Evaluating Carbonates

38

lution in our ability to evalu-

munity of interpreters

g carbonate reservoirs to

Classic Interpretation Problems: Evaluating Carbonates

Mahmood AkbarMario PetricolaMohamed WatfaAbu Dhabi, United Arab Emirates

Mohammed BadriMouhab ChararaCairo, Egypt

Austin BoydDenver, Colorado, USA

Bruce CassellRoy NurmiDubai, United Arab Emirates

Jean-Pierre DelhommeClamart, France

Mike GraceDallas, Texas, USA

Bill KenyonRidgefield, Connecticut, USA

Jon RoestenburgJakarta, Indonesia

ian Singer, Schlumberger Wireline & Testing,i, India; and Eric Standen, Schlumberger Wire-ing, Montrouge, France.le, ARI (Azimuthal Resistivity Imager), CMRle Magnetic Resonance tool), DSI (Dipolec Imager), EPT (Electromagnetic Propagation(Fullbore Formation MicroImager), FracView,, MDT (Modular Formation Dynamics Tester)hermal Decay Time) are marks of Schlum-

r behavior. This article tracks theation process, from carbonate rockon and petrophysical log evaluationechniques for measuring permeabil-hole and mapping large-scale flow and barriers.

ates for Beginnerstes and sand-shale rocks, or silici-are worlds apart. Whereas siliciclas- are composed of a variety of silica-grains that may have traveleds of miles from their source, carbon-s mainly consist of just two miner-ite and dolomite—and remain nearint of origin.1 Carbonates form inand deep marine settings, evaporiticakes and windy deserts.2 Most of thetes formed in past have shallow

INTERPRETATION

In recent years, there has been a small revo

ate carbonates. Novel technology and a com

determined to crack one hard nut are forcin

reveal many of their secrets.

For help in preparation of this article, thanks to HervéAnxionnaz, Schlumberger Wireline & Testing, Clamart,France; Jean-Louis Chardac, Schlumberger Wireline &Testing, Dubai, UAE; Bob Dennis, Schlumberger Wire-line & Testing, Wattayh, Oman; Philip Cheung, ArnaudEtchecopar and Ollivier Faivre, Schlumberger Wireline &Testing, Clamart, France; Brian Hornby, SchlumbergerCambridge Research, Cambridge, England; Rachel Korn-berg, GeoQuest, Gatwick, England; William Murphy andPabitra Sen, Schlumberger-Doll Research, Ridgefield,Connecticut, USA; Oberto Serra, Consultant, Paris,

France; JulNew Delhline & TestIn this artic(CombinabShear SoniTool), FMI GeoFrameand TDT (Tberger.

Carbonate reservoirs account for 40% oftoday’s hydrocarbon production, andbecause of several elephant fields in theMiddle East they are expected to dominateproduction through the next century. There-fore, understanding carbonate reservoirsand producing them efficiently havebecome industry priorities and are likely toremain so.

Current efforts in carbonate exploitationfocus on correctly targeting new wells, fre-quently horizontal, to optimize productionfrom untouched reserves and on ensuringthat massive water injection schemes deliveran effective sweep of the reservoir. In sup-port of these efforts, geoscientists are tryingto decipher the enigma of carbonate rock’scomplex pore space and understand howpermeability barriers and conduits affect

reservoiinterpretdescriptito new tity downconduits

CarbonCarbonaclastics, tic rocksbased hundredate rockals—calctheir poshallow basins, lcarbona

Oilfield Review

39January 1995

nDistribution of carbonates. Areas of modern sediments in shallow water are shown in yellow and orange, modern deepwater sedi-ments in black-hatched lines. Prolific carbonate oil basins are shown in red-hatched lines. [Adapted from Wilson, reference 2; Blatt H,Middleton G and Murray R: Origin of Sedimentary Rocks. Englewood Cliffs, New Jersey: Prentice-Hall, Inc. (1972): 410; and Perrodon A:Dynamics of Oil and Gas Accumulations. Pau, France: Bulletin des Centres de Recherches Exploration-Production Elf-Aquitaine (1983): 215.]

nUnfilled interparticle porosity (black).Holocene oolite, Great Salt Lake, Utah,USA. Cross-polarized light photograph.

1. An exception is the less-abundant class of rocks called calcareous sandstones, or carbonate arenites,which form when carbonate rock is broken up bywind or water, then transported and deposited. Cal-careous sandstones exhibit many of the structural and petrophysical characteristics of siliciclastic sandstones, while retaining carbonate mineralogy and microporosity.

2. For a review of carbonate geology:Wilson JL: Carbonate Facies in Geologic History.Berlin, Germany: Springer-Verlag, 1975.Reeckmann A and Friedman GM: Exploration for Carbonate Petroleum Reservoirs. New York, New York, USA: John Wiley & Sons, 1982.Tucker ME and Wright VP: Carbonate Sedimentology.Oxford, England: Blackwell Scientific Publications,1990.

3. Scholle PA: A Color Illustrated Guide to CarbonateRock Constitutents, Textures, Cements and Porosities.Tulsa, Oklahoma, USA: American Association ofPetroleum Geologists, 1978.

marine origins, but the most widespreadtype of modern carbonate is formed in deepwater (above). Silica-based rocks generallystand up to the rigors of geologic time,undergoing only minor alteration, or diagen-esis. Their depositional record is preserved,with bedding planes on outcrops and sub-surface correlations between wells clearlyrecognizable. The grains are regularlyshaped, and the pore space, though compli-cated, remains intergranular.

Carbonate rocks, on the other hand, arechemically unstable and undergo substan-tial alteration such as mineral dissolutionand dolomitization—the replacement of cal-cium carbonate by magnesium carbonate.Carbonates house a jumble of complex parti-cles, including a huge variety of biologicalorigin, and an even more complex porespace (Photographs, reprinted with permis-sion from Scholle, reference 3, are posi-tioned throughout this article).3 This compli-cates the tracking of facies across a carbonatereservoir and the assessment of the productiv-ity of a given carbonate formation.

The typical carbonate rock is made ofgrains, matrix and cement. Grains are eitherskeletal fragments of small organisms or par-ticles precipitated from calcium-rich water.The latter includes a variety of small, accre-tionary grains identified according to theirsize, origin and internal structure.

Matrix is the lithified mud of depositionthat fills most of the space not occupied bygrains. In carbonates, fine mud has severalsources—chemical precipitation, breakingof skeletal material into finer material,

40°

60°

20°

0°

20°

40°

60°

Reef Shelf carbonate Deep carbonate Carbonate oil province

0.23 mm

replacement of limestone by dolomite; sty-lolitization—the formation of stylolites,irregular planes of discontinuity betweenrock units due to compaction-related pres-sure solution; and fracturing—the planarbreaking up of rock mass due to stress.

Time and diagenesis generally workagainst the preservation of porosity (nextpage). Young carbonates usually haveporosities around 60%. Old carbonateshave just a percent or two. Reservoir car-bonates survive with porosities of 5 to 15%largely because the presence of hydrocar-bon impedes further destruction of porosity.The typically prolonged and extensive dia-genesis in carbonates also usually obscuresthe provenance and history of the rock.

Reservoir DescriptionHow does the reservoir geologist use thesedescriptions to help plan the optimumexploitation of a carbonate reservoir? Identi-fying and classifying carbonates are crucialin two key tasks. First, assessment of thereservoir’s paleoenvironment builds a broadunderstanding of likely reservoir geometry.Then, detailed well-to-well correlation oflithofacies helps construct a detailed three-dimensional picture.

Clues to paleoenvironment come fromevery available source—seismic surveys,outcrop studies, cuttings and core analysis,and logs, including those from the latestgeneration of electrical imaging tools, whichcan capture the wellbore likeness to a reso-lution of about 5 mm [0.19 in.] The mainpaleoenvironment indicators are:•Lithology—This provides a general idea of

the depositional setting. The presence ofclastic rocks indicates an external sourceof sediment, while their absence indicatesan environment free of external influence.

•Rock texture—The Dunham classificationof texture provides some idea of theenergy of deposition. Grain size variationsalso point to the sequence of deposition.For example, a fining upward sequencemay indicate a relative sea level rise, ormarine transgression. A coarseningupward sequence probably indicates arelative sea level drop.

•Sedimentary structure—Large-scale sedi-mentary structures are more difficult tosee in carbonates than in siliciclastics, butwhen identified, they offer powerful cluesto the depositional environment. Exam-ples are crossbeds in eolian dunes or solu-

remains of algae, and others. On lithifica-tion, mud becomes a very fine-grained cal-cite called micrite.

Cement describes crystalline material thatforms in most of the space remainingbetween grains and matrix or betweengrains themselves, binding them. Cementmay have a variety of crystal sizes depend-ing on its composition, the conditions ofcrystallization and the spaces to be filled.

Crucial to the interpretation of carbonatesis classifying the numerous ways grains andmatrix coexist. Progress in categorizing thesecomplexities surged in the late 1950sbecause of pressure within oil companies tobetter understand their carbonate assets. Theclassification that has stood the test of timemost successfully is by Robert Dunham.4

Dunham classifies a spectrum of rocktypes based on the internal structure and tex-ture of the rock (below). Mudstone consistsmainly of matrix in which relatively fewgrains are suspended. Wackestone is alsomatrix-supported but has more grains. Pack-stone has enough grains for them to startproviding support—matrix fills the remainingnonpore space. Grainstone has plenty ofgrains providing support and includes pro-gressively less matrix. Finally, boundstonedescribes carbonate rocks in which the origi-nal material provided support during deposi-tion, such as in reefs. Crystalline describesrock that has lost its depositional fabricbecause of diagenetic recrystallization, forexample, dolomitization.

Dunham’s classification provides someclue to the energy of deposition. The mud-based mudstone and wackestone are

40 Oilfield Review

deposited in low-energy settings. Packstoneand grainstone would appear to be fromhigh-energy deposition, but given significantdiagenesis, these grain-supported rockscould equally well have been deposited asmud-supported agglomerations and thenthrough compaction and chemical alter-ation transformed to their present state. Thedifficulty in classifying carbonates to reflectboth their current state and depositional his-tory demonstrates how dominant diagenesisis in forming the final carbonate rock.

Diagenesis may be divided into five mainmechanisms: compaction—the reduction ofpore space in response to tighter grain pack-ing as overburden increases; carbonatedegradation—the destruction of carbonatematerial through chemical dissolution andmicritization, the transformation of largecrystals into small ones; carbonate aggrada-tion—the construction of carbonate materialthrough precipitation of cement betweengrains, and recrystallization, such as the

Mudstone Wackestone Packstone Grainstone Boundstone Crystalline

Less than10% grains

More than10% grains

Mud-supported

Contains mud,clay and fine silt-size carbonate

Original components not bound together duringdeposition

Depositional texture recognizable

Grain-supported

Lacks mudand is grain-supported

Originalcomponentswere boundtogether

Depositionaltexture notrecognizable

nDunham’s classi-fication of carbon-ates, based on theinternal structureof the rock.

nCalcite-filled interparticle porosity(black) in oolite. Upper MississippianPitkin Limestone, Oklahoma, USA. Cross-polarized light photograph.

0.07 mm

tion grooves caused by irregular dissolu-tion of the surface of a carbonate rock.

•Biofacies—Identification of the wide vari-ety of skeletal grains, burrows and moldsmay pinpoint precise geological time andsettings. Ages and habitats of hundreds ofcarbonate creatures have been tabulatedfor this purpose.

•Nonskeletal content—Grains formed byprecipitation and accretion provide apowerful indicator of depositional setting.For example, homogeneous pelletsdevelop in quiet lagoons, while concentri-cally layered ooids occur mainly in active,shallow environments.

•Authigenic minerals—The cement andminerals that form in the rock after depo-sition provide some additional clues. Thepresence of pyrite suggests reducing(deoxidizing) conditions; glauconite indi-cates marine conditions; organic matterindicates little reworking.Slowly evidence accumulates, and the

origin and evolution of the reservoirbecome less conjectured and more certain.In comparison, the mapping of lithofacies isa detailed, nuts-and-bolts task, but no lesschallenging. Unlike siliciclastics, carbon-ates usually carry no bedding signaturesthat allow ready well-to-well correlationacross a field. Even dipmeter correlationsacross the borehole can be elusive. Thereare exceptions. Eolian carbonate deposits,for example, display continuity and bed-ding signatures exactly like their siliciclasticequivalents. More frequently, though, car-bonates exhibit a mix of features such asfractures, breccia, stylolites and vugs.

Helping the reservoir geologist recognizeand catalog these features in wells drilled

41January 1994

nCreation and destruction of carbonate porosity due to compaction and diagenesis,as a function of age. Inset—Transformation of a shell to create various molds andcasts, as it undergoes different combinations of burial, filling and dissolution.(Adapted from Reeckmann and Friedman, reference 2.)

4. Dunham RJ: “Classification of Carbonate RocksAccording to Depositional Texture,” in Ham WE (ed):Classification of Carbonate Rocks. Tulsa, Oklahoma,USA: American Association of Petroleum Geologists,1962.

DissolutionFissuresVugsCaverns

FractureBrecciaJoints

4. Erosional Porosity

Late

buria

l

diage

nesis

Calcite spar

Infillings

3. Pressure-and Temperature- Related Porosity

Pressuresolution

Compaction

Tectonic activityFracture

Calcite

Marine watersAragoniteMagnesium-calcite

Fresh water Cemen

t

Cement

RecrystallizationIntercrystalline

DissolutionVugsChannels

Burrowingorganisms

1. InitialPorosity

2. EarlyDiageneticPorosity

Low energyFenestralIntramicrite

High energyFrameworkIntraparticleInterparticle

Formation of pores

Depositional environmentSynsedimentary cement

Internal sediment

Boring organisms

Micrite

Lime mudMicrodebrisPeloids

Destruction of poresG

eolo

gic

time

Early diagenesis

50250 75 100Porosity, %

Original shell

Buried unfilled Buried filled Filled beforeburial

Dissolve shell material

Fill cavity Bury insedimentFill cavity

Recrystallization

Ove

rbur

den

&te

cton

ics

nSubhorizontal stylolites (wide darkbands) and inclinedfractures (narrowdark lines) in a Mid-dle East carbonateformation.

nMottled fabric with thin producing zone at XX88 ft. Dark color isinterpreted as mud-filled porosity. Light color is grains and matrix.

42 Oilfield Review

nSwarms of stylolites in a Mississippian carbonate, with twocore points visible.

XX93

XX92

XX94

Dep

th, f

t

XX88

XX87

XX89

Dep

th, f

t

Dep

th, f

t

XX15

XX17

with water-base mud is the FMI FullboreFormation MicroImager tool, which pro-vides a picture of most of the borehole with192 small current-emitting buttons mountedon four pads and four flaps. In the images,light color denotes high resistivity, indicatingrock grains or hydrocarbon-filled pores, anddark color indicates low resistivity such aswater-filled pores or shale. The images areno substitute for core analysis, but rather acomplement to them. Other evidence is fre-quently needed to corroborate an interpreta-tion, for example to decide whether a darkpatch is porosity or shale. However, anexperienced interpreter of FMI images canglean strong evidence of numerous types ofcarbonate features down to the centimeterscale (this page and next page).5

A recent trend in FMI interpretation hasbeen toward quantitative analysis of theimages. One processing method automati-cally extracts five facies types based on atextural classification by Nurmi et al.6 Thefacies are uniform zones of constant conduc-tivity or resistivity; layered zones of alternat-ing conductive and resistive layers; zones

5. Serra O: Formation MicroScanner Image Interpreta-tion. Houston, Texas, USA: Schlumberger EducationalServices, 1989.

6. Nurmi R, Charara M, Waterhouse M and Park R:“Heterogeneities in Carbonate Reservoirs: Detectionand Analysis Using Borehole Electrical Imagery,” inHurst A, Lovell MA and Morton AC (eds): GeologicalApplications of Wireline Logs. London, England: Geological Society of London, 1990.

nSecondary micrite formation in the form of large,light-colored areas.

XX27

XX26

XX28

Dep

th, f

t

43January 1995

nUnconformity between overlying shale and mid-Cretaceouscarbonate. Large, irregular dark features below the unconfor-mity are voids created by extensive dissolution during subaerialexposure.

nBreccia caused by collapse of a limestone cavern, indicatedby interconnected channels between sharp fragments of rock.This type of breccia porosity often makes for prolific production.

nLarge calcite concretion, that may helpelucidate depositional environment.

nMottled fabric of an Upper Cretaceouscarbonate reservoir in the Middle East.

Dep

th, m

96.2

96.4

96.6

Dep

th, f

t

XX84

XX85

XX86

XX60

XX59

XX61

Dep

th, f

t

19.0

18.2

18.6

Dep

th, f

t

Dep

th, f

t

XX05

XX10

Layered

Resistive with isolatedconductiveareas

Resistive with isolatedconductiveareas

Conductive with isolatedresistive areas

Cond. Area C

Cond. Area B

THLY

REPC

C O P Cmmho

mmho

0

0

0

0.1

0 100

250

250

ft

%A B C

with interwoven or contiguous conductiveareas, interpreted as interconnected porosity;resistive zones with isolated conductiveareas, interpreted as nonconnected pores;and conductive zones with isolated resistiveareas, caused for example by nonconductingcalcite or anhydrite nodules. Such a zonationcan be rapidly calculated from the imagesand lends itself readily to facies mappingacross a field (below).

A more advanced processing method actu-ally delineates identifiable objects such asrock grains or pores on the images.7 This isquite a challenge because picking the edge ofan object depends somewhat on overallimage intensity, which varies. The solution isto equate object boundaries with inflectionpoints in image intensity. This approach isincorporated in SPOT—Secondary PorosityTyping—prototype software running on Geo-

Quest’s GeoFrame platform. Current SPOTprocessing can yield the boundaries of bothresistive (light color) and conductive (darkcolor) features (next page, top). In tests madeon laboratory rock samples bored with“pores” of varying but precisely known diam-eter, the processing has given accurate andconsistent pore delineation.

Once resistive and conductive features aredelineated, then all manner of quantitativeinformation can be computed, such as theiraverage size, the spatial density of the fea-tures, the total area on the image covered bythe features, and the degree to which likefeatures are connected. We will later

address how these new parameters maycontribute to understanding of rock porosityand permeability.

The average sizes of resistive and conduc-tive features have recently been used to helpidentify Dunham rock types in an Occiden-tal Oil Company field in Indonesia and thuscontribute to facies mapping (page 46, bot-tom).8 In this interpretation, resistive featurescorrespond to carbonate coral framework orgrains, while conductive features corre-spond to pores or micritic matrix. On a log,the average sizes of the two types of featuresare played back together. The interpretationproceeds by noting the separation betweenthe curves and also their absolute magni-tudes.

Mudstone is interpreted when separationis at a maximum. This occurs when theaverage size of conductive features peaks—that is, micritic cement dominates—andthe average size of resistive features—orgrains—drops. Wackestone is interpretedwhen the average size of resistive featuresincreases, while the average size of conduc-tive features remains about the same. Pack-stone is interpreted when the average size ofresistive features peaks. And finally, grain-stone is interpreted when the sizes ofconductive and resistive features becomeequal. This broad-brush methodology hasbeen verified against macrofacies descrip-tions from cores in two wells in the field.Furthermore, the frequency with which thetwo curves mirror one another appears toindicate the frequency of a complete depo-sitional cycle—from low-energy mudstoneto high-energy grainstone.

44 Oilfield Review

nImplementation of Nurmi’s porosity classification through automatic processing. TheFMI image is processed to yield resistive and conductive segments (B). Segments thatare not continuous across the image are eliminated (C). Conductive area within eachprocessed image is plotted as a log (blue for B, red for C). Average conductivity for resis-tive segments in B (green) and for conductive segments in B (orange) are shown with thevariation in thickness of these segments across the image width (black).

7. Delhomme JP: “A Quantitative Characterization ofFormation Heterogeneities Based on Borehole ImageAnalysis,”Transactions of the SPWLA 33rd AnnualLogging Symposium, Oklahoma City, Oklahoma,USA, June 14-17, 1992, paper T.

8. Roestenburg JW: “Carbonate Characterization andClassification from In-Situ Wellbore Images,” pre-sented at the 23rd Annual Convention of the Indone-sian Petroleum Association, Jakarta, Indonesia, Octo-ber 4-6, 1994.

nUnfilled intraparticle porosity (black)within a large coral fragment. Holoceneback-reef beach sediment, Belize,British Honduras. Cross-polarized lightphotograph.

0.26 mm

nCross-polarized light photographs showing porosity. Left: Reduced interparticle and intraparticle porosity (black) in foraminiferaand mollusks. Pleistocene Key Largo Limestone, Florida, USA. Middle: Intercrystal porosity (black) in a fine- to medium-crystallinereplacement dolomite. Middle Eocene Avon Park Limestone, Florida. Right: Moldic porosity (black). Pleistocene Miami oolite, Florida.

45January 1995

nSPOT processing on FMI images delineates either resistive or conductive inclusions, orspots. The inclusions can then be analyzed to provide quantitative parameters such asmedian size of inclusion, density of inclusions per foot or meter, average area percent-age of inclusions, and even a porosity estimate and connectivity parameter.

0.24 mm 0.24 mm 0.26 mm

XX62

0 30 60 90 120 150 180 210 240 270 300 330 360

XX63

XX64

XX65

XX66

Number ofresistive

spots per ft

Percentage ofimage covered

by resistivespots,%Orientation

0 050 50

Dep

th, f

t

Without images, mapping facies followinggamma ray and other log signatures canoften prove unreliable. The safest bet, shortof coring every borehole, is to collectivelyinterpret all available log data, initially cali-brating the interpretation results to coredata. An example of this approach can befound in a study by the Indian Oil and Natu-ral Gas Commission (ONGC) and Schlum-berger that recently addressed a complexMiddle Eocene carbonate formation in off-shore India.9

In this study, the first step was to identifyfacies in the four cored wells according toDunham’s classification. This required theanalysis of 120 thin sections, 12 polishedsections and 6 scanning electron micro-scope images. This petrologic interpretationwas then integrated statistically with five logmeasurements made in the same wells—density, neutron porosity, sonic travel time,gamma ray and saturation. Matching the logmeasurements to the facies descriptionsrevealed clear links between weighted com-binations of log data and the Dunham classi-fication (above, left). However, rather thanDunham’s four, the logs recognized fivefacies types, the last of which alwaysoccurred within a wackestone zone but atdepths where no core was retrieved. Thisfacies was termed wackestone+. With logscalibrated to a core facies description, afacies interpretation could be made directlyfrom logs in all the remaining uncored wells,and then facies mapped between wells.

Petrophysical EvaluationFacies determination from logs is hardenough, but the challenge of establishingpetrophysical parameters such as saturationand permeability is even more daunting.The reason lies squarely with the complexdiagenesis and resulting convoluted poresystems of most carbonate rocks. Log ana-lysts divide porosity into primary and sec-ondary components, with primary existingat the time of rock formation and secondaryappearing as the rock matures and diagene-sis prevails. The more detailed classificationof Choquette and Pray exposes the immensediversity in both shape and size of carbon-ate pores (next page, top).10

The variety in pore type explains why per-meability answers remain so elusive. Vugsand their cousins may make for high poros-ity, but a consistent pore connectivity, usu-ally taken for granted in sandstones, may ormay not be present. Worse yet, the chaos

46 Oilfield Review

nFacies interpretation according to Dunham’s classification using statistical calibrationbetween log data and petrologic analysis from cores in an offshore carbonate field,India. The calibration allows facies to be mapped across the field using other loggedwells without cores.

nFacies interpretation in an Occidental Oil Company carbonate field in Indonesiausing SPOT processing to automate Dunham’s classification from FMI images. Dunhamrock type is interpreted by comparing the relative magnitudes of the average size ofresistive and conductive inclusions. Interpretation has been verified from cores.

XX70

30 mm 2 30000 30 mm 2 30000

XX72

XX74

ConductiveAverage Size

ConductiveAverage Size

Resistive Average size

Resistive Average size

FMI ImageFMI Image

GrainstoneGrainstone

Mudstone

W ackestone

Well 1 Well 2

Dep

th, f

t

0 40 -2 2 0 100 – + 1.0 100

Dep

th, f

t

XX50

XX60

XX70

XX80

�

GammaRay S

ynth

etic

Sei

smic

Trac

e

Per

cent

Rec

over

ed

Dia

gene

sis

Cor

eD

escr

iptio

n

LogFacies

Packstone

Wackestone

Grainstone

145BOPD

1660BOPD

Sonic-RtOverlay

Ohm-m

reigns at all scales. In sandstones, small 1/2-in.[1.27-cm] plugs bored from cores usuallyprovide samples homogeneous enough forestimating average permeability. In carbon-ates, however, sometimes not even a wholepiece of core can be regarded as representa-tive. The discrepancy between permeabilitiesmeasured at different scales may be relatedto heterogeneity or to anisotropy.11 The onlysure way of estimating reservoir-scale per-meability is by using wireline, drillstem orproduction tests. This was the approachtaken in the second phase of the Indianstudy, in which nine well tests in two wellsestablished a link between carbonate faciestype and permeability.

Each carbonate facies type was allowed apermeability value, to be determined. Then,for each test, the well’s flow capacity calcu-lated during the test was matched with thesum of the individual flow capacities of thewell’s various facies types. Each facies’ flowcapacity was the product of the facies type’sunknown permeability and its cumulativethickness in the well. The result of the matchwas a range of permeabilities for each faciestype, two types—grainstone and wacke-stone+—being particularly permeable. Pro-duction logs in one well confirmed the pro-ductivity of wackestone+ (right).

A much earlier study, predating imagingtechnology, also recognized clear differ-ences in permeabilities of the rock types ofthe Dunham classification. This study first

10. Choquette PW and Pray LC: “Geologic Nomenclatureand Classification of Porosity in Sedimentary Carbon-ates,” AAPG Bulletin 54 (February 1970): 207-250.

11. Ayan C, Colley N, Cowan G, Ezekwe E, Goode P,Halford F, Joseph J, Mongini A, Obondoko G, Pop Jand Wannell M: “Measuring Permeability Anisotropy:The Latest Approach,” Oilfield Review 6, no. 4 (Octo-ber 1994): 24-35.

9. Vashist N, Dennis RN, Rajvanshi AK, Taneja HR,Walia RK and Sharma PK: “Reservoir Facies and TheirDistribution in a Heterogeneous Carbonate Reservoir:An Integrated Approach,” paper SPE 26498, presentedat the 68th SPE Annual Technical Conference andExhibition, Houston, Texas, USA, October 3-6, 1993.

47January 1995

nChoquette and Pray’s classification ofcarbonate porosity: Fabric-selective poros-ity includes interparticle porosity occurringbetween grains; intraparticle porosityoccurring within original skeletal grains;intercrystal porosity occurring within crystallized micrite and/or dolomite;moldic porosity that results from the dissolution of grains; large-scale frame-work porosity called fenestral porosity that usually results from dissolution ofalgal mat deposits; shelter porosity thatdescribes pore space resulting from shelteroffered by large overlaying grains; andgrowth-framework porosity that is the natural outcome of organic processes such as coral reefs.

Nonfabric-selective porosity includesfracture porosity; channel porosity causedby extensive leaching; vug porosity thatresults from extensive dissolution of mate-rial and retains no evidence of the originalhost grain; and man-size cavern porositythat results from highly extensive and pro-longed leaching. Porosity that could fall ineither camp includes breccia, boring, bur-rowing and shrinkage porosity.

nMatching welltests with Dunham-style evaluation inan Indian offshorecarbonate field toprovide permeabil-ity values for mainfacies types. Cumu-lative flow profile(left) and flow con-tribution from eachlayer (middle) con-firm high perme-ability of wacke-stone+ (right).

Wackestone +

Wackestone

Cumulative Flow FaciesEntry

Packstone

Grainstone

Dep

th, f

t

XX60

100 % %100 00

XX70

Fabric-selective Not fabric-selective Fabric-selective or not

Interparticle

Intraparticle

Intercrystal

Moldic

Fenestral

Shelter

Growth-framework

Fracture

Channel

Vug

Cavern*

Breccia

Shrinkage

Burrow

Boring

*Cavern applies to man-sized or larger pores of channel or vug shapes

used a rudimentary log interpretationmethod to distinguish one rock type fromanother (below, left).12 Once the type wasidentified, a relevant porosity-permeabilityrelationship was applied at each depth tocalculate permeability from porosity logs.The procedure resulted in far better agree-ment with core permeability measurementsthan had previously been obtained.

In general, there are two ways to establishelusive petrophysical parameters such aspermeability from log data. One is to link theparameter statistically to log data, calibratingthe link with measurements of the parametermade in the field or laboratory. The calibra-tion can be in just one well or an entire field.An example is the Indian offshore studywhere well test results were linked to a statis-tically derived facies interpretation. The vari-ety of such statistical methods is immense

and currently extends to the use of neuralnetworks that attempt to mimic and evenimprove on our inherent ability to recognizepatterns in diverse data.13

The other approach is to somehowdirectly measure something about the rock’spore space, ideally from logs, and then tiethis in with sought-after petrophysicalparameters such as saturation and perme-ability. To this end, the newest measurecomes from FMI images, again thanks toSPOT processing. The proportion of animage delineated as pore space leadsdirectly to a new estimate of porosity, sub-ject of course to the interpretation that darkareas of the image are indeed pores.

In a well drilled through a carbonate reser-voir in Italy, SPOT-derived porosity compareswell with porosity conventionally interpretedfrom neutron and density logs (below, right).

In much of the logged interval, the twoporosities agree well, while elsewhere poros-ity derived from the FMI images is substan-tially less than conventional porosity. Thiscould be due to the FMI tool respondingonly to pores larger than the 5-mm resolutionof the tool and missing smaller intergranularand micritic pores. Interestingly, zones wherethe two porosities differ coincide with zonesflagged with a secondary porosity index bythe SPOT interpretation.

Another SPOT calculation is connectivity,an elaborately conceived but necessarilylimited attempt to quantify the degree ofconnection between pores identified onimages. A limitation is imposed becausetwo-dimensional images can say only somuch about three-dimensional connectivity.Nevertheless, SPOT connectivity has suc-cessfully predicted the productivity of oil

48

nPermeability-porosity relationships for carbonate rocks follow-ing Dunham classification.

nFrom a carbonate field in Italy: an FMI image (left track) allowsSPOT calculation of percentage area of conductive objects. Com-parison (middle track) between SPOT porosity (orange curve) andneutron-density porosity (black curve) shows good agreement.Archie’s exponent “m” (blue curve), computed from Rxo and EPTElectromagnetic Propagation Tool logs, and connectivity of porespace computed from FMI image (gold curve) also correlate (righttrack), except where there is secondary porosity (yellow).

Per

mea

bilit

y, m

d

1000

10

100

1

0.1

0.01

WackestonesMudstones

Vugg

y dolo

mite

Wacke

stone

Pack

ston

es,

Lime

mud

Gra

inst

one,

por

e

diam

eter

dec

linin

g

Moldic grainstone

Parti

cle d

iamet

er =

100

µ

Parti

cle

diam

eter

= 3

00µ

Par

ticle

dia

met

er =

500

µ

Leac

hed

chan

nels

in re

efal

rock

Coccolith chalk pore

diameter < 1µ

0 10 20 30Porosity, %

Fractures LeachingCompaction and cementing

Dep

th, m

X40

X50

X60

X70

0 25 0 5

0 50 Conductive Area%

Porosity

0 25

Variable m

SecondaryPorosity Index

FMIConnectivity

and gas wells in Texan and OklahomanOrdovician carbonates with vuggy, con-nected porosity (above).

Without images, the commonest approachto pore geometry lies through considerationof Archie’s law with its cementation expo-nent m:

in which Rt , Rw and φare, respectively, thewater-filled formation resistivity, connatewater resistivity and porosity.14 Early on,researchers realized that the cementationexponent captured something about thepore space, particularly its tortuosity, andthus could serve to estimate permeability aswell as interpret resistivity logs. Several the-oretical expressions for permeability basedon m have been developed, this being arecent example:

k = 126.7 φm R2 millidarcies,

in which R is an “effective” pore radius inmicrons.15

The exponent m measures reasonably con-stant at about 2 for sandstones, as it does forsimilarly constructed oolitic carbonates. Butotherwise in carbonate rock, it wanders allover. In fractured carbonate rock m tends to

1, and in rocks with nonconnecting vugs mrises to 3, 4 or higher (below, left).16 A par-ticularly copious study on Qatar carbonatesby Focke and Munn shows not only how mvaries with porosity—it varies a greatdeal—but also how that functionalitydepends on permeability.17 The challenge inusing m to evaluate a carbonate thereforedepends on being able to reliably estimatethe exponent at any depth, rather than usean arbitrary value, generally 2, derived fromobservations on sandstones.

Guidelines for achieving this were firstoffered by Lucia of Shell Oil in 1981.18

Using samples from carbonate reservoirs inTexas, USA and Alberta, Canada, Lucianoted that m depended unambiguously onthe proportion of the rock’s porosity comingfrom unconnected vugs. Estimate that fromcore samples, he suggested, and a likely mcould be derived for selected intervals inthe well.

But a more versatile method was soondevised that permitted estimating m foot byfoot. This made use of a new logging mea-surement—high-frequency electromagneticpropagation travel time, or tpl .19 Like theresistivity log, tpl responds to water-filledporosity, but does so without an exponent.Combining resistivity and tpl thereforeallows elimination of porosity for a continu-ous evaluation of m. The results of such anm computation transformed the accuracy ofcarbonate evaluation in a number of MiddleEast fields (next page, top). The methodol-ogy was later extended to take advantage ofyet another wireline measurement, the TDTThermal Decay Time log, permitting thecontinuous evaluation of not just m, but alsothe saturation exponent n.20

The exponent n appears in Archie’s law

adapted for hydrocarbon-bearing rock:

49January 1995

nData fromOrdovician carbon-ate oil reservoirs inOklahoma andTexas, USA, indi-cating a correla-tion between wellproduction andconnectivityderived from FMIimages. The datapertain to wellsknown not to inter-sect fracture andstylolite systems.

nConjectured variation of Archieexponent m in fractured and vuggyrocks. (Adapted from Watfa andNurmi, reference 16.)

12. Nurmi RD and Frisinger MR: “Synergy of Core Petro-physical Measurements, Log Data, and Rock Exami-nation in Carbonate Reservoir Studies,” paper SPE11969, presented at the 58th SPE Annual TechnicalConference and Exhibition, San Francisco, Califor-nia, USA, October 5-8, 1983.

13. Wiener JM, Rogers JA, Rogers JR and Moll RF: “Predicting Carbonate Permeabilities from Wireline Logs Using a Back-Propagation Neural Network,”Expanded Abstracts, SEG 61st Annual InternationalMeeting and Exposition, Houston, Texas, USA,November 10-14, 1991: 285-288.Mohaghegh S, Arefi R, Ameri S and Rose D: “Design and Development of an Artificial NeuralNetwork for Estimation of Formation Permeability,”paper SPE 28237, presented at the SPE PetroleumComputer Conference, Dallas, Texas, USA, July 31-August 3, 1994.

14. “Archie’s Law,” The Technical Review 36, no. 3 (July 1988): 4-13.

15. Watfa M and Youssef FZ: “An Improved Techniquefor Estimating Permeability in Carbonates,” paperSPE 15732, presented at the 5th SPE Middle East OilShow, Manama, Bahrain, March 7-10, 1987.

16. Watfa M and Nurmi R: “Calculation of Saturation,Secondary Porosity and Producibility in ComplexMiddle East Carbonate Reservoirs,” Transactions ofthe SPWLA 28th Annual Logging Symposium, Lon-don, England, June 28-July 2, 1987, paper CC.

17. Focke JW and Munn D: “Cementation Exponents (m) in Middle Eastern Carbonate Reservoirs,” paperSPE 13735, presented at the Middle East Oil Techni-cal Conference and Exhibition, Bahrain, March 11-14, 1985.

18. Lucia FJ: “Petrophysical Parameters Estimated fromVisual Descriptions of Carbonate Rocks: A FieldClassification of Carbonate Pore Space,” paper SPE10073, presented at the 56th SPE Annual TechnicalConference and Exhibition, San Antonio, Texas,USA, October 5-7, 1981.

19. Freeman DW and Henry KC: “Improved SaturationDetermination with EPT,” paper SPE 11466, pre-sented at the SPE Middle East Oil Technical Confer-ence, Manama, Bahrain, March 14-17, 1983.Amin AT, Watfa M and Awad MA: “Accurate Estima-tion of Water Saturations in Complex CarbonateReservoirs,” paper SPE 15714, presented at the 5th SPE Middle East Oil Show, Manama, Bahrain, March 7-10, 1987.

20. Watfa M: “Using Electrical Logs to Obtain the Satu-ration Exponent (n) in the Archie Equation,” paperSPE 21415, presented at the SPE Middle East OilTechnical Conference and Exhibition, Bahrain,March 2-5, 1994.

Rt = Rw

φm Swn ,

Rt = Rw

φm ,

28

Oil

prod

uctio

n, B

OP

D

0

FMI Connectivity

100

200

300

400

500

24201612840

1.0 1.5 2.0 2.5 3.0

Cementation exponent, m

Por

osity

, p.u

.

Fractures Vugs

Fractureporosity

φ nc =

10

50

40

30

20

10

8

6

4

2

1

0.8

0.5

φ nc = 5

φ nc = 2.5φ nc = 2

φ nc = 1

Non connectedporosity

φ nc =

0.5φ

f = 0.1φf = 1

φf = 2.5

φf = 5

φf = 10

in which Sw is water saturation. Like theexponent m, n also runs into trouble in car-bonates, sometimes varying dramaticallyfrom the conventionally assumed value of2.21 This has been shown in several sets ofexperiments on cores, the most recent byBouvier (below).22 Petrophysicists suspectthe likely cause of discrepancy is tiny micro-pores in the micritic matrix. Most probably,these small pores still contain original waterwhile the large pores contain oil. It is alsoprobable that the micrite remains water wet,while the grains have become oil wet. Bothphenomena would explain why carbonateformations producing only oil sometimesexhibit low resistivities more characteristicof a water-bearing formation. Essentially, thewater-filled micropores provide a short-cir-cuit to the survey current.

In summary, the classic Archie approachfor analyzing the complex pore geometriesof carbonate is fraught with obstacles,which have been only partially overcome.

New Logging TechniquesToday, two new techniques—nuclear mag-netic resonance (NMR) logging and Stone-ley wave logging—offer new perspectiveson carbonate permeability and pore struc-ture. The theoretical foundations for bothtechniques have been known for years, butuntil recently neither has received adequatetechnical implementation. That is changingwith the introduction of the CMR Combin-able Magnetic Resonance tool and the DSIDipole Shear Sonic Imager tool.

In nuclear magnetic resonance, sharpmagnetic pulses are used to momentarilyreorient hydrogen molecules away from theambient magnetic field direction. After eachpulse subsides, the hydrogen moleculesrealign themselves with the ambient field,oscillating about it as they do so. Observingthese oscillations permits measuring howmany hydrogen molecules relax after theimposed magnetic pulse and also the rate atwhich they realign to the ambient field,called the relaxation.

The implications for logging are dramatic.The measurement of how many hydrogenmolecules relax provides a measure ofporosity, and the relaxation times indicatethe size of pores containing the hydrogenmolecules. Relaxation times are short insmall pores because the hydrogenmolecules are near the grain surface whereinteraction with surface charges speedsrelaxation. Relaxation times are longer inlarge pores. Measuring the spectrum ofrelaxation times—so-called T2 relaxation

50 Oilfield Review

nArchie exponent“m,” determinedfrom EPT and resis-tivity logs, com-pared with sec-ondary porosity,determined by sub-tracting sonicporosity from neu-tron-density poros-ity—in the Arabcarbonate forma-tion. In the purelimestone zone atthe top, the twoparameters matchwell. In thedolomite zone atthe bottom, theydiverge, indicatingthe presence ofunconnectedvuggy porosity.[Reprinted fromNurmi et al.: “Typecasting Hetero-geneities,” MiddleEast Well EvaluationReview, no. 10(1991):16-32.]

nCarbonate rock’sresistivity-versus-saturation behav-ior for three car-bonate samples,indicating depen-dence on pore-sizedistribution. Thestraight-linebehavior at highsaturations is likelydominated bylarge pores, whilethe behavior at lowsaturations is dom-inated by smallpores. Resistivityindex is defined asthe ratio of rockresistivity at anarbitrary satura-tion to rock resistiv-ity when water sat-urated. (Adaptedfrom Bouvier, refer-ence 22.)

50

Res

istiv

ity in

dex

Water saturation, %

10

60 70 80 90 100

8

6

4

2.5

1.5

1Microns

0.04 63

XX00

2.00

0 Secondary Porosity, p.u.

Cementation Exponent, m

Lithology Analysis

Fluid Volume

25

100.00

0 100

%

%

Residual oil

Moved oil

Limestone

Dolomite

Water

0.007.00

Dep

th, f

t

XX50

7.1 p.u.1.1 md

Core Sample 2

5.9 p.u.0.2 md

Core Sample 1

T2, msec10,0001000100101

12.2 p.u. 7.5 md

3 days

6 hours

1.5 hours

0 hours

Core Sample 3

times—resulting from each pulse promisesto give an indication of the range of poresizes in reservoir rock. In sandstones, com-parisons between T2 relaxation times andmercury porosimetry, a standard lab tech-nique for evaluating pore sizes—pore necksizes to be precise—are generally good(above).23 This indicates that in sandstones,there is a predictable relationship betweenpore and pore neck sizes. Researchers areconducting similar measurements on car-bonates, but results so far have not shownthe same predictable relationship.

The CMR measurement is made from apad-type tool with permanent magnets thatprovide an ambient field focused entirely into

the formation. This rules out the possibility ofa borehole signal, a problem that plaguedearlier technology that used instead the muchweaker and unfocused earth’s field. Eliminat-ing the borehole signal used to require theexpensive and unpopular technique of dop-ing the entire mud column with magnetite.The new tool’s depth of investigation is about1 in. [2.5 cm], and a dead zone directly infront of the pad avoids most effects frommudcake or rugosity. Vertical resolution isjust 6 in. [15 cm], facilitating comparisonswith the high-resolution FMI logs.

Recent CMR logs run in carbonate forma-tions in West Texas coupled with laboratorymeasurements on cores from the wells illus-trate exciting possibilities for overall petro-physical evaluation.24 The formations inquestion are partly dolomitized carbonateswith a good deal of nonconnected vuggyporosity. In addition, silt layers create verti-cal permeability barriers. The main interpre-

tation challenge is to estimate at any depthwhat percentage of porosity actually con-tributes to production. This requires beingable to discount the minute pore space inthe silt and also any vuggy porosity that isnot connected.

T2 spectra were measured on water-satu-rated cores both before and after they hadbeen centrifuged to expel all produciblewater (below). Before centrifuging, the spec-tra show water-filled porosity covering thefull range of pore sizes, while spectra aftercentrifuging no longer show the large poresizes, since the water has been expelledfrom them. Equating the porosity differencebetween the two spectra with the volume ofwater expelled during centrifuging estab-lished a T2 cutoff of 95 msec to divide largefrom small pores. Applying this cutoff tospectra measured by the CMR tool providedan estimate of small-pore porosity that cor-related well with silt intrusions evaluated

51January 1995

21. “Archie II: Electrical Conduction in Hydrocarbon-Bearing Rock,” The Technical Review 36, no. 4(October 1988): 12-21.

22. Bouvier L: “Les Saturations en Réservoir Carbonatéà Double Porosité: Réconciliation des Mesures Lab-oratoires et Diagraphies,” Pétrole et Techniques 375(October 1992): 21-24.

23. Morriss CE, MacInnis J, Freedman R, Smaardyk J,Straley C, Kenyon WE, Vinegar HJ and Tutunjian PN:“Field Test of an Experimental Pulsed Nuclear Mag-netism Tool,” Transactions of the SPWLA 34thAnnual Logging Symposium, June 13-16, 1993,Calgary, Alberta, Canada, paper GGG.

24. Chang D, Vinegar H, Morriss C and Straley C: “Effec-tive Porosity, Producible Fluid and Permeability inCarbonates from NMR Logging,” Transactions of theSPWLA 35th Annual Logging Symposium, Tulsa,Oklahoma, USA, June 19-22, 1994, paper A.

nT2 spectra for threesamples during labo-ratory centrifuging todrive water from thepores. As centrifug-ing continues for atotal duration ofthree days, water isgradually expelledfrom large pores,leaving some waterin the small ones.

nDecaying signal amplitude from adownhole nuclear magnetic resonancemeasurement (top) is transformed to a T2relaxation spectrum (middle), and theninterpreted in terms of rock pore geome-try (bottom). (Adapted from Morriss et al.,reference 23.)

Sig

nal a

mpl

itude

0 100 200 300 400

Sig

nal d

istri

butio

n

0.1 1.0 10 100 1000

Por

osity

vol

ume

Pore diameter, microns1 10 100

NMR Measurement

T2 Spectrum

Pore Geometry

Time, msec

T2, msec

0

1

0

8

from other logs. Following visual analysis ofthe cores, a second cutoff at 750 msec wasselected to isolate vugs from intergranularporosity. This was also applied to spectrameasured downhole, providing a log ofvuggy porosity (below).

Recent laboratory work on core samplesfrom the carbonate Mubarraz field in AbuDhabi, UAE, confirms the potential of NMRmeasurements.25 A challenge in this area isto distinguish small micropores in themicrite matrix from the much larger pro-ductive intergranular pores. Analyzing 20samples from two wells, a team of Schlum-berger and Abu Dhabi Oil Company geo-scientists found that micropores were cor-rectly identified using a relaxation timecutoff of 190 msec on laboratory-measuredT2 spectra. Furthermore, permeable grain-stone facies could be distinguished fromlower-permeability packstones and mud-stones with a cutoff of 225 msec. Finally,the NMR data could be interpreted to givemore accurate permeability estimates thanthose obtained from conventional porositylogs. The CMR logging tool is currentlybeing tested in Abu Dhabi, and expecta-tions are high that similarly impressiveresults will be obtained in boreholes.

Another logging tool, the DSI imager,gains direct entry to permeability by physi-cally moving fluid through the formation.This is achieved when low-frequency tubewaves—called Stoneley waves—propagate

up and down the borehole. The Stoneleywave preserves most of its energy in theborehole, but in permeable formationssome energy is attenuated when wave pres-sure pushes fluid from the borehole into theformation, similar to a quick, small-scalewell test. This slows the velocity of thewave by an amount that can be related tothe ratio of formation permeability to fluid

52 Oilfield Review

nMoldic porosity (purple). Formerly a fos-siliferous micrite, the micrite has beenreplaced by dolomite and the more solu-ble calcitic fossils dissolved to leave amoldic porosity. Upper Eocene OcalaLimestone, Florida. Gypsum plate added.

nT2 spectra for vuggy carbonates showing dual-porosity system, and a log of vuggyporosity obtained from evaluating the right end of the NMR spectra only, above a cut-off of 750 msec.

25. Kenyon WE, Takezaki H, Straley C, Sen PN, HerronM, Matteson A, Petricola MJ: “A Laboratory Study ofNuclear Magnetic Resonance Relaxation and itsRelation to Depositional Texture and PetrophysicalProperties—Carbonate Thamama Group, MubarrazField, Abu Dhabi,” paper SPE 29886, presented atthe SPE Middle East Technical Conference and Exhi-bition, Bahrain, March 11-14, 1995.

26. Winkler KW, Liu H-L and Johnson DL: “Permeabilityand Borehole Stoneley Waves: Comparison BetweenExperiment and Theory,” Geophysics 54 (January1989): 66-75.

27. Petricola M and Frignet B: “A Synergetic Approach toFracture and Permeability Evaluation from Logs,”paper SPE 24529, presented at the 5th Abu DhabiPetroleum Conference and Exhibition, Abu Dhabi,UAE, May 18-20, 1992.

28. Cassell B, Badri M and Faulhaber J: “PermeabilityPrediction Based on Anelastic Attenuation UsingDipole Shear and Low Frequency Monopole Sourcesin a Carbonate Reservoir in Saudi Arabia,” presentedat the GEO-94 Middle East Geosciences Exhibition &Conference, Bahrain, April 25-27, 1994.

viscosity. Given a viscosity for the boreholefluid, in well-controlled circumstances suchas laboratory measurements or boreholeswith no mudcake, the permeability canthen be estimated.26

The DSI tool generates Stoneley waveswith a special monopole transmitter at fre-quencies of 600 Hz to 5 kHz, ideal fortube-wave logging and a quantum leapahead of previous technology equippedwith transmitters operating in the 10 to 20kHz range. Recent estimation of permeabil-ity using Stoneley-wave velocity as obtainedfrom the DSI tool shows impressive agree-ment with core permeability measurementsin an Abu Dhabi carbonate reservoir (nextpage, top right).27

The method of obtaining permeabilityusing Stoneley-wave velocity requiresknowing the formation’s density and shearvelocity. A second method establishes per-meability from the Stoneley wave withoutother data. This method is based on observ-ing how permeability attenuates Stoneley-wave energy by directly comparing signalsfrom near and far receivers.28 Attenuation is

8.7 p.u.0.7md

X200

X150

X100

XX5010-1 1 10 102

kNMR, md

20

Porosity, p.u.

10 0

Core

Vug porosity

Vug

1 100 10,000T2, msec

11.9 p.u.13 md

10.9 p.u.0.2 md

6.6 p.u.0.07md

Dep

th, f

t

0.18 mm

greater at higher frequencies, so the com-parison is more sensitive if measured at thehigh end of the Stoneley-wave frequencyspectrum. Excellent agreement has beenobserved in Middle East carbonate reser-voirs between permeability estimatesobtained using this second method and pro-duction logs and core data (bottom).

Research continues into improving Stone-ley-wave permeability, for example in

53January 1995

nFilled fenestral porosity in a blue-greenalgal biolithite. Porosity may be due to airspaces in crinkled mat sediments. UpperSilurian Limestone, Pennsylvania, USA.

XX50

Fluid Analysis

X100

X150

X200

X250

X300

X350

Dep

th, f

t

100 50 0 Low High

Permeability index fromStoneley velocity

CumulativeStoneley Perm.Spinner

Permeability index fromStoneley attenuation

HighLow0 10 20

Residual oil

Moved oil

Water

Caliper

Permeability index fromStoneley velocity

Permeability index fromStoneley attenuation

XX900

X1000

X1100

X1200

X1300

X1400

Dep

th, f

t Core permeability

Permeability

nPermeability estimated fromStoneley-wave attenuation withoutrecourse to other log data, comparedwith permeabilityfrom Stoneley-wavevelocity, in a MiddleEast carbonatereservoir. Integratedpermeability showsan excellent matchwith a flowmeterproduction log.

nPermeabilityevaluated fromStoneley-wavevelocity and atten-uation comparedwith core measure-ments in a MiddleEast carbonatereservoir. Evalua-tion using velocityrequires formationdensity and shear-velocity measure-ments.

0.26 mm

accounting for the presence of mudcake,which almost certainly interferes with thetube wave’s ability to move formation fluids.Large-Scale FeaturesMapping reservoirs at the large scale andunderstanding their complex petrophysics atthe small scale are all part of the challengefacing reservoir geologists and engineers.But in carbonates, additional care must betaken to recognize and evaluate two typesof medium- to large-scale features that arecaused by overburden and tectonic stresses.Either can dramatically affect reservoir per-formance, creating heterogeneous oranisotropic behavior where none might oth-erwise be suspected. These two features arestylolites and fractures.29

Stylolites occur in any sedimentary forma-tion, but are particularly common in car-bonates—picture the thin, sawtooth “veins”visible on polished marble tiles and floors.Stylolites are easily recognized on outcropsand cores as irregular planes of discontinu-ity between rock units. Formed during com-paction, probably through the mechanismof pressure solution, stylolites concentratefine-grained insoluble residue along theirirregular seams. They are usually assumedto act as permeability barriers, but somecore measurement results confirm that stylo-lites can develop permeability. Identifyingthem and evaluating their imact on perme-ability are therefore top priorities for thereservoir engineer.

Borehole imaging has greatly facilitatedthe identification of stylolites downhole(right and next page). Viewed with the FMI

54 Oilfield Review

nFilled shelter porosity beneath a largemollusk fragment. Pliocene and Pleis-tocene Marl, Florida.

nThree types ofstylolites identifi-able on FMIimages. Dark col-ored stylolites (left),probably filledwith clay; stylolitesassociated with alight band (below),are probably resis-tive calcite. Somestylolites are asso-ciated with exten-sional fractures(next page, left).

0.29 mm

XX32

XX33

XX34

XX35

Dep

th, f

t

XX80

XX81

XX82

Dep

th, f

t

tool, they appear in three common varieties.First, some stylolites exhibit undulating butslightly irregular surfaces and are filled withdark, therefore conductive material, proba-bly clay. A second group of stylolites seemsto have an associated band of light color,most likely resistive calcite. The third type ofstylolite clearly shows associated exten-sional fractures caused by excessive over-burden stress.

The question remains: Which stylolitesform permeability barriers and which donot? Until recently, there has been no sureway of deciding. Now, answers are obtain-able from a third-generation wireline testingtool, the MDT Modular Formation Dynam-ics Tester tool. Unlike previous wirelinetesters, this tool permits testing betweenprobes set as far apart as 8 ft [2.4 m], a large

enough interval to comfortably straddle astylolite. In such tests recently performed inthe Middle East, MDT measurements indi-cated that stylolites previously assumed tobe completely impermeable may in fact be

partly conductive to fluid flow.If stylolites generally impede flow, frac-

tures almost always enhance it. Indeed,some reservoirs, particularly carbonateones, rely exclusively on fractures toachieve commercial levels of production.Before the advent of wireline imaging tech-niques, detecting fractures was difficult andcharacterizing anything about them wasalmost impossible. That bleak outlookchanged dramatically with the introductionof the FMI and DSI tools. The recently intro-duced ARI Azimuthal Resistivity Imager toolalso makes an important contribution infracture detection.

Briefly, all three tools contribute to frac-ture interpretation, but each alone may notprovide a complete picture.30 On FMIimages, open fractures filled with invadingwater-base mud of high conductivity arerecognizable as dark and usually frag-mented sinusoid traces. With the help ofinteractive FracView image processing, theinterpreter can reliably pinpoint fractures,calculate their dip and azimuth, and esti-mate spatial density at the borehole. Addi-tional analysis of image resistivity near thefracture can also lead to an estimation offracture aperture.31

With simple models of fracture geometry,the combined log information may providean effective fracture permeability. This canthen be integrated with permeability esti-mates for the unfractured part of the rock toyield a permeability for the whole rock. In

55January 1995

nReduced fracture porosity (black). Upper Jurassic Limestone, Germany.Cross-polarized light photograph.

29. For general reference:Nelson RA: “Analysis of Anisotropic Reservoirs,” inGeologic Analysis of Naturally Fractured Reservoirs.Houston, Texas, USA: Gulf Publishing Company,1985.

30. Cheung PS-Y and Heliot D: “Workstation-BasedFracture Evaluation Using Borehole Images,” paperSPE 20573, presented at the 65th SPE Annual Tech-nical Conference and Exhibition, New Orleans,Louisiana, USA, September 23-26, 1990.

31. Luthi SM and Souhaité P: “Fracture Apertures fromElectrical Borehole Scans,” Geophysics 55 (July1990): 821-833.

0.22 mm

XX88

XX89

XX90

Dep

th, f

t

XX91

the Rocky Mountains, where a low-porositycarbonate reservoir depends on fractures forproduction, such a combined permeabilityhas been successfully compared to perme-ability obtained from drill-stem tests (left).

There are a few caveats, however, to frac-ture interpretation using FMI resistivityimages. First, the calculated fracture aper-ture seems to be influenced by the fluidoriginally filling the fracture—fractures inwater zones appear systematically widerthan nearby fractures in hydrocarbon zones.It is suspected that invasion fails to removeall hydrocarbon from the walls of the frac-ture, thereby making the fracture look thin-ner to electrical imaging techniques.32 In adepleted carbonate field being exploited foradditional oil using horizontal wells, thisphenomenon has been put to good use inidentifying fractures that are likely to allowwater breakthrough (next page).

Second, the FMI tool is a relatively shal-low measurement, and this limits the tool’sability to distinguish natural fractures thatcontribute to reservoir performance fromdrilling-induced fractures that do not. Cer-tain types of drilling-induced fractures areeasily recognized by their geometry—forexample, vertical fractures oriented perpen-dicular to the least horizontal stress andtherefore intersecting a vertical boreholeover a lengthy interval. Nonvertical drilling-induced fractures, however, are harder todistinguish and may be easily confused withthe natural variety. Fracture identification inhighly deviated and horizontal wellsbecomes harder still.

The ARI tool provides some added depthof investigation but a poorer along-boreholeresolution, and as a result, fewer fracturesare detected. However, ARI image process-ing provides some clue to fracture depth aswell as aperture, although neither is unam-biguously determined.33 The two parametersare genetically linked, so the tool responseto a fracture enables an estimate of one ofthe parameters once a value for the other istaken.

Greater depth of investigation, up to sev-eral meters, is provided by the DSI tool thatdetects open fractures in the same way thatit senses a permeable formation—byemploying the Stoneley wave to physicallypulse mud into and out of the fracture.34

However, there is a commensurate deterio-ration in resolution along the borehole axis,to about 1.5 m [4.9 ft], showing closelyspaced fractures as a single fracture. Like theFMI measurement, the Stoneley wave per-mits the evaluation of fracture aperture,though again this may actually represent the

56 Oilfield Review

nFracture evaluation and estimation of fracture aperture from an FMI image in aRocky Mountain carbonate reservoir. This information was used to derive an effectivepermeability that matched well test data. Squares and triangles signify a level of confi-dence in fracture identification, with squares having a higher level than triangles.

nFilled fracture porosity, with conjugateset of large fracture veins. Middle Ordovi-cian Limestone, Pennsylvania.

nEnlarged moldic porosity (black). Notelarge mollusk mold with upper edgeenlarged. Other grains include mainlymiliolid foraminifera. Upper OligoceneLimestone, Florida. Cross-polarized lightphotograph.

0.26 mm 0.22 mm

Mean aperture

XX00

XX10

0.001 mm 10.00D

epth

, ft

cumulative apertures of several neighboringfractures. Comparisons between fractureaperture estimated from the two techniqueshave shown good agreement in metamor-phic volcanics at a UK waste disposal site.On the down side, the Stoneley wave yieldsno information on fracture dip or azimuth.

There may be further to go in fractureinterpretation, but a comparison betweenthe techniques of ten years ago and those oftoday reveals the extraordinary advancesachieved by novel wireline technology anda spirited community of interpreters. It is alevel of improvement common to all areasof carbonate interpretation. Today’sexploitation in increasingly complex anddifficult reservoirs has gained from a verita-ble revolution in formation evaluation. Longmay it continue. —HE, LS

57January 1995

nFracture aper-tures determinedfrom FMI images intwo subhorizontalwells in a fracturedand partlydepleted reservoir.The apertures inthe top exampleshow four zoneswhere aperture ishigher than else-where. This wasinterpreted to indi-cate the fracturesare water-filledand liable to waterout early. Aftercompletion, the topwell produced with50% water cut; thebottom well pro-duced no water.

32. Standen E, Nurmi R, ElWazeer F and Ozkanli M:“Quantitative Applications of Wellbore Images toReservoir Analysis,” Transactions of the SPWLA 34thAnnual Logging Symposium, Calgary, Alberta,Canada, June 13-16, 1993, paper EEE.

33. Faivre O: “Fracture Evaluation from QuantitativeAzimuthal Resistivities,” paper SPE 26434, presentedat the 68th SPE Annual Technical Conference andExhibition, Houston, Texas, USA, October 3-6,1993.

34. Hornby BE, Johnson DL, Winkler KW and PlumbRA: “Fracture Evaluation Using Reflected Stoneley-Wave Arrivals,” Geophysics 54 (October 1989):1274-1288.

Oil/watercontact

Potential water entry

Oil/watercontact

m

Mea

n a

pert

ure,

mm

300

.001

10

Mea

n a

pert

ure,

mm

.001

10

400 500

400 500 600m