver oriented “inline” with thatsource and also by the dipole receiver oriented“offline” (above right).15 This example showsrecording of flexural waves at 13 receiverstations with eight receivers distributed in a ringat each station.16

In isotropic formations, flexural wavesgenerated by a dipole source remain polarized inthe plane of the source and are detected only onthe dipole receiver aligned in that plane.However, in anisotropic formations, the flexuralwave splits into fast and slow componentsaligned with the formation anisotropy. Unless thetool axes are fortuitously aligned with the

formation’s fast and slow directions, flexural-wave energy will be recorded by the offline aswell as the inline receivers.

The directions, or azimuths, of fast and slowshear or flexural waves can be seen in a crossed-dipole log. Creating a crossed-dipole log is amultistep process. The first step is decompo-sition and recombination of the waveforms

> Inline and offline response on azimuthally distributed receivers from aborehole flexural wave in an anisotropic formation. The flexural wavewas excited by firing of the X-dipole transmitter, Tx, shown at the bottom.In this TIH medium, the flexural wave splits into fast and slow waves withcomponents of particle motion on all receivers, not just those alignedwith the tool X-axis.

Tool axis

x

x’y

θ y’Tool orientationrelative to formation

Formation fastshear-wave axis

Ty

Tx

Dipoletransmitter

pair

Undisturbedborehole

Low-frequencyboreholeflexural wave(exaggerated)

R5x

R6x

R7x

R8x

R9x

R10x

R11x

R12x

R13x

R4x

R3x

R2x

R1x

R13y

R12y

R11y

R7y

R6y

R5y

R4y

R3y

R2y

R1y

R10y

R9y

R8y Receiver-8 ring

Receiver-9 ring

Receiver-10 ring

Receiver-11 ring

Receiver-12 ring

Receiver-13 ring

Receiver-7 ring

Receiver-6 ring

Receiver-5 ring

Receiver-4 ring

Receiver-3 ring

Receiver-2 ring

Receiver-1 ring

Receiver array

15. Offline is sometimes referred to as crossline.16. Pistre V, Kinoshita T, Endo T, Schilling K, Pabon J,

Sinha B, Plona T, Ikegami T and Johnson D: “A ModularWireline Sonic Tool for Measurements of 3D (Azimuthal,

Radial, and Axial) Formation Acoustic Properties,”Transactions of the SPWLA 46th Annual LoggingSymposium, New Orleans, June 26–29, 2005, paper P.

> Shear-wave splitting in a vertical borehole in a TIH medium with vertical fractures. No matterhow the dipole source is oriented relative to thefast and slow directions of the medium, the shearwave will split into fast and slow components.The fast component aligns parallel to the plane of the fractures, while the slow component alignsperpendicular to the plane of the fractures.

Dipolereceivers

Dipolesource

FastS-wave

SlowS-wave

Sourcepulse

58171schD06R1.qxp.p41.ps 5/22/06 4:44 PM Page 41

acquired on all sensors at each receiver stationto yield four waveforms corresponding to theinline and offline responses at every depth to thetwo orthogonal dipole transmitters. Next, thesewaveforms are mathematically rotated to putthem in a coordinate system consistent with thedirections of maximum and minimum offlinewaveform energy.17 Then, the waveforms corre-sponding to fast- and slow-shear orientationsundergo semblance processing to obtain the fastand slow shear-wave slownesses.18 Zones withequal fast and slow shear-wave slownesses areisotropic, while zones with large differencesbetween fast and slow shear-wave slownesses arehighly anisotropic.

The slownesses of the fast and slow S-wavesand the P- and Stoneley waves—the four slow-nesses that can be measured by sonic logging inan anisotropic medium—are transformed intofour anisotropic moduli. These four moduli canalmost characterize the simplest of anisotropicmedia. TIV and TIH media require five moduli to

be fully characterized. For more complex types ofanisotropy, more measurements are required,such as P-waves propagating in differentazimuths or inclinations, or S-waves travelingvertically and horizontally. Surface seismic andborehole seismic surveys often can provide thisinformation.

InhomogeneityFormation properties may vary not only withmeasurement direction, as in anisotropic forma-tions, but also from place to place, in what arecalled inhomogeneous, or equivalently, hetero-geneous, formations. As with anisotropy,detecting and quantifying inhomogeneity usingacoustic waves will depend on the type offormation variation and its geometry relative tothe borehole axis.

Standard sonic logging can characterizeformation properties that vary along theborehole. Early sonic-logging tools run in vertical

boreholes identified inhomogeneities in the formof boundaries between horizontal layers (see“History of Wireline Sonic Logging,” page 32).Other heterogeneities, such as high-permeabilityzones or open fractures that intersect theborehole, can be detected using Stoneley waves,as described earlier.

Formation properties that vary away from theborehole, or along the radial axis, are evidence ofthe drilling process and are more difficult toassess. The drilling process removes rock andcauses the in-situ stresses to redistribute, orconcentrate, around the borehole in a well-knownelastic manner.19 In addition, drilling not onlybreaks the rock that is removed to form theborehole, but also may mechanically damage avolume of rock surrounding the hole.20 This type ofdamage is called plastic deformation, in contrastto elastic, or reversible, deformation. In additionto plastic deformation, drilling fluid may reactwith clays, causing swelling and altering near-wellbore velocities. Mud that invades pore spacedisplaces formation fluids that probably havedifferent properties, also altering sonic velocities.Drilling-induced variations may be more gradualthan variations across layer interfaces.

42 Oilfield Review

17. Alford RM: “Shear Data in the Presence of AzimuthalAnisotropy: Dilley, Texas,” Expanded Abstracts, 56th SEGAnnual International Meeting, Houston (November 2–6,1986): 476–479.Brie A, Endo T, Hoyle D, Codazzi D, Esmersoy C, Hsu K,Denoo S, Mueller MC, Plona T, Shenoy R and Sinha B:“New Directions in Sonic Logging,” Oilfield Review 10,no. 1 (Spring 1998): 40–55.

18. Esmersoy C, Koster K, Williams M, Boyd A and Kane M:“Dipole Shear Anisotropy Logging,” Expanded Abstracts,64th SEG Annual International Meeting, Los Angeles(October 23–28, 1994): 1139–1142.Kimball and Marzetta, reference 10.

19. Winkler KW, Sinha BK and Plona TJ: “Effects ofBorehole Stress Concentrations on Dipole AnisotropyMeasurements,” Geophysics 63, no. 1(January–February 1998): 11–17.

20. Winkler KW: “Acoustic Evidence of Mechanical DamageSurrounding Stressed Borehole,” Geophysics 62, no. 1(January–February 1997): 16-22.

21. Zeroug S, Valero H-P and Bose S: “Monopole RadialProfiling of Compressional Slowness,” prepared forpresentation at the 76th SEG Annual InternationalMeeting, New Orleans, October 1–3, 2006.

22. Sinha B, Vissapragada B, Kisra S, Sunaga S,Yamamoto H, Endo T, Valero HP, Renlie L and Bang J:“Optimal Well Completions Using Radial Profiling ofFormation Shear Slownesses,” paper SPE 95837,presented at the SPE Annual Technical Conference andExhibition, Dallas, October 9–12, 2005. Sinha BK: “Near-Wellbore Characterization Using RadialProfiles of Shear Slownesses,” Expanded Abstracts, 74th SEG Annual International Meeting, Denver (October 10–15, 2004): 326–329.

23. Chang C, Hoyle D, Watanabe S, Coates R, Kane R,Dodds K, Esmersoy C and Foreman J: “Localized Mapsof the Subsurface,” Oilfield Review 10, no. 1 (Spring1998): 56–66.

24. Hornby BE: “Imaging of Near-Borehole Structure UsingFull-Waveform Sonic Data,” Geophysics 54, no. 6 (June 1989): 747–757.

> Compressional and shear radial profiles in an anisotropic inhomogeneousformation. The profile of variation in compressional slowness (Track 4) is created bytomographic reconstruction based on tracing rays through a modeled formationwith properties that vary gradually away from the borehole. The percentagedifference between observed slowness and slowness of the unaltered formation isplotted on color and distance scales to indicate the extent of difference away fromthe borehole. In these sandstones, identifiable from the gamma ray log in Track 2,compressional slowness near the borehole varies by up to 15% from far-fieldslowness, and the variation extends to more than 12 in. [30 cm] from the boreholecenter. The borehole is shown as a gray zone. Shear radial profiles show thedifference between fast shear-wave slowness and far-field slowness (Track 1) andthe difference between slow shear-wave slowness and far-field slowness (Track 3).Large differences in shear slowness extend out to almost 10 in. [25 cm] from theborehole center. The radial variation in compressional and shear velocities isdrilling-induced.

X,480

X,490

Mea

sure

d de

pth,

ft

ft

Distance fromBorehole Center

ft

Distance fromBorehole Center

ft0 20 2

0 250 25

Distance fromBorehole Center

10 110gAPI

GammaRay

Fast ShearDifferential

%0 25

2 0

Slow ShearDifferential

%

CompressionalDifferential

%

58171schD06R1.qxp.p42.ps 5/22/06 4:44 PM Page 42

Spring 2006 43

Alteration in near-wellbore properties cancause velocities to increase or decrease relativeto the unaltered, or far-field, formation. Usually,drilling-induced damage reduces formationstiffness, causing velocities to decrease near theborehole. However, when drilling fluid replacesgas as the pore-filling fluid, the resultingformation is stiffer, so compressional velocityincreases near the borehole.

Radial alteration of rocks and fluids affectscompressional and shear velocities differently.Alteration that reduces stiffness of the rockfabric, such as drilling-induced cracking orweakening, causes both P and S velocities todecrease. However, a change in pore fluid haslittle effect on S velocity, while P velocity maychange dramatically. For example, when drillingfluid replaces gas, P-wave velocity increases, butS-wave velocity is relatively unaffected.Complete characterization of radial inhomo-geneity requires analysis of radial variation ofcompressional and shear slownesses.

A radial compressional-slowness profile isgenerated by collecting P-wave data for multipledepths of investigation, from near the wellbore tothe unaltered, far-field formation. This requiresrecordings from a wide range of transmitter-

receiver spacings. Ray-tracing techniques invertthe refracted compressional arrivals to yieldcompressional slowness versus distance from theborehole.21 The difference between near-wellbore compressional slowness and far-fieldcompressional slowness can be plotted alongwith depth of radial alteration (previous page).In this example, radial variations of shearslownesses are also plotted.

Radial variations in shear slowness arequantified through inversions of the broadbanddispersions of flexural and Stoneley modes.22 Athigh frequencies, these dispersive modesinvestigate the near-wellbore region, and at lowfrequencies, they probe the unaltered formationfar from the borehole. Dispersion data from awide range of frequencies help produce the most reliable radial profiles of variations in shear slowness.

Some of the most challenging inhomo-geneities to characterize are those that do notintersect the borehole. These may be verticalfractures or faults near a vertical borehole orsedimentary interfaces near a horizontal well.Detecting such inhomogeneities requires amethod that looks deep into the formation andthat is able to detect abrupt changes information properties.

The sonic-imaging technique, sometimescalled the borehole acoustic reflection survey,provides a high-resolution directional image ofreflectors up to tens of feet from the borehole(left).23 Consequently, this technique hassignificant potential application in horizontalwells. To create an image, the tool recordswaveforms of relatively long duration from themonopole transmitters. Receivers must bedistributed around the tool to allow the azimuthsof the reflections to be distinguished.

Complex data processing similar to thatdesigned for surface seismic surveys is applied ina multistep process. First, a compressional-velocity model of the region in the vicinity of theborehole is created using the P head waves.Then, to extract reflected energy, the traditionalsonic arrivals, including P and S head waves andStoneley waves, must be filtered from thewaveforms for each shot. The filtered traces areinput to depth migration, a process that positionsreflections in their correct spatial location usingthe velocity model.

The migration process formally converts a setof amplitude and traveltime measurements intoa spatial image of the formation. This can beviewed as a triangulation process in which thedistance and the dip of a reflector relative to theborehole are determined by the signals recordedat receivers at different TR spacing. Thereceivers at different azimuths around theborehole measure different distances to areflector depending on the azimuth and the dipof the reflector relative to the borehole.

The sonic-imaging technique was developedin the 1980s, but results have improved withadvances in sonic tools and processingmethods.24 The technique has been used to imagesteeply dipping beds from near-verticalboreholes and sedimentary boundaries fromhorizontal wells. For examples of sonic imagingand other applications of sonic measurementssee “Sonic Investigations In and Around theBorehole,” page 14. –LS

> Sonic-imaging data-acquisition geometry. Designed to detect layerboundaries and other inhomogeneities roughly parallel to the borehole, thesonic-imaging technique records reflected signals (red rays) from interfacestens of feet away. Borehole signals (black rays) must be filtered out.

Coal bed

Reflectedsignals

Borehole signals

58171schD06R1.qxp.p43.ps 5/22/06 4:44 PM Page 43

44 Oilfield Review

From Inner Earth to Outer Space

Joel Lee GrovesJohn SimonettiStefan VajdaWolfgang ZieglerPrinceton Junction, New Jersey, USA

Jacob I. TrombkaGoddard Space Flight CenterGreenbelt, Maryland, USA

For help in preparation of this article, thanks to Edward Durner, Steve Meddaugh, Jim Roderick and Joel Wiedemann, Princeton Junction, New Jersey. EcoScope is a mark of Schlumberger.Teflon is a mark of E.I. du Pont de Nemours and Company.

In the 1930s, Conrad and Marcel Schlumberger began development of tools and sensors

to explore Earth’s inner space. Some 75 years later, similar detectors are helping

scientists investigate the fundamental nature and origin of objects in outer space.

On a cold day in February 2001, a spacecraftlanded on 433 Eros, an asteroid between theorbits of Mars and Jupiter. The spacecraft hadcompleted its five-year journey to investigatefundamental questions about the nature andorigin of near-Earth objects for the first time.

The technical demands of the Near EarthAsteroid Rendezvous NEAR-Shoemaker (NEAR)mission were immense. A multidisciplinary teamof US National Aeronautics and SpaceAdministration (NASA) scientists and engineersdrew from many scientific and industrialresources, including the predominantly inner-Earth-focused oil and gas industry.

Applying technologies developed for oil andgas exploration to scientific endeavors is not anew practice. Oilfield technologies have oftenbeen applied in the interest of science. Forexample, deep-drilling projects conducted onland and in most major oceans of the world havecontributed to our understanding of Earth’s pastas well as its future.

Engineers and scientists with the interna-tionally funded Ocean Drilling Program begansubsea drilling operations in 1961 to explore thehard outer layer of the Earth’s crust, orlithosphere. Scientists used tools and techniquesdeveloped for oil and gas exploration to documentcontinental drift and to generate a substantialquantity of data relating to plate tectonics.1

> Distant spiral galaxy. The Hubble Space Telescope captured this image of light that left the spiralgalaxy NGC1300 more than 69 million years ago. Barred spirals differ from normal spiral galaxies inthat the arms of the galaxy do not spiral all the way into the center, but are connected to the twoends of a straight bar of stars containing the nucleus at its center. At Hubble’s resolution, fine details,some of which have never before been seen, show disk, bulge and nucleus throughout the galaxy’sarms. The nucleus shows its own distinct spiral structure that is about 3,300 light-years across. Theimage was constructed from exposures taken in September 2004 by the Advanced Camera for Surveys.(Image courtesy of NASA.)

1. Andersen RN, Jarrard R, Pezard P, Williams C andDove R: “Logging for Science,” The Technical Review 36,no. 4 (October 1988): 4–11.

2. Kerr RA: “Signs of a Warm, Ice-Free Arctic,” Science 305,no. 5691 (September 17, 2004): 1693.

3. For more on deep-ocean drilling: 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’sSecrets,” Oilfield Review 16, no. 4 (Winter 2004/2005):24–37.

4. Acceleration is often expressed in units of g-force (gn),which is defined as 9.80665 m/s2, approximately equal tothe acceleration due to gravity on the Earth’s surface atsea level.

58171schD07R1.qxp.p44.ps 5/25/06 10:48 PM Page 44

Spring 2006 45

In 2004, engineers drilling in the Arctic Oceanat the crest of the Lomonosov ridge providedpreliminary evidence that the Arctic was ice-freeand warm about 56 million years ago.2 Scientistsanalyzed cores recovered from the drilling projectto help determine when, why and how the Arctictemperature changed. They also gained insightinto current global-warming trends.3

Understanding the fundamental processesthat occur deep within the Earth’s crust hascontributed to our knowledge of many inner-earth events, including volcanic activity, platetectonics, weather fluctuations, and chemicaland thermodynamic processes that lead tomineral deposition.

Hydrocarbons are most often found inforbidding environments. Tools and sensors arestressed to their limits as boreholes are drilleddeeper into the Earth’s crust where high temper-ature and pressure and excessive vibrations arecommon, and stress and shock forces reachthousands of times the acceleration of gravity(gn).4 Tools and instruments must also surviveextreme thermal cycles, from the cold surface ofthe Arctic to temperatures higher than 204°C[400°F] in the downhole environment. Drilling,logging and measurement instruments haveevolved to meet these challenges. Today, oil andgas E&P tools and instruments are designed andthoroughly tested for extended exposure to theseharsh environments.

Similarly, the forces encountered whilelaunching and accelerating a vehicle into spacecan be traumatic to equipment components. Forexample, the shock of pyrotechnic-stage separa-tion can reach over 4,000 gn, stressing both thevehicle and its payload. Once in space, depend-ing on orientation relative to the Sun, tempera-ture extremes range from more than 100°C[212°F] to below -200°C [-328°F]. Because of theneed to operate in harsh environments, the toolsand instrument packages designed for deep-welldrilling are inherently applicable to otherchallenging environments, such as outer space.

Whether exploring inner space for scientificpurposes, searching for oil and gas or probing thevastness of outer space, the desire to explore hasdriven the history of modern civilizations. Thisdrive led, at least in part, to the conquest of themoon in the 1960s, marking the beginning of anew generation in space exploration and travel.More recently, spacecraft, such as the HubbleSpace Telescope (HST), aided by technologiesdeveloped for oil and gas exploration, havepeered from Earth orbit ever more sharply anddeeply into the universe beyond our solar system(previous page).

58171schD07R1.qxp 5/26/06 7:22 PM Page 45

As we move from exploration of inner spaceto that of outer space, the tools and techniquesdeveloped for exploration deep beneath theEarth’s surface are helping to uncover themysteries of our solar system and the far reachesof space. In this article, we discuss a few of therecent contributions made to space explorationby the scientists and engineers of the petroleumindustry. Although the mission of the NEARspacecraft has ended, oilfield technology aboardthe HST and the Cassini-Huygens Saturn probecontinues to expand our knowledge and chartour way forward in the quest for knowledge.

Keeping the Hubble on TargetThroughout history, our understanding of theuniverse has been limited by what we could see.The invention of the telescope enhanced ourvision and allowed observations by Copernicus,Kepler and Galileo in the 16th and 17th centuries

to show that the Earth was not the center of theuniverse.5 By the 18th century, the developmentof the telescope helped scientists investigate thecosmos. Increasingly bigger and better telescopeshave routinely discovered and documentedplanets, stars and nebulae that are invisible tothe naked eye.

As recently as the beginning of the 20thcentury, most astronomers still believed that theuniverse consisted of a single galaxy, the MilkyWay—a collection of stars, dust and gas in thevastness of space. However, the universe as weknew it changed in 1924 when Americanastronomer Edwin Hubble used the 2.54-m [100-in.] Hooker Telescope on Mount Wilson,near Los Angeles, to observe billions of othergalaxies beyond the Milky Way.6

For astronomers like Edwin Hubble, therehas always been a major obstacle to a clear viewof the universe—the Earth’s atmosphere. Gasesand airborne particulates in the atmosphere blur

visible light, cause starlight to scintillate, ortwinkle, and hinder or totally absorb infrared,ultraviolet, gamma ray and X-ray wavelengths .

To minimize atmospheric distortion,scientists built observatories on mountaintopsand away from the areas of highly radiated light,or sky glow, found near large cities. This effortmet with varying levels of success. Today,adaptive optics and other image-processingtechniques have minimized, but not totallyeliminated, atmospheric effects.7

In 1946, Princeton astrophysicist LymanSpitzer documented the potential benefits of atelescope in space, well above Earth’satmosphere. Then, following the launch of theSoviet satellite Sputnik in 1957, NASA placed twoorbital astronomical observatories (OAO) intoEarth orbit. The OAOs made a number ofultraviolet observations and established thebasic principles for the design, manufacture andlaunch of future space observatories.8

Scientific, governmental and industrialgroups continued the move toward extrater-restrial exploration by planning the next stepbeyond the OAO program. Spitzer gathered thesupport of other astronomers for a large orbitaltelescope, later called the Hubble SpaceTelescope, and in 1969, the National Academy ofSciences approved the project.9

NASA’s Goddard Space Flight Center inGreenbelt, Maryland, USA, was responsible forscientific instrument design and ground controlfor the space observatory. In 1983, the SpaceTelescope Science Institute (STScI) wasestablished at The Johns Hopkins University inBaltimore, Maryland. The staff of STScI managedthe telescope’s observation time and data. NASAchose the Marshall Space Flight Center inHuntsville, Alabama, USA, as the lead NASA fieldcenter for the design, development andconstruction of the space telescope. Perkin-ElmerCorporation, now Hughes Danbury OpticalSystems, developed the optical telescope assemblyand the fine-guidance sensor (FGS) system.

On April 24, 1990, after numerous projectdelays, the space shuttle Discovery lifted off fromEarth carrying the HST in its cargo bay. Thefollowing day, the school-bus-size space telescopewas deployed in low-Earth orbit (above left). Free of atmospheric distortion, the gianttelescope mirror began its mission of gatheringphotons from as far away as the edge of the known universe.

46 Oilfield Review

> Servicing the Hubble Space Telescope (HST). The Space Shuttle Discovery,mission STS-82, lifts the HST from its service bay after the second Hubbleservice mission. With a launch weight of 11,340 kg [25,000 lbm], the Hubble’smain structure is 13 m [42.6 ft] long and 4.27 meters [14 ft] wide. Its twin solar arrays span 13.7 meters [45 feet] when deployed. The telescopeitself is a reflecting configuration termed a Cassegrain, comprising a 2.4-m[94.5-in.] primary mirror, and a 30-cm [12.2-in.] secondary mirror. (Imagecourtesy of NASA.)

58171schD07R1.qxp.p46.ps 5/25/06 10:48 PM Page 46

Spring 2006 47

Critical to the performance of the HST isstaying on target for extended periods of time.Electromagnetic waves emitted from distantobjects are often faint or weak, so the HST muststay perfectly positioned while the photons arebeing collected in sufficient quantities to form an image. To accomplish this, engineers used the Schlumberger oilfield photomultiplier-tubetechnology to design the FGS system.10

An FGS is essentially a targeting cameracapable of making celestial measurements,locking onto guide stars and providing data formaneuvering the telescope.11 Two FGSs are usedto point the telescope at an astronomical targetand hold that target in the telescope’s field ofview; the third FGS can then be used forastrometry measurements.12

The FGS system can maintain pointingaccuracy to 0.007 arcseconds, allowing thetelescope’s pointing-control system (PCS) tokeep the Hubble telescope on target duringcamera exposure times of 10 hours or more.13 ThePCS combines a number of different sensorsubsystems to achieve this milliarcsecondpointing accuracy. This level of accuracy andprecision is comparable to training a laser beamon a target the size of a thumbnail from adistance of 442 km [275 miles].

Within the housing of each FGS instrumentare two orthogonal white-light, shearinginterferometers, their associated optical andmechanical elements and four Schlumberger S-20 photomultiplier tubes (PMTs) (above right).14

These PMTs are based on the same ruggedconstruction as those used in well-logginginstruments. The photocathode was manu-factured using the same technology as tubes usedin oilfield service applications. For use on theHST, the PMTs were designed to be sensitive overa spectral range of 400 to 700 nanometers (nm),with an efficiency of approximately 18% at theblue end of the electromagnetic spectrum anddiminishing linearly to about 2% at the red end.

Each FGS interferometer consists of apolarizing beam splitter followed by two Koestersprisms. To measure the direction of the lightemitted by a guide star, the pairs of Koestersprisms are oriented perpendicular to oneanother. The angle of the wavefront in the X andY planes gives the precise angular orientation ofthe guide star relative to the HST’s optical path.These data, once fed into the PCS, are used tocontrol the telescope orientation relative to aguide star.

5. NASA—Hubble’s Conception: http://hubble.nasa.gov/overview/conception-part1.php (accessed April 18, 2006).

6. NASA, reference 5.7. Adaptive optics is a technology used to improve the

performance of optical systems by reducing the effectsof rapidly changing optical distortion typically resultingfrom changes in atmospheric conditions. Adaptive opticsworks by measuring the distortion and rapidlycompensating for it using either deformable mirrors ormaterial with variable refractive properties.

8. Smith RW: The Space Telescope–A Study of NASA,Science, Technology and Politics. New York City:Cambridge University Press, 1989.

9. Smith, reference 8.10. For more on photomultiplier tubes: Adolph B, Stoller C,

Brady J, Flaum C, Melcher C, Roscoe B, Vittachi A andSchnorr D: “Saturation Monitoring With the RSTReservoir Saturation Tool,” Oilfield Review 6, no. 1(January 1994): 29–39.

11. Space Telescope Science Institute–FGS History:http://www.stsci.edu/hst/fgs/design/history (accessedMarch 14, 2006).A guide star is one of many bright stars used fortelescope positioning and triangulation.

12. Astrometry is a branch of astronomy that deals with thepositions of stars and other celestial bodies, theirdistances and movements.

13. A second of arc, or arcsecond, is a unit of angularmeasurement that comprises one-sixtieth of anarcminute, or 1⁄3,600 of a degree of arc or 1⁄1,296,000 ≈ 7.7x10-7

of a circle. It is the angular diameter of an object of 1unit diameter at a distance of 360x60x60/(2π) ≈ 206,265units, such as (approximately) 1 cm at 2.1 km.

14. Interferometers were first used by Michaelson, who wonthe Nobel Prize in 1907 for his work using an opticalinterferometer to accurately measure the speed of light.

> Guiding Hubble. Light from the HST Optical Telescope Assembly (OTA) is intercepted by a pickoffmirror in front of the HST focal plane and directed into the fine-guidance system (FGS) (left). The lightrays are collimated, or made parallel, and then compressed by an aspheric collimating mirror andguided to the optical elements of the star selector assembly. Small rotations of the star selector A and B assemblies alter the direction of the target’s collimated beam, and hence the tilt of the incidentwavefront with respect to the Koesters prism (right). As the wavefront rotates about Point B, therelative phase of the transmitted and reflected beams change as a function of angle alpha. When thewavefront’s propagation vector is parallel to the plane of the dielectric surface, the relative intensitiesof the two emergent beams detected by the photomultiplier tubes will be equal. When alpha is notzero, the intensities of the left and right output beams will be unequal and the PMTs will recorddifferent photon counts, thus providing the telescope guidance control system with data allowing forpointing correction. [Images courtesy of NASA and The Johns Hopkins University Applied PhysicsLaboratory (JHUAPL).]

Star selectormirrors

Correction group

Deviation prism

Pickoff mirrorFilters (5 in wheel)

Beam-splitter prism

Koesters prism

Doublet lens (4)

Photomultiplier tubewith pinhole lensassembly (4)

Aspheric collimatingmirror

PMT B

Field stop Field stopField lensField lens

Positivedoublet

KoestersprismDielectric

beamsplitter

Incident wavefront

Alpha

A C

D

B

Positivedoublet

PMT A

Optical bench

58171schD07R1.qxp.p47.ps 5/25/06 10:48 PM Page 47

In addition to guiding the HST, the accuracyof FGS sensors makes them useful for high-precision astrometric measurements. Thesemeasurements help scientists determine theprecise positions and motions of stars. The FGSsensors can provide star positions about 10 timesmore precisely than measurements made withground-based telescopes. Scientists use astro-metric measurements to help define wobble inthe motion of stars that might indicate thepresence of a planetary companion (below left).The motions of stars can also determine whethera star pair represents a true binary star system,or simply an optical binary.15

Aided by elements of oilfield technology, theHubble Space Telescope continues its worktoday. Scientists are using instruments like theHST to search the far reaches of the universe anduncover secrets of the past while reaching intoour future.

Asteroids—Up Close and PersonalA little closer to home, technologies developedfor oilfield use are helping scientists exploreasteroids in our solar system. These large piecesof rock are primordial objects left over from theformation of the solar system. Some scientistshave suggested that asteroids are the remains ofa protoplanet that was destroyed in a massivecollision. However, the prevailing view is thatasteroids are leftover rocky matter that neversuccessfully coalesced into planets.

Scientists theorize that the planets of thesolar system formed from a nebula of gas anddust that coalesced into a disk of dust grains

around the developing Sun. Within the disk, tinydust grains coagulated into larger and largerbodies called planetesimals, many of whicheventually accreted into planets over a period of100 million years. However, beyond the orbit ofMars, gravitational interference from Jupiterprevented protoplanetary bodies from growing todiameters larger than about 1,000 km[620 miles].16

Most asteroids are concentrated in an orbitalbelt between Mars and Jupiter (below). Thesespace rocks orbit the Sun as planets do, but theyhave no atmosphere and very little gravity. Theasteroids in the belt comprise a significant

48 Oilfield Review

> True binary stars. Each of the two stars in atrue binary system orbits around the center ofmass of the system. Kepler’s laws of planetarymotion govern how each star orbits the center of mass. At aphelion (A), each of the two starsare the farthest apart in their respective orbits.At perihelion (C), the stars are the closest.

+

+

+

+

A

B

C

D

> Main asteroid belt. The asteroid belt is a region of the solar system falling roughly between theplanets Mars and Jupiter where the greatest concentration of asteroid orbits can be found. The mainbelt region contains approximately 93.4% of all numbered minor planets. Trojan asteroids occupy tworegions centered 60° ahead of and behind Jupiter. Several hundred Trojans are known out of a totalpopulation that includes an estimated 2,300 objects bigger than 15 km [9 miles] across and manymore of smaller size; most do not move in the plane of the planet’s orbit but rather in orbits inclinedby up to 40°.

Mars

Mercury

VenusEarth

Jupiter

Trojanasteroids

Trojanasteroids

Astronomical units

0 2.7 5.21.5

Mainasteroidbelt

58171schD07R1.qxp.p48.ps 5/25/06 10:48 PM Page 48

Spring 2006 49

amount of material—putting all of the asteroidstogether would form a body about 1,500 km [930miles] in diameter, roughly half the size ofEarth’s moon.17

Not all asteroids are far away in the asteroidbelt. Some, called near-Earth asteroids (NEAs),have orbits that bring them close to Earth.Astronomers believe NEAs to be fragmentsejected from the main asteroid belt by asteroid-asteroid collisions or by gravitational perturba-tions from Jupiter. Some NEAs could also be thenuclei of dead, short-period comets.

Since many asteroids have historically struckEarth and its moon, understanding theircomposition and origin may be a key to our pastas well as our future. Scientists believe that thechemical building blocks of life and much ofEarth’s water may have arrived on asteroids orcomets that bombarded the planet in the earlystages of its development (above left). Onewidely accepted theory suggests that an asteroidmeasuring at least 10 km [6 miles] across,impacted the Earth some 65 million years ago,causing mass extinctions among many life forms,including the dinosaurs.

Astronomers suspect that the approximately800 NEAs found to date represent only a smallpercentage of their total population. The largestpresently known is 1036 Ganymede, with anapproximate diameter of 41 km [25.5 miles].NEAs with diameters greater than 1 km[0.6 miles] are known as potentially hazardousasteroids, suggesting that should they strikeEarth, they could threaten life as we know it.

Of the more than 700 known potentiallyhazardous asteroids, one of the largest isToutatis, an asteroid that is nearly 1.6 km[1 mile] long and orbits around the Sun withinone-half degree of Earth’s orbital plane. InDecember 1992, Toutatis passed within0.024 astronomical units (AU), or 9.4 lunardistances from Earth.18 Then, on September 29,2004, Toutatis’s orbital path brought it within0.01 AU of Earth—the closest approach of anylarge asteroid in the 20th century.

Although astronomers have known aboutasteroids for nearly 200 years, until recently, theirbasic properties, their relationship to meteoritesfound on Earth and their origins remained amystery. NASA and the scientific community,driven by both the desire to understand asteroidsand the threat to Earth presented by NEAs morethan 1 km in diameter, set in motion the plans forthe NEAR project.

A Mission of Many FirstsIn 1990, NASA introduced a new program ofplanetary missions called the Discovery program.By 1991, the first mission was chosen—arendezvous with near-Earth asteroid 433 Eros.The Johns Hopkins University Applied PhysicsLaboratory (JHUAPL) was chosen to manage theproject, and in 1995, the NEAR spacecraft wasshipped to the Kennedy Space Center in Florida.19

Discovered in 1898, the NEA Eros is one ofthe largest and best-observed asteroids.20 Withdimensions 33 by 13 by 13 km [21 by 8 by8 miles], Eros is about the size of Manhattan,New York, USA (above). It accounts for nearlyhalf of the volume of all near-Earth asteroids.

15. The term binary star refers to a double-star system, or aunion of two stars into one system based on the laws ofattraction. Any two closely spaced stars might appearfrom Earth to be a double-star pair when, in fact, theyare a foreground and background star pair widelyseparated in space. These systems are typically referredto as optical binaries.

16. NASA–Eros or Bust: http://science.nasa.gov/headlines/y2000/ast08feb_1.htm (accessed April 14, 2006).

17. NASA, reference 16.18. NASA/ Jet Propulsion Laboratory–Asteroid 4179 Toutatis:

http://echo.jpl.nasa.gov/asteroids/4179_Toutatis/toutatis.html (accessed April 14, 2006).An astronomical unit (AU) is equivalent to the distancefrom the Earth to the Sun, or approximately 149,000,000 km[92,500,000 miles].

19. The NEAR spacecraft was renamed NEAR–Shoemakerto honor planetary geologist Eugene Shoemaker(1928–1997).

20. Farquhar RW: “NEAR Shoemaker at Eros: MissionDirector’s Introduction,” Johns Hopkins APL TechnicalDigest 23, no. 1 (2002): 3–5.

> Impacting the Earth. An asteroid impacting the Earth some 49,000 year ago scarred the Earthleaving this 1.2-Km [0.7-mile] crater. This view from the Space Shuttle shows the dramatic expressionof the crater in the arid landscape of Arizona, USA. (Image courtesy of the Earth Sciences and ImageAnalysis Laboratory, NASA Johnson Space Center, STS040_STS040-614-58.)

> A large asteroid. The outline of Eros (red) issuperimposed on the island of Manhattan, New YorkCity, showing the relative size of the asteroid.

Manhattan

The footprint ofasteroid Eros

58171schD07R1.qxp.p49.ps 6/1/06 5:16 PM Page 49

The large S-type potato-shaped asteroid is one ofthe most elongated asteroids. It orbits aroundthe Sun, rotating on its axis once every 5.27hours, with a perihelion of 1.13 AU and anaphelion of 1.78 AU (top).21

NEAR departed Earth for asteroid Eros onFebruary 17, 1996, riding on top of a Delta-IIlaunch vehicle. One year later, on February 18,1997, NEAR reached its most distant point fromthe Sun, 2.18 AU, setting a new distance record

for a spacecraft with instrumentation powered bysolar cells.

By the end of its five-year mission, NEARwould produce an impressive list of spacecraftfirsts: the first spacecraft with instrumentationsolely powered by solar cells to operate beyondthe orbit of Mars, the first to encounter a C-typeasteroid, the first to encounter a near-Earthasteroid, the first to orbit a small body, and thefirst spacecraft to land on a small body.

NEAR—The Scientific MissionPrior to the NEAR mission, our knowledge ofasteroids came primarily from three sources:Earth-based remote sensing, data from theGalileo mission flybys of the two main-belt S-type asteroids 951 Gaspra and 243 Ida, andlaboratory analyses of meteorites recovered afterimpact with the Earth.

Although astronomers theorize that mostmeteors result from the collision of asteroids,they may not be completely representative of allmaterials that comprise NEAs.22 Clear linksbetween meteorite types and asteroid typesproved difficult to establish.23

Some S-type asteroids appear to be fragmentsof bodies that underwent substantial melting anddifferentiation, while others consist of whatappears to be nonmelted primitive materials likechondrites.24 Scientists believe that nonmelted S-type asteroids may have preserved thecharacteristics of the solid material from whichthe inner planets accreted.

The Galileo mission flybys provided the firsthigh-resolution images of asteroids in the early1990s. Images revealed complex surfaces coveredby craters, fractures, grooves and subtle color variations (left).25 However, Galileo’sinstrumentation was not capable of measuringelemental composition, so prior to the NEARmission, scientists continued to be unsure of therelationship between ordinary chondrites and S-type asteroids.

Mission engineers believed that the NEARdata, when combined with those from the Galileoflybys, would help scientists understand therelationship between S-type asteroids and othersmall bodies of the solar system. The NEARmission’s primary objectives were to rendezvouswith, achieve orbit around and conduct the firstscientific exploration of a near-Earth asteroid.

The NEAR SpacecraftEngineers designed NEAR’s systems to be solar-powered, simple and highly redundant.26 OnboardNEAR were five instruments designed to makedetailed scientific observations of the grossphysical properties, surface composition andmorphology of Eros. These five were the multi-spectral imager (MSI), near-infrared spectrom-eter (NIS), magnetometer (MAG), NEAR laserrangefinder (NLR) and the combined X-ray,gamma ray spectrometer (XGRS) (next page).

The MSI imaged the surface morphology ofEros with spatial resolutions down to 5 m[16.4 ft], while scientists used the NIS tomeasure mineral abundances at a spatial

50 Oilfield Review

> Approaching Eros. This image of the southern hemisphere of Eros offers a long-distance look at theasteroid’s cratered terrain. (Image courtesy of NASA/JHUAPL.)

> Asteroids close up. Shown are views of the three asteroids that had been imaged at close rangeby spacecraft prior to NEAR’s arrival at Eros. The image of Mathilde (left) was taken by the NEARspacecraft on June 27, 1997. Images of the asteroids Gaspra (middle) and Ida (right) were taken bythe Galileo spacecraft in 1991 and 1993, respectively. All three objects are presented at the samescale. The visible part of Mathilde is 59 km wide by 47 km long [37 by 29 miles]. (Images courtesy ofNASA/JHUAPL.)

Mathilde Gaspra Ida

58171schD07R1.qxp.p50.ps 5/25/06 10:49 PM Page 50

Spring 2006 51

resolution on the order of 300 m [984 ft]. TheMAG was used to define and map intrinsicmagnetic fields on Eros.

Scientists used the NLR to enhance thesurface morphology profiles derived from NEAR’simaging camera. The NLR is a laser altimeterthat measures the distance from the spacecraftto the asteroid surface by sending out a shortburst of laser light and then recording the timerequired for the signal to return from theasteroid. The ranging data were used toconstruct a global shape model and a globaltopographic map of Eros with a spatial resolutionof about 5 m.

The XGRS was the primary tool used forsurface and near-surface elemental analysis ofEros. Scientists combined data from the XGRS,MSI and the NIS to produce global maps of Eros’ssurface composition.

Development of the complex XGRS systembegan about three years prior to launch. Theinstrument was designed to detect and analyzeX-ray and gamma ray emissions from the asteroidsurface from orbital altitudes of 35 to 100 km [22 to 62 miles]. Although spectroscopy ofremote surfaces is possible during spacecraftflyby operations, measurements made whileorbiting allow longer observation times andproduce higher quality spectral data.

X-rays emitted from the Sun shining on Erosproduce X-ray fluorescence from the elementscontained in the top 1 mm [0.04 in.] of theasteroid’s surface. In the absence of anysignificant atmosphere that might otherwiseabsorb X-ray emissions, elements fluoresce atenergy levels that are characteristic of specificelements. Scientists used the X-ray fluorescenceenergy detected in the 1- to 10-keV level to infersurface elemental composition.

The XRS subunit consists of three identicalgas-filled proportional counters that provide alarge active surface area and therefore thesensitivity required for remote sensing. Similardetectors have been used on lunar orbitalmissions and most recently on Apollo missions.

The X-ray gas tubes are not particularlysensitive to temperature change, since themultiplication effect depends more on thenumber of gas molecules than the gas pressure.However, the gain in the gas tubes is sensitive tovoltage variations.

Gamma ray spectrometry provides acomplementary measurement of near-surfaceelemental composition. The gamma rayspectrometer (GRS) detects discrete-line gammaray emissions in the 0.1- to 10-MeV energy range.

At these energy levels, oxygen [O], silicon [Si],iron [Fe] and hydrogen [H] become excited, orradioactively activated, from the continual influxof cosmic rays. The GRS also detects naturallyradioactive elements such as potassium [K],thorium [Th] and uranium [U]. The measure-ments have been used for years in oil and gaswell logging to determine the physical andelemental composition of reservoir rock.

Unlike the low-energy X-rays, gamma rays arenot as easily absorbed and therefore can escapefrom regions beneath the surface, allowing theGRS to reveal elemental composition to depthsas much as 10 cm [4 in.] below the surface. By comparing elemental analysis from the XRSand GRS, scientists inferred the depth andextent of the dust layer, or regolith, covering thesurface of Eros.27

21. Asteroids are classified based on reflectance spectrumand light-reflection characteristics, or albedo, which areindicators of surface composition. S-Type (silicaceous)asteroids are more prevalent in the inner part of themain asteroid belt, while C-Type (carbonaceous)asteroids are found in the middle and outer parts of thebelt. Together, these two types account for about 90% ofthe asteroid population.Perihelion and aphelion are the orbital points nearestand farthest from the center of attraction—in this case,the Sun.

22. A meteorite is a solid portion of a meteoroid thatsurvives its fall to Earth. Meteorites are classified asstony meteorites, iron meteorites and stony ironmeteorites, and further categorized according to theirmineralogical content. They range in size frommicroscopic to many meters across. Of the several tensof tons of cosmic material entering Earth’s atmosphereeach day, only about one ton reaches the ground.

23. Cheng AF, Farquhar RW and Santo AG: “NEAROverview,” Johns Hopkins APL Technical Digest 19, no. 2(1998): 95–106.

24. Chondrites are a type of stony meteorite made mostly ofiron- and magnesium-bearing silicate minerals.Chondrites are the most common type of meteorite,accounting for about 86% that fall to Earth. Theyoriginate from asteroids that never melted, or underwentdifferentiation. As such, they have the same elementalcomposition as the original solar nebula. Chondritesderive their name from the fact that they containchondrules—small round droplets of olivine andpyroxene that apparently condensed and crystallized inthe solar nebula and then accreted with other materialsto form a matrix within the asteroid.

25. Cheng et al, reference 23.26. Cheng et al, reference 23.27. Regolith is a layer of loose material, including soil,

subsoil and broken rock, that covers bedrock. On Earth’smoon and many other bodies in the solar system, itconsists mostly of debris produced by meteorite impactsand blankets most of the surface.

> NEAR spacecraft systems. NEAR’s basic design and primary systems are shown. (Image courtesyof NASA/JHUAPL.)

Gamma rayspectrometer

X-rayspectrometer

X-ray solarmonitors

NEAR laser rangefinderNear-infraredspectrometer

Multispectral imager

Aft deck

Forward deckSide panels

Propulsion system

Solar panel

58171schD07R1.qxp.51.ps 7/5/06 5:18 PM Page 51

The GRS central detector assembly is based ona ruggedized thallium-activated [Tl] sodium iodide[NaI] scintillator unit used in oilwell loggingoperations, designed and built by Schlumberger(below). NaI-based scintillators are widely used indownhole logging-tool applications to makemeasurements of density, natural radioactivity and elemental spectra. As an example, theEcoScope multifunction logging-while-drilling tooluses a NaI detector to make while-drillingspectroscopy measurements.28 Other logging toolsuse different materials.

Interactions of gamma rays with solidmaterials depend on the energy of the gamma raysand on the density and the atomic number of thematerials being investigated. These interactionscan be classified by the level of energy absorbedby the substrate material.

At lower energy levels, the photoelectriceffect, or Compton scattering, is prevalent. Inthis case, only a fraction of the gamma ray energy is deposited, and the rest leaves thematerial as low-energy photons. At higher

gamma ray energy levels, above 3 MeV, pairproduction becomes dominant.29

Identification of elemental compositions isperformed primarily by measuring the charac-teristic photoelectric energy of individualnuclear varieties when excited by an externalradiation source, such as solar wind or othercosmic rays. At higher energy levels the pair-production mechanism generates well-definedspectra. As such, the most accurate GRSmeasurements were made during periods of highsolar-flare activity when gamma ray energy levelswere at their highest.

To improve the elemental identificationcapability of the GRS, an active detector cup shieldwas designed especially for NEAR. It wasfabricated from a single bismuth germanate[BGO] crystal. The dense BGO cup acted as anactive scintillator while providing direct andpassive shielding from the local gamma rayenvironment and reducing unwanted back-ground signals.

The new design replaced the more expensiveand less reliable long booms used in other missionsto reduce unwanted signals from the activation ofthe spacecraft body itself by cosmic radiation. TheGRS also provided sensitivity to the direction fromwhich the gamma rays were coming.

Detour to a C-Type AsteroidIn early December 1993, NEAR mission managersat The Johns Hopkins University Applied PhysicsLaboratory reviewed a list of asteroids that mightbe in close proximity to NEAR’s flight path (nextpage, top). Asteroid 253 Mathilde was found to bewithin 0.015 AU, or about 2.25 million km[1.4 million miles], of NEAR’s planned orbitalpath. Engineers calculated that with slightchanges in NEAR’s planned trajectory, it couldencounter 253 Mathilde with only a 57 m/s[187 ft/s] change in velocity, well within thespacecraft’s velocity margin.30

Although the dark asteroid was discovered in1985, little was known about Mathilde. Newastronomical observations from ground-basedtelescopes showed it to be a C-type asteroid withan unusual rotation period of 15 days, almost anorder of magnitude slower than most otherknown asteroid rotation periods.

NEAR encountered Mathilde on the way toEros after five trajectory-correction maneuversabout 2 AU from the Sun.31 At this distance,available power from the spacecraft’s solar-powered system was down nearly 75%. With thislimited power, astronomers could use only theMSI to explore the surface of the asteroid, andradio-tracking data, before and after approach,to help determine the mass of the asteroid.

During the flyby, Mathilde exerted a slightgravitational pull on the NEAR spacecraft.Because of Mathilde’s mass, the gravitationaleffects on NEAR’s path were detectable in thespacecraft’s radio-tracking data.

Data from radio-tracking mass estimatesalong with volume approximations helpedscientists calculate the asteroid’s approximatedensity of 1.3 ± 0.3 g/cm3 [81.16 ± 18.73 lbm/ft3].Because of the asteroid’s spectra, Mathilde wasbelieved to be similar in composition tocarbonaceous-chondrite meteorites. However,Mathilde’s density was half of that expected,implying either a high internal porosity orsignificant void space within the asteroid.

Scientists imaged Mathilde over a 25-minuteperiod during the spacecraft’s approach at adistance of 1,200 km [746 miles] and a speed of9.93 km/s [22,213 mi/h]. A total of 534 images

52 Oilfield Review

> XGRS imaging systems. The combined X-ray, gamma ray spectrometer system (XGRS) is shownmounted on the NEAR spacecraft (top left). Shown on the right side of the XRGS instrument is thegamma ray spectrometer. The assembly is mounted to the aft deck of the NEAR spacecraft (top right).The sensor assembly (bottom left) contains the NaI(Tl) detector that is positioned within the bismuthgermanate (BGO) cup shield to reduce unwanted background signals by almost three orders ofmagnitude. The Schlumberger photomultiplier tubes (PMTs) at each end convert the light output ofthe scintillation detectors into electrical signals. (Image and diagram courtesy of NASA/JHUAPL.)

Teflon spacers

Support

Aft deck

Thermalspacers

Gamma raydetector

Connector

Clamp

Spring

SpringTeflon wedge

SmallPMT

Opticalcoupling

NaI (Tl)crystal

BGOshield

LargePMT

Opticalcoupling

58171schD07R1.qxp.52.ps 7/5/06 5:18 PM Page 52

Spring 2006 53

were obtained during this interval at resolutionsranging from 200 to 500 m [656 to 1,640 ft] (above).

Images obtained during the flyby of Mathildeshow an asteroid with a heavily cratered surface.At least four giant craters have diameters thatare comparable to the asteroid’s mean radius of26.5 km [16.5 miles]. The magnitude of theimpacts required to create craters of this size issignificant. Scientists suspect that Mathilde didnot break apart during any of these impacts

because of the asteroid’s high porosity.Laboratory data suggest that cratering in highlyporous targets is governed more by compactionof the target material than by fragmentation andexcavation.32 Cratering processes governed bystructural properties such as porosity producecraters with steep walls, crisp rims and with littleejecta, similar to those imaged on Mathilde.

The images also show Mathilde is remarkablyuniform. The NEAR observations revealed noevidence of any regional albedo, or spectral

variations, implying a homogeneous composition.Further, the measured albedo was consistentwith ground-based telescopic observations.

Although significant data were gained by theMathilde flyby, numerous questions about C-typeasteroids remain unanswered. Mathilde’s densitywas inconsistent with common carbonaceous-chondrite meteorites found on Earth, and theasteroid’s surface appears homogeneous. So, thequestion remains: what connection, if any, existsbetween dark asteroids and meteors found in thesolar system?

Detecting Gamma Ray BurstsGamma ray bursts (GRBs) have remained one ofthe great mysteries of astrophysics since theirdiscovery more than 30 years ago. NASA’s HubbleSpace Telescope made the first observation of anobject associated with a GRB that was detected bythe Italian BeppoSAX satellite in February 1997.33

Scientists believe that GRBs result frommassive explosions in the distant universe thatrelease waves of high-energy photons. GRBsseem to occur daily and emanate from randomparts of the sky. GRBs represent the mostpowerful events known in the universe, emittingin one second as much energy as the Sun willemit in its lifetime. Spectroscopic analyses offaint, but long-lasting GRB optical afterglowshave, in a number of cases, indicated Dopplershifts in the red spectrum that indicate acosmological origin of GRBs.34 Time is critical infollow-up observation efforts, since GRBafterglows fade quickly, in the radio as well asoptical spectrum, making it difficult forastronomers to locate the emission source.

28. For more on while-drilling spectroscopy measurements:Adolph B, Stoller C, Archer M, Codazzi D, el-Halawani T,Perciot P, Weller G, Evans M, Grant J, Griffiths R,Hartman D, Sirkin G, Ichikawa M, Scott G, Tribe I andWhite D: “No More Waiting: Formation Evaluation WhileDrilling,” Oilfield Review 17, no. 3 (Autumn 2005): 4–21.

29. Pair production is the chief method by which energyfrom gamma rays is observed in condensed matter.Provided there is enough energy available to create thepair, a high-energy photon interacts with an atomicnucleus and an elementary particle and its antiparticleare created.

30. Dunham DW, McAdams JV and Farquhar RW: “NEARMission Design,” Johns Hopkins APL Technical Digest 23,no. 1 (2002): 18–33.

31. Cheng et al, reference 23.32. Domingue DL and Cheng AF: “Near Earth Asteroid

Rendezvous: The Science of Discovery,” Johns HopkinsAPL Technical Digest 23, no. 1 (January-March 2002):6–17.

33. The Johns Hopkins University Applied PhysicsLaboratory–Near Spacecraft Gets Unexpected View ofMysterious Gamma-Ray Burst: http://www.jhuapl.edu/newscenter/pressreleases/1998/gamma.htm (accessedApril 5, 2006).

34. NASA–Automatic NEAR-XGRS Data Processing Systemfor Rapid and Precise GRB Localizations with theInterplanetary Network: http://gcn.gsfc.nasa.gov/gcn/near.html (accessed April 5, 2006).

> Destination Eros. The NEAR spacecraft was successfully launched inFebruary 1996, taking advantage of the unique alignment of Earth and Erosthat occurs only once every seven years. A Delta-II rocket placed NEAR intoa two-year Earth gravity-assist trajectory. The gravity-assist maneuver decreasedthe aphelion distance while increasing the inclination from 0 to about 10°.

Sun

Earthorbit

Earth swingby01/22/981,186-km altitude

Eros arrival01/09–02/06/99

Deep-spacemaneuver03/07/97∆V = 215 m/s

Erosorbit

Launch 02/17/96C3 = 25.9 km2/s2

> A quick look at asteroid Mathilde. This view of 253 Mathilde, taken from adistance of about 1,200 km, was acquired shortly after the NEAR spacecraft’sclosest approach to the asteroid. Showing on Mathilde are numerous impactcraters, ranging from more than 30 km [18 miles] to less than 0.5 km [0.3 miles]in diameter. Raised crater rims suggest that some of the material ejected fromthese craters traveled only short distances before falling back to the surface;straight sections of some crater rims indicate the influence of large faults orfractures on crater formation. Mathilde has at least five craters larger than 20 km [12 miles] in diameter on the roughly 60% of the body viewed during theNEAR flyby. (Image courtesy of NASA/JHUAPL.)

20 km

58171schD07R1.qxp.p53.ps 5/25/06 10:49 PM Page 53

Since 1993, astronomers have used speciallyinstrumented spacecraft to help identify thesource of GRBs. These include the Ulyssesspacecraft and several spacecraft near the Earth:the BeppoSAX, Wind observatory, the ComptonGamma-Ray Observatory (CGRO) and the RossiX-Ray Timing Explorer. Unfortunately, these near-Earth spacecraft are too close to each other toallow a definitive triangulation of burst locations.

The loss of the Pioneer Venus orbiter andMars Observer in the early 1990s meant thatastronomers lacked a third detector source foraccurate triangulation of deep-space GRBs. Theaddition of the NEAR spacecraft to theinterplanetary network greatly increased theprobability of associating a GRB with a particularsource using optical and radio telescopes.

The GRS onboard NEAR was not originallyintended to begin its work until the spacecraftreached Eros. However, while en route to Eros,simple software changes to the XGRS systemallowed scientists to use the spectrometer forGRB detection. By adding the NEAR spacecraftto the GRB interplanetary network (IPN) andtaking advantage of significant improvements intelemetry rate and computational capability,NEAR helped reduce GRB detection andtriangulation times from months to seconds.

As an example, gamma ray detectors on theNEAR and Ulysses spacecraft first recordedgamma ray burst GRB000301C on March 1,2000.35 Initially, the sky coordinates of the burstwere not well-defined, but with data from theNEAR and Ulysses spacecraft, an area of the skyabout 4.2 arcminutes wide and 180 degrees inlength was identified as the potential source. Asecond position from the Rossi X-Ray TimingExplorer reduced the error to 4.2 degrees longand 8.7 arcminutes wide. Triangulation of thethree data points further narrowed the gammaray emission zone to within a 50 arcminutesquare, thus allowing a much quicker search ofthe sky by the HST and ground-based telescopes.

Over a 15-month period from December 1999to February 2001, the IPN, including NEAR,detected over 100 GRBs.36 Of these, 34 werelocalized rapidly and precisely enough to allowoptical and radio telescope follow-up observa-tions. The suspected GRB emission locations weredetermined with accuracies of the order ofseveral arcminutes. One of the most interestingresults was the detection of a GRB originating inthe southern constellation Carina. Opticalobservations of an extreme red-shift indicatedthat the source of the GRB was about 12.5 billionlight-years from Earth, making it the mostdistant GRB yet detected.

Unlocking the Secrets of ErosThe NEAR spacecraft entered Eros orbit onFebruary 14, 2000, beginning its one-year missionto explore Eros. Orbital characteristics rangedfrom elliptical to circular and took NEAR within35 km [22 miles] of the surface of Eros. Then,almost six years to the day after launch,engineers at JHUAPL brought NEAR’s mission toits culmination with a successful controlleddescent to the surface of Eros.

Although the primary mission of NEAR was toinvestigate the mineralogy, composition,magnetic fields, geology and origin of Eros, NEARobtained much more detailed information duringits orbital encounter with Eros.

Images, laser altimetry and radio-sciencemeasurements provided strong evidence thatEros is a consolidated, yet fractured asteroidwith a regolith cover varying dramatically indepth from near zero to as much as 100 m[328 ft] in some areas.37 Scientists believe thatthe presence of joined and well-defined craters isindicative of cohesive strength within theasteroid. Surface images show the geometricrelationship of grooves and cuts in the surface,suggesting that the rock is competent and not aloosely bound agglomeration of smaller rocks.

The gravity field on Eros appeared to beconsistent with that expected from a uniform-density object of the same shape. The measureddensity of Eros indicates that it has a bulkporosity of 21 to 33%, implying that even thoughthe asteroid’s mass is uniformly distributed, it issignificantly porous and potentially fractured,but to a lesser extent than Mathilde.

Imaging at resolutions of a few centimetersper pixel revealed a complex and active regoliththat has been significantly modified andredistributed by gravity-driven slope processes.High-albedo features noted in images takenaround crater walls that slope in excess of 25°were often 1.5 times brighter than theirsurroundings, indicating recent changes insurface characteristics due to regolith sloughing(above right).38

Silicate mineralogy analysis performed by theNIS was consistent with ordinary chondritemeteorites. Spatially resolved measurements of theasteroid’s surface provided no evidence for mineralcompositional variation. Scientists believe that thespectral uniformity of Eros may have resulted froma uniformly high degree of space weatheringcaused by micrometeorite bombardment.

The NEAR spectrographs, XRS, GRS and NISmeasured the elemental and mineral composi-tion of Eros. Data acquired by the XRS duringorbiting showed calcium, aluminum, magnesium,iron and silicon abundances consistent withordinary chondrite and certain primitiveachondrite meteorites. However, the level ofsulfur typical of chrondritic meteorites wasabsent or depleted on Eros.

Although the surface of Eros appears to beelementally homogeneous, the XRS instrumentcan measure only surface composition, so it isunknown if the sulfur depletion is a surface effector consistent through the core of the asteroid. Ifthe sulfur depletion is consistent across the bulkof the asteroid, this would imply an associationwith primitive achondrite meteorites.

The orbital GRS measurements had lowersignal levels than predicted, so the elementalratios with the highest precision were measuredafter landing. GRS data showed the Mg/Si andSi/O ratios and the abundance of K to beconsistent with chondritic meteorite values, butfound Fe/Si and Fe/O levels to be lower thanwhat would be expected with chrondriticmeteorites. Since these measurements were

54 Oilfield Review

> Close-approach Eros crater wall. Material on theinner wall of the crater in the center of the imageis brighter than the surrounding regolith and isthought to be subsurface material exposed whenoverlying, darker regolith slid off. The field ofview is 1.2 km [0.7 miles] across, taken from 38km [24 mi] above Eros. (Images courtesy ofNASA/JHUAPL.)

58171schD07R1.qxp.p54.ps 5/25/06 10:49 PM Page 54

Spring 2006 55

made after landing and the GRS instrument canprobe tens of centimeters below the surface,these measurements reflect a volume of about1 m3 [35.3 ft3] around the detector. From theGRS data alone, scientists could not determinewhether the Fe depletion is a global composi-tional property of Eros or a localized property ofthe landing zone.

Although the XGRS system observed Erosduring a one-year orbital period, the useful timefor data collection was considerably shorter.Engineers were limited by the angular require-ments of the solar panels relative to the sun,telemetry time and periods when the surface ofEros was properly lit by the Sun. In the end,scientists found that the best quality composi-tional data were acquired during low-altitudeorbits and after landing on Eros (right). OnceNEAR was on the surface, the gamma rayspectrometer obtained in-situ measurements ofthe regolith for a period of about 14 days.39

The surface composition of Eros suggests thatthe asteroid is similar in bulk composition to arange of meteorites that have experiencedminimal thermal alteration since their formationat the birth of the solar system. Scientists believethat Eros is primitive in its chemical compositionand has not experienced differentiation into acore, mantle and crust. Differences between XRSand GRS data in Fe/Si ratio and an apparentdeficiency of sulfur at the surface of Eros couldreflect either alteration processes in the regolithduring the last millions to billions of years orpartial melting in the first 10 million years ofsolar system history.

These spectral measurements providedscientists with a new set of questions. While thespectral observations are consistent with anordinary chondritic meteorite composition, themeasurements did not establish an undisputed linkbetween Eros and a specific meteorite type. Thequestion remains whether Eros is unrelated to anyknown meteorite type, or is actually a chondritetype at depth, below the surface layers that mayhave been altered by weathering processes.

35. NASA–Amateurs Catch a Gamma-Ray Burst:http://science.nasa.gov/headlines/y2000/ast14mar_2m.htm (accessed April 5, 2006).

36. Trombka JI et al: NASA Goddard Space Flight Center:http://www.dtm.ciw.edu/lrn/preprints/4631trombka.pdf(accessed April 5, 2006).

37. Domingue and Cheng, reference 32.38. Domingue and Cheng, reference 32.39. Trombka et al, reference 36.

> Landing on Eros. The location of NEAR Shoemaker’s planned landing site (top right) is shown in thisimage (yellow circle) mosaic taken on December 3, 2000, from an orbital altitude of 200 km [124miles]. NEAR’s imaging systems were recording (bottom 4 images) as the spacecraft performed acontrolled landing on the surface of Eros. At a range of 1,150 m, NEAR captured an image that spans54 m [177 ft] of the asteroid’s surface (1). The large rock at the lower left corner of the imagemeasures 7.4 m [24 ft] across. NEAR then recorded images at ranges of 700 m (2), 250 meters (3),followed by the last image before landing (4) at a range of 120 m [394 ft]. The field of view in this finalimage measures 6 m [20 ft] across. The large rock at the top of the image measures 4 m [12 ft]across. The streaky lines at the bottom indicate loss of signal as the spacecraft touched down on theasteroid during image transmission. Once on the surface, the GRS system generated gamma rayspectrum data for a period of seven days (graph, top left). These scientific data were the first evercollected on the surface of an asteroid. The gamma ray instrument has two sensors (red and bluelines) that detected clear signatures of key elements in the composition of Eros. These data, whichsurpass in quality all the data accumulated by this instrument from orbit, helped the NEAR scienceteam relate the composition of Eros to that of meteorites that have fallen to Earth. (Images courtesyof NASA/JHUAPL.)

1

2

3

4

Final Eros images: range 1,150 m (3,773 ft)

Final Eros images: range 700 m (2,300 ft)

Final Eros images: range 250 m (820 ft)

Last, closest image of Eros

106

105

104

0 2,000 4,000Energy, keV

IronPotassium

Silicon

Oxygen

NEAR Surface GRS Spectra

Iron

Rela

tive

inte

nsity

6,000 8,000 10,000

103

Silicon Oxygen

Iron

Outer detector

Inner detector

58171schD07R1.qxp.p55.ps 5/25/06 10:49 PM Page 55

Scientists were surprised that Eros appearsto have little or no magnetic field. Mostmeteorites, including chondrites, tend to bemore magnetized than Eros. Perhaps its lowlevels of iron and the fact that it never has beenheated to melting play a role in thisdifferentiation. The spectral homogeneity of Eroscombined with gravity-field measurements,structural characteristics and indications ofstructural coherence suggests that Eros is acollisional fragment of a larger parent body.

The NEAR mission, a mission of many firsts inNASA’s Discovery Program, substantiallyincreased our knowledge of primitive bodies inour solar system. Although the data returned byNEAR have revealed many secrets of asteroids,many questions remain unanswered, and morewill be learned from future missions.

Exploring Gas GiantsThe goal of the Cassini mission is to exploreSaturn, its many known moons and those yet tobe discovered. Managed by NASA’s Jet PropulsionLaboratory (JPL) in Pasadena, California, USA,Cassini is a joint endeavor of NASA, theEuropean Space Agency (ESA) and the Italian

space agency, Agenzia Spaziale Italiana (ASI). Itis one of the most ambitious efforts in planetaryspace exploration.40

Because of the low level of sunlight reachingSaturn, solar arrays are not feasible as a powersource. Engineers employed a set of radioisotope-thermoelectric generators similar to those used

56 Oilfield Review

40. NASA/Jet Propulsion Laboratory–Cassini Mission toSaturn: http://www.jpl.nasa.gov/news/fact_sheets/cassini.pdf (accessed April 13, 2006).

> Preparing Cassini for flight. Technicians reposition and level the Cassini orbiter in the Payload Hazardous Servicing Facility at the Kennedy Space Centerin July 1997, after stacking the craft’s upper equipment section on the propulsion module (left). The orbiter’s primary systems are shown (right). (Imagescourtesy of NASA/JPL.)

4-m high-gain antennaLow-gain antenna (1 of 2)

11-m magnetometerboom

Radio/plasma wavesubsystem antenna(1 of 3)

Remote sensinginstruments

445 N engine (1 of 2)

Radar bay

Huygens Titanprobe

Radioisotopethermoelectricgenerator (1 of 3)

> Imaging Saturn’s rings. The Ultraviolet Imaging Spectrograph (UVIS) is a setof telescopes used to measure ultraviolet light from the Saturn system’satmospheres, rings and surfaces. The UVIS has two spectrographic channelsor instruments: the extreme ultraviolet channel and the far ultraviolet (FUV)channel. Each instrument is housed in aluminum cases, and each contains areflecting telescope, a concave grating spectrometer and an imaging, pulse-counting detector. The UVIS also includes a high-speed photometer (HSP)channel, a hydrogen-deuterium absorption cell (HDAC) channel and anelectronic and control subassembly. (Image courtesy of NASA/Laboratory for Atmospheric and Space Physics.)

HDAC

FUV spectrograph

HSP

58171schD07R1.qxp.p56.ps 5/25/06 10:49 PM Page 56

Spring 2006 57

on the previous Galileo and Ulysses missions.With these systems, heat from the natural decayof plutonium-238 is used to generate electricityto operate Cassini’s systems.

The Cassini spacecraft is equipped with18 instruments, 12 on the orbiter and another sixon the Huygens probe, which is designed toseparate from the main spacecraft andparachute through the atmosphere of Titan,Saturn’s largest moon. The 12 instruments on theorbiter are currently conducting in-depth studiesof Saturn, its moons, rings and magneticenvironment (previous page, bottom).

Key to Cassini’s science mission is theUltraviolet Imaging Spectrograph (UVIS), aninstrument based on Schlumberger sensors andpackaging, and designed to operate in harshenvironments like those found in oil and gaslogging operations (previous page, right). TheUVIS is now helping scientists determineatmospheric chemistry, the nature of clouds andring systems, and the atmospheric energybalance on Saturn and its moon Titan.

The UVIS comprises a set of telescopes thatmeasure ultraviolet light from the Saturnsystem’s atmospheres, rings and surfaces. Theinstrument has two spectrographs: the farultraviolet channel (FUV), 110 to 190 nm, and theextreme ultraviolet channel (EUV), 56 to 118 nm.

The FUV and EUV channels in the UVISspectrometer require different detectors tooptimize sensitivity to the wavelength rangerequired by the Cassini project. In cooperationwith the Laboratory for Atmospheric and SpacePhysics (LASP) at the University of Colorado,Schlumberger designed the detector response tomeet these requirements.

The FUV detector was assembled using acesium iodide photocathode with a magnesiumfluoride window. This detector was vacuum-sealed and included an integrated pump thatmaintained an ultrahigh vacuum during thespacecraft assembly and launch. Once in space,the detector was equalized to the vacuum ofspace for the voyage to Saturn.

The EUV detector utilizes a potassiumbromide photocathode and has no window sincetransmission of all known substances is very poorin this short wavelength range. Fortunately,potassium bromide is a very robust photocathodeand can be exposed to dry air for the short timerequired for testing and assembly. Once in thevacuum of space, the detector cover was opened,allowing light to enter the instrument.

Both detectors utilize specially selectedmicrochannel plates (MCP). MCP technology hasa long history in spaceflight imaging instruments.

Quality-control procedures during manufacturingallowed only MCPs with very low-defect densitiesto be used for final assembly. Once an MCP wasavailable, LASP and Schlumberger scientistsworked together during the final assemblyprocess. The units were then transported toNASA laboratories for final testing.

Two FUV and two EUV detectors that met thestringent quality requirements for space travel toSaturn were assembled at the SchlumbergerPrinceton Technology Center (PTC) in NewJersey. One pair of detectors was designated asflight units while the second set was kept inreserve as a backup.

The UVIS also includes a high-speedphotometer (HSP) channel, a hydrogen-deuterium absorption cell (HDAC) channel andelectronic and control subassemblies. Scientistsare using the HSP to make stellar occultationmeasurements of the structure and density ofmaterial in Saturn’s rings.

Cassini was launched on October 15, 1997,from Cape Kennedy, Florida, aboard a TitanIVB/Centaur rocket, the most powerful launchvehicle in the US fleet (above). After Cassini wasplaced in orbit around Earth, the upper stagefired to send Cassini on an interplanetarytrajectory that would eventually deliver thespacecraft to Saturn.

> Launching Cassini. A Titan IVB/Centaur launch vehicle propelled the Cassini spacecraft and its attachedHuygens probe into space from Cape Kennedy Air Station’s Launch Complex 40, Florida. Visible in thisview are the 20-m [66-ft] long, 5-m [17-ft] wide payload atop the vehicle holding the Cassini spacecraft.Cassini’s planned interplanetary flight path (chart inset) began with launch from Earth on October 15,1997, followed by gravity assist flybys of Venus, Earth and Jupiter. The gravity-assist flybys of thedifferent planets are designed to increase the spacecraft’s velocity relative to the Sun so it can reachSaturn. With the gravity-assist trajectory, it took more than 61⁄2 years for the Cassini spacecraft toarrive at Saturn. (Images courtesy of NASA.)

Orbit of Jupiter

Saturn Arrival07/01/04

Jupiter swingby12/30/00

Orbit ofSaturnOrbit of Venus

Launch 10/15/97

Orbit of Earth

Deep-space maneuver12/03/98

Earth swingby 08/18/99

Venus swingby 06/24/99

Venus swingby 04/26/98

58171schD07R1.qxp.p57.ps 5/25/06 10:49 PM Page 57

Cassini flew twice past Venus, then once pastEarth and Jupiter. The spacecraft’s speedrelative to the Sun increased as it approachedand swung around each planet, giving Cassini thecumulative boost it needed to reach Saturn withminimal fuel consumption. After reachingSaturn, Cassini fired its main engine for about 96minutes, reducing the spacecraft’s speed andallowing it to be captured in an orbit aroundSaturn. On January 5, 2005, Cassini released theEuropean-built Huygens probe toward Titan.

Journey to a Distant MoonWith a diameter larger than the planet Mercury,Titan is one of the most interesting moons in thesolar system. The surface of this moon lieshidden beneath an opaque atmosphere morethan 50% denser than that of Earth (left).

Titan’s atmosphere is filled with a brownish-orange haze composed of complex organicmolecules falling like rain from the sky to thesurface. Most scientists agree that conditions onTitan are too cold for life to have evolved—although there are theories concerning thepossibility of life forms in covered lakes of liquid hydrocarbons warmed by the planet’sinternal heat.

The Huygens probe entered Titan’s atmos-phere on January 14, 2005, deployed itsparachutes and began its scientific observationsduring a descent through the moon’s denseatmosphere lasting close to 21⁄2 hours (belowleft).41 Instruments onboard the probe detected asurface temperature of 94K at the landing site.Images taken by the probe while descendingshowed surface channels that appeared toindicate rain or fluid flow, possibly in the form ofliquid methane. Ridges as tall as 100 m wereobserved near the landing area (next page, top).High quantities of methane were detected in thelower atmosphere, with nitrogen predominatingin the upper atmosphere. Oxygen was notdetected probably because it is tied up as frozenwater. This would also prevent the formation of carbon dioxide.

Laboratory tests recreated the impactmeasurements derived from the onboardpenetrometer. These tests indicate that thesurface in the landing area may be composed offine particles with a thin crust. Accelerometermeasurements suggest the probe settled 10 to15 cm [4 to 6 in.] into the surface. Heat frominstruments then evaporated liquid methane inthe soil and released it around the spacecraft asmethane gas. The Huygens probe continued

58 Oilfield Review

> Titan image. In this infrared view of Titan, features on the leading hemisphereare shown, including the bright, crescent-shaped Hotei arcus (right of center),often referred to as “the smile” by researchers. The view is centered on thebright region called Xanadu. The image was taken with the Cassini spacecraftnarrow-angle camera using a spectral filter sensitive to wavelengths ofinfrared light centered at 938 nm. The image was acquired at a distance ofapproximately 1.3 million km [800,000 miles] from Titan. (Image courtesy ofNASA/JPL/Space Science Institute.)

> Descent to Titan. The Huygens probe analyzed Titan’s atmosphere andrecorded a significant amount of data and images on its journey to thesurface of Titan. (Image courtesy of NASA/JPL.)

1,000

500

300

192

Altit

ude,

km

0 2.5

Time, hours after entry

Entry speed: 6.2 km/s

Peak deceleration: 10gn to 25gn

Main chute deploys

Drogue chute deploys

Probeperformssurfacescience

Deceleratorjettisons

Instrumentinlet-portopens170

0

58171schD07R1.qxp.p58.ps 5/25/06 10:49 PM Page 58

Spring 2006 59

making measurements and transmitting data toCassini for 72 minutes after landing until powerlimitations and deterioration of the spacecraftdue to extreme surface conditions on Titanresulted in loss of signal.

Exploring the Ringed PlanetAside from Titan, more moons of greater varietyorbit Saturn than any other planet. So far,observations from Earth and by spacecraft havefound Saturnian satellites ranging from smallasteroid-size bodies to those as large as Titan.

Saturn is the second-largest planet in thesolar system. Like the other gaseous outerplanets—Jupiter, Uranus and Neptune—it hasan atmosphere made up mostly of hydrogen andhelium, and like them, it is ringed. Saturn’sdistinctive bright rings are made up of ice androck particles ranging in size from grains of sandto small houses.

Although the face of Saturn appears calm, theplanet has a windswept atmosphere where anequatorial jet stream blows at 1,800 km/h[1,118 mi/h], and swirling storms churn beneaththe cloud tops. Early explorations by NASA’sPioneer 11 spacecraft in 1979, and the Voyager 1and 2 spacecraft in 1980 and 1981, found Saturnto have a huge and complex magnetic environ-ment where trapped protons and electronsinteract with each other, the planet, the ringsand the surfaces of many of Saturn’s moons.

From Earth, Saturn’s rings appear as only afew monolithic bands, while in reality, theyconsist of thousands of rings and ringlets, withparticles sometimes arranged in complicatedorbits by the gravitational interaction of smallmoons previously unseen from Earth (right).Scientists are using data from the UVIS indetailed computer models to simulate thecomplex motion of these rings.

Second in size only to Jupiter, Saturn hasmore than 750 times the volume of Earth.Combined with the planet’s low density,less than half that of water, its fast rotationpromotes a bulge of material near the equator.Saturn is shaped like a flattened ball; its pole-to-pole diameter is only 108,728 km [67,560 miles],compared to about 120,536 km [about74,898 miles] for the equatorial diameter.

41. European Space Agency–Cassini-Huygens:http://huygens.esa.int/science-e/www/object/index.cfm?fobjectid=36396 (accessed April 13, 2006).

> Exploring Saturn’s rings. Images taken during the Cassini spacecraft’s orbit around Saturn showcompositional variation in Saturn’s rings (top). The red in the image indicates sparser ringlets thatprobably comprise “dirty,” and possibly smaller particles than those in the icier turquoise ringlets. The red band roughly three-fourths of the way outward is known as the Encke Gap. This image wastaken with the Ultraviolet Imaging Spectrograph (UVIS) instrument, which is capable of resolving therings to show features up to 97 km [60 mi] across, roughly 100 times the resolution of ultraviolet dataobtained by the Voyager 2 spacecraft. The false-color view of Saturn’s A ring (bottom left) was alsotaken by the UVIS. The ring is the bluest in the center, where the gravitational clumps are the largest.The thickest black band in the ring is the Encke Gap, and the thin black band farther to the right is theKeeler Gap. The insert (bottom right) is a computer simulation about 150 m [490 ft] across, illustratinga clumpy region of icy particles in the A ring. (Images courtesy of NASA/JPL/University of Colorado.)

> Under Titan’s atmosphere. The perspective view of the surface of Titan near the Huygens probelanding site (top) is color-coded, with blue the lowest altitude and red the highest. The total areacovered by the image is about 1 by 3 km [0.6 by 2 miles]. A pair of images (inset) was acquired fromthe Huygens descent imager/spectral radiometer. The left image was acquired from 14.8 km [9 miles]above the surface with the high-resolution imager and the right from 6.7 km [4 miles] altitude with themedium-resolution imager. (Images courtesy of ESA/NASA/JPL/University of Arizona/USGS.)

58171schD07R1.qxp.p59.ps 5/25/06 10:49 PM Page 59

Unlike rocky inner planets such as Earth,Saturn has no surface on which to land. Aspacecraft descending into its atmosphere wouldsimply find the surrounding gases becomingdenser, and the temperature progressivelyhotter; eventually the craft would be crushed andmelted. Detailed analysis of Saturn’s gravita-tional field leads astronomers to believe that thedeepest interior of Saturn must consist of amolten rock core about the same size as theplanet Earth, but much denser.

Spectroscopic studies by the Voyagerspacecraft found Saturn to be made up of about94% hydrogen and 6% helium. Hydrogen andhelium are the primary constituents of all thegiant gas planets, the Sun and the stars. Gravityat the top of Saturn’s clouds is similar to thatnear the surface of Earth. The temperature nearthe cloud tops is about -139°C [-218°F],increasing toward the planet’s core due toincreased atmospheric pressure. At the core,Saturn’s temperature is predicted to be about10,000°C [18,000°F].

On June 21, 2005, the UVIS detected auroralemissions from both Saturn’s northern andsouthern poles (above right).42 These emissionsare believed to be similar to Earth’s NorthernLights yet are invisible to the naked eye.Ultraviolet images captured the entire oval of theauroral emissions from hydrogen gas excited byelectron bombardment. Time-lapse imagesindicate that aurora lights are dynamic,responding rapidly to changes in the solar wind.

New MoonsThere were only 18 known moons orbiting Saturnwhen the Cassini spacecraft began its mission toSaturn in 1997. During Cassini’s seven-yearjourney, Earth-based telescopes uncovered 13more moons. Soon after the spacecraft reachedSaturn, the Cassini team discovered two moretiny moons, Methone and Pallene. The two newmoons are approximately 3 km [1.8 miles] and4 km [2.5 miles] across.

Scientists suspected that more tiny Saturnianmoons might be found within the gaps in Saturn’srings. On May 1, 2005, using a sequence of time-lapse images from Cassini’s cameras, astrono-mers confirmed the presence of a tiny moonhidden in a gap in Saturn’s A ring.43 The images

show the tiny object in the center of the KeelerGap and the wavy patterns in the gap edges thatare generated by the moon’s gravitationalinfluence (above).

The new object, Daphnis, is about 7 km[4 miles] across and reflects about half the lightfalling on it—a brightness that is typical of theparticles in the nearby rings. As Cassinicontinues to explore Saturn and its moons,scientists expect to uncover more of the secretsof this vast planetary system.

Signs of an AtmosphereAlthough the moon Enceladus is covered with icecomposed of water, like Saturn’s other moons, itdisplays an abnormally smooth surface with veryfew impact craters. With a diameter of only500 km [310 mi], Enceladus would fit into the state of Arizona. Yet despite its small size,Enceladus exhibits one of the most interestingsurfaces of all the icy satellites. Enceladusreflects about 90% of the incident sunlight as if

60 Oilfield Review

42. Laboratory for Atmospheric and Space Physics–Cassini-UVIS Mission to Saturn and Titan:http://lasp.colorado.edu/cassini/whats_new/ (accessed April 13, 2006).

43. NASA/Jet Propulsion Laboratory–Cassini Finds anActive, Watery World at Saturn’s Enceladus: http://www.nasa.gov/mission_pages/cassini/media/cassini-072905.html (accessed April 13, 2006).

44. NASA/Jet Propulsion Laboratory, reference 43.

> Perturbations caused by a tiny moon. This image confirmed earlier suspicions that a small moonwas orbiting within the narrow Keeler Gap in Saturn’s A ring. The Keeler Gap is located about 250 km[155 miles] inside the outer edge of Saturn’s A ring, which is also the outer edge of the bright mainrings. The new moon, Daphnis, is about 7 km across and reflects about 50% of incident sunlight.Scientists predicted the moon’s presence and its orbital distance from Saturn after July 2004, whenthey saw perturbations in the ring structure of the Keeler Gap’s outer edge. These images wereobtained with the Cassini spacecraft narrow-angle camera on May 1, 2005, at a distance ofapproximately 1.1 million km [680,000 miles]. (Image courtesy of NASA/JPL/Space Science Institute.)

Moon

Perturbationscaused by moon

> The southern lights of Saturn. Images of Saturn obtained by Cassini’s UVISshow auroral emissions at its poles similar to Earth’s Northern Lights. The twoUV images are the first from the Cassini-Huygens mission to capture the entire“oval” of the auroral emissions at Saturn’s southern pole. They also showsimilar emissions at Saturn’s north pole. In these false-color images, bluerepresents aurora emissions from hydrogen gas excited by electronbombardment, while red-orange represents reflected sunlight. These imageswere taken 1 hour apart; during this time the brightest spot in the left auroraimage fades and a bright spot appears in the middle of the aurora in the rightimage. (Images courtesy of NASA/JPL/University of Colorado.)

58171schD07R1.qxp.p60.ps 5/25/06 10:49 PM Page 60

Spring 2006 61

covered with fresh-fallen snow, placing it amongthe most reflective objects in the solar system.Although Enceladus was previously thought to bea cold and dead rock mass, data from the Cassinispacecraft indicate evidence of ice volcanism,which might explain its smooth surface features.

In July 2005, Cassini’s instruments detected acloud of water vapor over the moon’s southernpole and warm fractures where evaporating iceprobably supplies the vapor cloud.44 So far,Enceladus is the smallest body found thatdisplays evidence of active volcanism. Scientiststheorize that warm spots in the moon’s icy andcracked surface are probably the result of heatfrom tidal energy like the volcanoes on Jupiter’smoon Io. Its geologically young surface of water-base ice, softened by heat from below, resemblesareas on Jupiter’s moons, Europa and Ganymede.

Cassini flew within 175 km [109 miles] ofEnceladus on July 14, 2005. Data collectedduring that flyby confirm an extended anddynamic atmosphere. This atmosphere was firstdetected by Cassini’s magnetometer during adistant flyby earlier in 2005 (above left).

Cassini’s magnetometer detected distur-bances in the magnetic field caused by smallcurrents of ionized gas from the atmospherearound this moon. These could be detected bythe instrument long before imaging instrumentscould be applied to confirm this finding.

As Cassini approached this small body,imaging instruments were able to makemeasurements that showed gas composition,further confirming the presence of anatmosphere. The ion and natural massspectrometers and the UVIS showed that thesouthern atmosphere contains water vapor(left). The mass spectrometer found that watervapor comprises about 65% of the atmosphere,with molecular hydrogen at about 20%. The restis mostly carbon dioxide and some combinationof molecular nitrogen and carbon monoxide. Thevariation of water-vapor density with altitudesuggests that the water vapor may come from alocalized source comparable to a geothermal hotspot. The ultraviolet results strongly suggest alocal vapor cloud. The fact that the atmospherepersists on this low-gravity world, instead ofinstantly escaping into space, suggests that themoon is geologically active enough to replenishthe water vapor at a slow, continuous rate.

High-resolution images show that the southpole has an even younger and more fracturedappearance than the rest of Enceladus, complete

> Shifting magnetic fields. This artist’s conception shows the detection of adynamic atmosphere on Saturn’s icy moon Enceladus. The Cassinimagnetometer is designed to measure the magnitude and direction of themagnetic fields of Saturn and its moons. During Cassini’s three close flybysof Enceladus on Febuary 17, March 9, and July 14, 2005, the instrumentdetected a bending of the magnetic field around Enceladus thought to becaused by electric currents generated by the interaction of atmosphericparticles and the magnetosphere of Saturn. The graphic shows the magneticfield observed by Cassini, as well as the predicted vapor cloud being ventedfrom the south pole of Enceladus. Cassini’s magnetometer observed bendingof the magnetic field consistent with its draping around a conducting object.(Image courtesy of NASA/JPL.)

Hot plasma flow

Saturn

Enceladus

Vapor cloud

> Indications of an atmosphere. On July 11, 2005, the Cassini ultraviolet imaging spectrographobserved the star Bellatrix as it passed behind Enceladus, as seen from the spacecraft. The starlightwas observed to dim when it got close to Enceladus, indicating the presence of an atmosphereisolated to the southern pole (A). The ultraviolet imaging spectrograph indicated that the atmospherewas water vapor, based on absorption features in the spectrum of the star. The colors show theundimmed star signal (blue) versus the dimmed star signal (red). As Bellatrix reemerged from behindEnceladus, there was no observed dimming of the starlight. In another occultation (B) of the starLambda Scorpius, no sign of an atmosphere was detected, implying that the atmosphere is localizedtoward the southern pole. (Image courtesy of NASA/JPL/Space Science Institute.)

A

Starlight

Bellatrix

Lambda Scorpius

Starlight

July February

B

58171schD07R1.qxp.p61.ps 5/25/06 10:49 PM Page 61

with icy boulders the size of large houses andlong, bluish cracks or faults (left).

Another Cassini instrument, the compositeinfrared spectrometer (CIRS), demonstratesthat the southern pole is warmer thananticipated (below left). Temperatures near theequator were found to reach a frigid 80K.Scientists believe that the poles should be evencolder because of the low level of energy receivedfrom the Sun. However, south polar averagetemperatures reached 85K, much warmer thanexpected. Small areas of the pole, concentratednear the fractures, are even warmer: higher than140K in some places.

Scientists find the temperatures difficult toexplain if sunlight is the only heat source. Morelikely, a portion of the polar region, includingobservable fractures, is warmed by heat escapingfrom the interior. Evaporation of this “warm” iceat several locations within the region couldexplain the density of the water-vapor clouddetected by Cassini’s instruments. How a 500-km[310-mile] diameter moon can generate thismuch internal heat and why it is concentrated atthe southern pole are still a mystery.

Similar to multiple well-logging instrumentsworking together deep beneath the Earth’ssurface, the discovery of an atmosphere onEnceladus resulted from an array of differentsensors working in synergy to acquire data andmaximize scientific value.

The Challenge of SpaceAdvances in technology, particularly during thelast 100 years, have helped change the way weview the Earth, our solar system and the universebeyond. From the E&P industry’s early begin-nings, engineers, geoscientists and many otherdedicated men and women have led the way inexploration of our inner space environment.Today, this same innovative spirit, and in manycases, similar technologies, are taking us beyondthe confines of Earth’s environment into the vastunknowns of outer space.

The examples presented in this article arejust a few of the contributions made by theoilfield service industry to space exploration. Inthe future, we can expect to see more terrestrialtechnology applied in the quest for extrater-restrial understanding. The late astrophysicistCarl Sagan wrote, “Imagination will often carryus to worlds that never were. But without it, wego nowhere.”45 It is this imagination andcreativity that have driven the E&P industry toexplore deep beneath the Earth’s surface andthat will inevitably launch the first drillingexpeditions to Mars and beyond. —DW

62 Oilfield Review

> A hot southern pole. This map represents the surface temperature ofEnceladus as seen by the composite infrared spectrometer. The observedtemperatures included an unexpected hot spot at the south pole. On averagethe region is 15K warmer than expected; in some places hot spots greaterthan 140K were observed. The hottest spots line up with the blue fracturestripes visible in the previous image (above). (Images courtesy of NASA/JPL/Goddard Space Flight Center.)

Enceladus Temperature Map

Predicted temperatures Observed temperatures

85

75

80

70

Tem

pera

ture

, kel

vin

65

> Imaging Enceladus. This view (top left) is a mosaic of four high-resolution images taken by theCassini spacecraft narrow-angle camera during its close flyby of Saturn’s moon Enceladus. The viewis about 300 km [186 miles] across and shows a myriad of faults, fractures, folds, troughs and craters.The images were taken in visible light at distances ranging from of 26,140 to 17,434 km [16,246 to10,833 miles]. The southern polar terrain of Enceladus (bottom left) appears strewn with greatboulders of ice in the wide-angle camera image; more details are shown in the high-resolution,narrow-angle camera image (inset). The two images were acquired at an altitude of approximately208 km [129 miles]. The enhanced color view of Enceladus (right) is principally of the southernhemisphere. The south polar terrain is marked by a striking set of ‘blue’ fractures and encircled by a conspicuous and continuous chain of folds and ridges. This mosaic is a false-color view thatincludes images taken at wavelengths from the ultraviolet to the infrared portion of the light spectrum.(Images courtesy of NASA/JPL/Space Science Institute.)

45. Sagan C: Cosmos. New York City: Carl SaganProductions and Random House (1980): 4.

58171schD07R1.qxp.p62.ps 5/25/06 10:49 PM Page 62

Spring 2006 63

J.L. Arroyo Franco, who is based in Reynosa, Mexico,is Team Leader for the Pemex Exploración yProducción Herreras area group (North of Mexico).Prior to transferring to this position in early 2006, hewas group geophysicist and then leader of theCuitlahuac group. He has worked on gravimetric and2D and 3D seismic studies for the company since 1980.He earned a degree in geophysical engineering fromUniversidad Nacional Autónoma de México in MexicoCity, and an MS degree in administration atUniversidad Autónoma de Chihuahua, Mexico.

Gopa S. De is a Research Consultant with ChevronEnergy Technology Company in San Ramon, California,USA. She began her career with Chevron Oil FieldResearch Company in 1982. Her major research inter-ests are sonic logging and rock physics. She has a PhDdegree in condensed matter physics from theUniversity of California, San Diego. Gopa is a memberof the American Physical Society, the Society ofExploration Geophysicists (SEG), the SEG ResearchCommittee and the SPE Reservoir Evaluation &Engineering (SPEREE) Review Board.

Joel Lee Groves is Principal Research Scientist at theSchlumberger Princeton Technology Center (PTC),New Jersey, USA. His major projects are pulsed-neutrongenerator systems, X-ray generators and multiphaseflowmeters. He holds a BS degree in physics and an MS degree in experimental physics, both from WestVirginia University, Morgantown, USA. He also has aPhD degree in nuclear physics from University ofIllinois at Urbana-Champaign, USA. He joinedSchlumberger-Doll Research (SDR) as a research scientist in Ridgefield, Connecticut, USA, in 1984 after10 years at the University of Illinois and ColumbiaUniversity, New York City. He was director of theNuclear Science department at SDR when he left tojoin PTC in 1995. While at PTC, he has worked as radi-ation safety officer, minitron manufacturing engineerand director of research and engineering.

Jakob B.U. Haldorsen received Cand. Mag. (BS) andCand. Real. (MS) degrees in physics from the Universityof Oslo in Norway. He then spent six years in researchand teaching at the University of Oslo and at theEuropean Organization for Nuclear Research (CERN)in Geneva, Switzerland. After joining Geco in 1981, heworked in many different positions, including R&Eproject manager in Oslo and then in Houston. AfterGeco became part of Schlumberger in 1987, he wastransferred to SDR in Ridgefield, Connecticut, as amember of the Geoacoustics department. Later hemoved to Schlumberger Cambridge Research (SCR) inEngland as a member of the Seismic department andthen to Geco-Prakla in Hannover, Germany, to work onalgorithmic and physics problems related to dataacquired in high-noise environments. Jakob returnedto SDR in 1995 as program leader for the SurfaceRadar program and is now a Scientific Advisor respon-sible for high-resolution formation imaging using sonicand borehole seismic tools.

David Linton Johnson joined the rock-physics programat SDR, Ridgefield, Connecticut, in 1979, and is currently Scientific Advisor and Program Manager ofthe Sensor Physics department. He is responsible forvarious linear and nonlinear borehole acoustic projectsand remains active in research on properties of granu-lar or porous media. David received his BS degree inphysics from the University of Notre Dame, Indiana,USA, and earned MS and PhD degrees in theoreticalphysics from the University of Chicago. Before joiningSchlumberger, he was a faculty member of NortheasternUniversity, Boston, Massachusetts, USA. David is theauthor of numerous publications and holds severalpatents. He is a Fellow of the American Physical Society.

Andreas Kayser worked for Schlumberger from 2003until he recently took a position with BP in Sunbury,England. As a development engineer at SCR, England,Andreas was responsible for X-ray tomography analysisand oversaw data acquisition, conversion, visualizationand interpretation. He was also involved in a broadrange of projects using Inside Reality* virtual realitytechnology and Petrel* seismic-to-simulation softwarefor interpretation, visualization and complex well plan-ning. Andreas then moved to Schlumberger Data andConsulting Services in Doha, Qatar, to work on FMI*Fullbore Formation MicroImager tool interpretation.He holds an MS degree in geology from PhilippsUniversity in Marburg, Germany.

Mark Knackstedt earned a BS degree from ColumbiaUniversity, New York City, and a PhD degree from RiceUniversity, Houston, both in chemical engineering. Heis Professor and Head of the Department of AppliedMathematics at The Australian National University,Canberra, and a Visiting Professor at the School ofPetroleum Engineering at the University of New South Wales in Sydney. His work has focused on thecharacterization and realistic modeling of disorderedmaterials. Mark’s primary interests lie in modelingtransport, elastic and multiphase flow properties anddevelopment of 3D tomographic image analysis forcomplex materials.

M.A. Mercado Ortiz is Oriental Area Manager for the Burgos Integral Asset for Pemex Exploración yProducción. He was previously manager for theCuitlahuac group.

Tom Plona, Scientific Advisor at SDR in Ridgefield,Connecticut, is currently working on new sonic-loggingmethods to enhance understanding of geomechanicsproblems. Since joining Schlumberger in 1976, he hasconducted basic rock-physics acoustic studies and participated in numerous acoustic-tool developmentprojects. Tom is serving as an SPWLA DistinguishedSpeaker for the year 2006 and is a prolific author whoholds numerous patents. He has a BS degree in physicsfrom Providence College, Rhode Island, USA, and MSand PhD degrees in physics from GeorgetownUniversity, Washington, DC.

Lasse Renlie, based in Stjørdal, Norway, is LeadPetrophysicist in the Halten/Nordland area of theNorwegian Sea for Statoil ASA. Before joining the company in 1998, he worked for Sintef PetroleumResearch in Trondheim. Lasse obtained an MS degreein physics and a PhD degree in borehole acousticsfrom The Norwegian University of Science andTechnology, Trondheim.

John Simonetti is a Technical Consultant forSchlumberger PTC in New Jersey, where he works onthe application of new materials for the improvementof nuclear tools. He joined the company in 1982 in theCeramic Product Development group and later workedin the materials technology section and on imagingdetectors, nuclear generators, new scintillator evalua-tion and neutron monitor development. John earned aBS degree from Rutgers, The State University of NewJersey, New Brunswick, and MS and PhD degrees fromPrinceton University, all in chemistry.

Bikash Sinha is a Scientific Advisor in the Mathematicsand Modeling program at SDR in Ridgefield,Connecticut. Since joining Schlumberger in 1979, hehas contributed to many sonic-logging innovations forgeophysical and geomechanical applications and tothe development of high-precision quartz pressuresensors. He is currently involved in the near-wellborecharacterization of mechanical damage and estima-tion of formation-stress parameters using boreholesonic data. Bikash received a B Tech degree (Hons)from the Indian Institute of Technology, Kharagpur,and an MS degree from the University of Toronto,Canada, both in mechanical engineering. He has aPhD degree in applied mechanics from RensselaerPolytechnic Institute, Troy, New York, USA. He hasauthored or coauthored more than 135 technicalpapers and received 22 US patents. An IEEE Fellow,Bikash received the 1993 outstanding paper award forinnovative design and development of quartz pressuresensors published in the IEEE Transactions onUltrasonics, Ferroelectrics, and Frequency Control.

Jacob I. Trombka is a Senior Goddard Fellow at theNASA Goddard Space Flight Center, Greenbelt,Maryland, USA. He is Team Leader for the X-ray/gammaray remote-sensing spectrometers for the Near EarthAsteroid Rendezvous (NEAR) spacecraft. At Goddard,he has worked on the development of remote-sensingspectrometers and has been principal investigator,team member or guest investigator on the US Apollo,Viking, WIND, SMM and Mars Observer programs and the Russian Luna, Mars, Phobos and Mars 1996programs. Jacob began his career at the NASA JetPropulsion Laboratory, working on the Ranger gammaray spectrometer and studying the applications of X-ray, gamma ray and neutron/gamma ray spectroscopyto planetary remote and in-situ geochemical analysissystems. The asteroid 1981 ET26 has been renamed(4990) Trombka for his work in that field. He is a prolific author and has received many awards. Heearned BS and MS degrees in physics from WayneState University, Detroit, Michigan, USA, and a PhDdegree in nuclear science from the University ofMichigan, Ann Arbor.

Contributors

An asterisk (*) is used to denote a mark of Schlumberger.

58171schD08R1.qxp 5/25/06 10:51 PM Page 63

64 Oilfield Review

Stefan Vajda is Principal Research Scientist and nowworks in nuclear detectors at the Schlumberger PTCin New Jersey. He holds an MS degree from theUniversity of Bucharest, Romania, and a PhD degreefrom the Institute for Nuclear Physics andEngineering, Bucharest/Magurele, Romania, both inphysics. Stefan joined the company in 1984, working asa research physicist in neutron and X-ray generators.He later worked on inorganic scintillator research,nuclear detectors and gamma ray spectroscopy. Hebuilt a gamma ray spectrometer for the NASA missionto asteroid 433 Eros, which was successfully completedin 2001. Stefan has written more than 40 papers andholds two patents.

Henri-Pierre Valero, Program Manager and SeniorResearch Scientist at SDR, Ridgefield, Connecticut, is involved in the development of algorithms for bore-hole acoustic tools. He joined the company in 1998 as a project engineer at Schlumberger KK, Fuchinobe,Sagamihara, Kanagawa, Japan, and worked on thedevelopment and implementation of sonic answerproducts until his transfer to SDR in 2004. Henri-Pierreholds a BS degree in science from Rennes University,France, an MS degree (Hons) in geoscience froml'Ecole Nationale Supérieure des Pétroles et Moteurs,Rueil-Malmaison, France, and a PhD degree (Hons) in geophysics from Institut de Physique du Globe deParis, France.

Stephen Williams is Formation Evaluation Adviser forNorsk Hydro ASA in Bergen, Norway. He is responsiblefor planning, execution and follow-up of formationevaluation programs on the company’s wells aroundthe world. He has held a number of formation evalua-tion positions since he joined Norsk Hydro in 1998.Before this, he spent 14 years with Schlumberger invarious assignments in operations, technical manage-ment, training and management in North and SouthAmerica, Europe, Scandinavia and the Middle East.Stephen earned BA and MA degrees in natural sci-ences from University of Cambridge in England.

Kenneth Winkler is a Principal Research Scientist atSDR, Ridgefield, Connecticut, working on microsonicprinciples, tools and interpretation techniques to pro-duce a high-resolution acoustic-velocity map of theborehole wall. His other projects include nonlinearacoustics, flow assurance and pore-pressure studies.He joined Schlumberger in 1979, setting up the SDRhigh-pressure rock properties laboratory. Since then, he has worked on various programs includinggeoacoustics, near-wellbore acoustics and high-resolu-tion microsonic imaging. Ken holds a BS degree inphysics from Rensselaer Polytechnic Institute, Troy,New York, and MS and PhD degrees in geophysics from Stanford University, California. He has served as a Distinguished Speaker of SPWLA, and he wastechnical editor of SPE Formation Evaluation from1993 to 1996. The author of many papers, he also holds several patents.

Murtaza Ziauddin, Schlumberger Principal Engineerat the Sugar Land Technology Center, works on matrixstimulation of sandstones and carbonates, CO2 seques-tration, hydraulic fracturing, and organic and inor-ganic scale. He led the development of Virtual Lab*geochemical simulation software for matrix acidizing,inorganic scale prediction and water-compatibilitytesting. He is involved in developing a predictive rheol-ogy model for polymer-base fracturing fluids and withacidizing models in StimCADE* well stimulation soft-ware and WellBook* software application for treat-ment design, execution and evaluation. Murtaza joinedSchlumberger in 1997 after receiving a BS degree fromthe University of Houston, and a PhD degree from theUniversity of Minnesota at Minneapolis, USA, both inchemical engineering. He has written many papersand holds several patents.

Wolfgang Ziegler, Principal Engineer forSchlumberger PTC, New Jersey, is currently workingon nuclear detector development, focusing on high-temperature applications and new materials. Hejoined the company at SDR, Ridgefield, Connecticut, in1992 to work on measurement concepts later imple-mented in EcoScope* multifunction logging-while-drilling tools and researching new detectors foroilfield applications. He transferred to PTC in 2000and continued his involvement in those projects.Wolfgang earned a Diploma in physics from MainzUniversity, and a PhD degree in experimental nuclearphysics from Darmstadt University, both in Germany.

Coming in Oilfield Review

Naturally Fractured Reservoirs. The presence ofnatural fractures in reservoirs can make otherwisetight rock productive, but fractures can also negativelyimpact porous and permeable reservoirs. This articleexamines naturally fractured reservoirs, and howindustry geoscientists and engineers detect, characterize and model them at various scales andthroughout all stages of reservoir development.

Subsidence and Compaction. Subsidence abovereservoirs may have enormous economic consequences,which may not be limited to damage to oilfield infra-structure. A large subsidence bowl may cause extensivedamage to surface structures, particularly in low-lyingareas. Compaction results from depleting formationsthat are mechanically weak; it is the cause of industry-related subsidence. The article describes the funda-mentals of compaction and subsidence, and includescase studies from several active fields.

Heavy Oil. Dwindling production of conventional oils,high prices and the need to augment reserves are revitalizing interest in heavy oil. This article reviewsfluid properties of heavy oil and potential productionscenarios, from mining to in-situ combustion. Casestudies demonstrate techniques for characterizingheavy-oil reservoirs, determining the best recoverymethod, constructing and completing wells, and monitoring production.

58171schD08R1.qxp 5/25/06 10:51 PM Page 64

06_OR_002_0

58171schD01R1.qxp 5/31/06 8:34 PM Page 2

SCHLUMBERGER OILFIELD REVIEW

SPRING 2006

VOLUME 18 N

UMBER 1

Spring 2006

Oilfield Review

Oilfield Technologies in Space

Sonic Advances

Acoustic Waves

High-Resolution Core Visualization

58171schD01R1.qxp 5/31/06 8:34 PM Page 1