Calcite cement in Permian deep-water sandstones, Delaware ... · stone in the upper Bell Canyon...

AUTHOR

Shirley P. Dutton � Bureau of EconomicGeology, Jackson School of Geosciences, Uni-versity of Texas at Austin, Austin, Texas 78713-8924; [email protected]

Shirley Dutton is a senior research scientist atthe Bureau of Economic Geology with researchinterests in sedimentary petrology, clastic diagen-esis, sedimentology, and reservoir characteriza-tion. She received a B.A. degree from the Universityof Rochester and an M.A. degree and Ph.D. fromthe University of Texas at Austin, all in geology. Shewas an AAPG Distinguished Lecturer in 1987.

ACKNOWLEDGEMENTS

This research was funded by the United StatesDepartment of Energy (DOE) under contractno. DE-FC22-95BC14936 and by the State of Texasunder State Match Pool 4201 and as part of theState of Texas Advanced Resource Recovery project.The DOE project manager was Daniel J. Ferguson.The Bureau of Economic Geology acknowledgessupport of this research by Landmark Graphicsvia the Landmark University Grant Program.Support for this work was provided in part by theJohn A. and Katherine G. Jackson School of Geo-sciences and the Geology Foundation at the Uni-versity of Texas at Austin. Project partners Conoco,Inc., and Orla Petco, Inc., are thanked for pro-viding data fromGeraldine Ford and East Ford fields,respectively. Special thanks to William A. Flandersof Transpetco Engineering for an insightful col-laboration on the study of East Ford field. JoseGuzman, Helena Zirczy, and Daniel Mendez pro-vided research assistance, and Sigrid Clift andJose Guzman described the cores from FordGeraldine Unit. Isotopic analyses were done atthe Stable Isotope Laboratory, directed by PeterSwart, at University of Miami’s Rosenstiel Schoolof Marine and Atmospheric Science. I benefitedfrom discussions with A. R. Dutton, J. W. Jennings,Jr., R. G. Loucks, K. L. Milliken, and S. C. Ruppel.Constructive reviews by peer reviewers David E.Eby, Joseph R. Studlick, and Robert C. Trenthamand AAPG editor Gretchen M. Gillis improvedthis article. Illustrations were prepared by thegraphics staff of the Bureau of Economic Geologyunder the direction of Joel Lardon, graphics man-ager. Lana Dieterich edited the manuscript. Pub-lication was authorized by the Director, Bureauof Economic Geology, John A. and Katherine G.Jackson School of Geosciences, University ofTexas at Austin.

Calcite cement in Permiandeep-water sandstones,Delaware Basin, west Texas:Origin, distribution, and effecton reservoir propertiesShirley P. Dutton

ABSTRACT

Calcite cement is the dominant control on reservoir quality inturbidite sandstones of the Upper Permian Bell Canyon For-mation, Delaware Basin. These well-sorted, very fine-grainedarkoseswere deposited in a basin-floor setting by channel-leveesystems terminating in broad lobes. Calcite cement distribu-tion in the East Ford and Geraldine Ford fields was mappedusing core, log, and thin-section data. Calcite is concentrated intightly cemented zones that are mostly less than 1 ft (0.3 m)thick. Areas that have high percentages of calcite-cementedsandstone (>20%) occur along the margins of the sandstonebodies, in overbank and lobe deposits, where sandstone pinchesout into siltstone. Areas that have the lowest percentage ofcalcite-cemented sandstone (<10%) occur where the sandstoneis thickest, in the channel facies.

Isotopic composition of the calcite (d13C= �1.8 to �3.0x[relative to the Peedee belemnite, PDB], d18O = �4.6 to�6.3x [PDB]) is consistent with the source of calcium car-bonate being from dissolution of detrital carbonate rock frag-ments and marine skeletal debris. Because internal sources ofcalcite were apparently insufficient to account for the cementvolume, cement components are interpreted as having beentransported into the sandstones from organic-rich basinal silt-stones and limestones. Feldspars buffered acidic formation wa-ters near where they entered the sandstone, resulting in cal-cite concentrated near the sandstone margins. The calciteformed near maximum burial depths of 4800 ft (1.5 km) and

AAPG Bulletin, v. 92, no. 6 (June 2008), pp. 765–787 765

Copyright #2008. The American Association of Petroleum Geologists. All rights reserved.

Manuscript received September 25, 2007; provisional acceptance November 19, 2007; revised manuscriptreceived January 14, 2008; final acceptance January 28, 2008.

DOI:10.1306/01280807107

temperatures of 104jF (40jC) from marine porewaters with d18O of approximately 0x (relativeto standard mean ocean water).

Most of the calcite-cemented zones are inter-preted as being concretions that extend no morethan a few meters laterally. Production data andgeophysical log correlations suggest that some ce-mentedzones are laterally continuous at least 1000 ft(300m) and cause vertical reservoir compartmen-talization. Laterally extensive calcite layersmay beassociated with the base of turbidite deposits.

INTRODUCTION

Calcite cement exerts a strong control on sand-stone reservoir properties, particularly in shallow-buried or young sandstones containing little quartzcement. Because calcite cement reduces reservoirporosity and permeability and affects fluid flowduring production, an improved understanding ofthe controls on calcite cement distribution has eco-nomic significance. Determining the origin of cal-cite cement in a sandstone can improve our abilityto predict its distribution. This case study of thereservoir-scale distribution of calcite cement inves-tigated Permian deep-water turbidite sandstonesof theDelawareMountainGroup todetermine the(1) origin of the cement, (2) the controls on ce-ment distribution, and (3) the effect of the cementon reservoir properties. The study focused on cal-cite cement distribution in the East Ford andGeral-dine Ford fields in the Delaware Basin, west Texas(Figure 1).

Many previous studies have examined the ori-gin of carbonate cement in deep-water turbiditesandstones (e.g., Boles and Ramseyer, 1987; Slattand Hopkins, 1990; Moraes and Surdam, 1993; Mc-Bride et al., 1995;Hendry et al., 1996; Boles, 1998;Silva et al., 1998; Sombra et al., 1998; Souza andSilva, 1998; Beaubouef et al., 1999; Anjos et al.,2000; Macaulay et al., 2000; Stewart et al., 2000).In these studies, various controls on cement dis-tribution in turbidites are identified, including con-tact with interbedded shales (Anjos et al., 2000),location of internal shale layers and mudclast lags(Moraes and Surdam, 1993), zones of permeabil-

ity contrast at the top or bottom of beds (McBrideet al., 1995), bioclast-rich beds (Hendry et al., 1996;Sombra et al., 1998) or channel lags (Beaubouefet al., 1999), zones of higher permeability and great-er fluid flow (Souza and Silva, 1998; Anjos et al.,2000), faults and fractures (McBride et al., 1995),salt tectonics and associated faulting and verticalfluid movement (Stewart et al., 2000), oil migra-tion along faults (Macaulay et al., 2000), contactswith carbonate-rich beds at the base of the reser-voir (Silva et al., 1998), location of the oil-watercontact (Souza and Silva, 1998), and oil biodeg-

radation at paleo-oil-water contacts (Watson et al.,1995).

Relatively few of these studies have sufficientdata to determine three-dimensional calcite cementdistribution at the reservoir scale or the effect ofcalcite cement on reservoir flow in turbidite reser-voirs.A studyof two-dimensional outcrops of Perm-ian turbidites of the Brushy Canyon Formation,west Texas, concludes that channel-lag depositscontain abundant bioclastic carbonate and have lowporosity because of preferential calcite cementation(Beaubouef et al., 1999). The lags are best preservedalong the margins of the channel and removed fromthe axes. Lags are also associatedwith surfaces that

Figure 1. Location map of East Ford and Geraldine Ford fieldsin the Delaware Basin and paleogeographic setting during theLate Permian (modified from Dutton et al., 1999).

766 Calcite Cement in Permian Deep-Water Sandstones

define channel stories, at the base and top of mas-sive sandstones. A halo of increased calcite cementaround the channels may result from the positionof channel-lag remnants (R. T. Beaubouef, 2001,personal communication).

Carbonate cement (calcite and dolomite) inLower Cretaceous turbidite sandstones of the Upa-nema field, Potiguar Basin, Brazil, is more laterallycontinuous in lobe deposits than in channel deposits(Moraes and Surdam, 1993). Because carbonate-cemented zones inmassive channel sandstones areassociated with discontinuous mudclast lags, theyare dispersed and of short lateral extent. In lobefacies, laterally extensive carbonate-cemented layersare associated with continuous shaly zones betweenturbidite depositional packages. The calcite cementenhanced permeability anisotropy inherited fromdepositional architecture because the cement is as-sociated with shaly zones.

The Cretaceous Namorado Sandstone of theCampos Basin, Brazil, is composed of massive tur-bidite sandstones that contain calcite-cementedintervals 4 to8 in. (10 to20 cm) thick (Sombra et al.,1998). The cemented zones have a vertical fre-quency of 1/m in cores and occur in bioclast-richlayers. Horizontal cores cut concretions up to 20 ft(6 m) long, but no periodicity was observed in thehorizontal distribution of concretions.Models thatbest fit the ratio of vertical permeability/horizontalpermeability (Kv/Kh) fromwell testswere the onesconstructedwith small concretions 1–2m (3–6 ft)long.

Calcite-cemented zones in massive Paleoceneturbidite sandstones of the Balmoral field, NorthSea, occur as strata-bound concretions in the mid-dle of continuous sandstones (Slatt and Hopkins,1990). Cement occurs in both lobe and channelsandstones, and average permeability is reduced byabout half in both facies when cemented and un-cemented samples are averaged together. Diagramspresented by Slatt andHopkins (1990) suggest thatcalcite-cemented zones are somewhat more abun-dant in lobe facies than in channel facies and thatcalcite-cemented zones are more common in thelower part of the reservoir.

Miocene turbidites at the North Coles Leveefield in the San Joaquin Basin, California, contain

carbonate-cemented zones that average 1 ft (0.3 m)in thickness and compose 6 to 8% of the total res-ervoir thickness (Boles and Ramseyer, 1987). Thecemented zones are not correlatedwith either grainsize or vertical position in upward-fining cycles, andthey are generally less than 300 ft (90 m) in lateralextent.

In this study of the East Ford and GeraldineFord fields in west Texas (Figure 1), closely spacedlog and core data provided reservoir-scale infor-mation on calcite cement distribution, and pro-duction data from the East Ford field revealed theinfluence of calcite cement on reservoir flow. Thestudy interval was the oil-producing Ramsey sand-stone in the upper Bell Canyon Formation, whichis undergoing CO2 flooding in both fields (PittawayandRosato, 1991;Dutton et al., 2003). This data setprovides a rare opportunity to determine the three-dimensional distribution of calcite cement in sub-surface reservoirs.

GEOLOGIC SETTING

Upper Permian (Guadalupian) Delaware Moun-tain Group strata compose a succession of slopeand basin deposits in the Delaware Basin in westTexas and southeast New Mexico. The DelawareBasinwas semirestricted, its south end partly opento the seaway and its north end surrounded by anextensivecarbonate shelf-and-reefcomplex(Figure1).TheBellCanyonFormation, the focus of this study,is the youngest formation in the Delaware Moun-tainGroup, and it is composed of interbedded sand-stones, siltstones, and limestones (Meissner, 1972;Fischer and Sarnthein, 1988; Gardner, 1992). Dur-ing sea-level highstands, organic-rich siltstones ac-cumulated on the basin floor by slow settling fromthe suspension of marine algal material and air-borne silt. Associated limestone tongues withinthe Bell Canyon Formation were deposited by sed-iment gravity flows that originated by slumping ofcarbonate debris along the flanks of the rapidly ag-grading carbonate platform (Gardner, 1992; Brownand Loucks, 1993a, b). During sea-level lowstands,siliciclastic sands were carried into the basin by tur-bidity currents.Claywas carried by thewindbeyond

Dutton 767

the Delaware Basin, accounting for the scarcityof detrital clay-sized sediment in the DelawareMountain Group deposits (Fischer and Sarnthein,1988).

TheEast Ford andGeraldine Ford fieldsproduceoil from the Ramsey sandstone, the youngest sand-stone in the Bell Canyon Formation (Figure 2).Depositional models of the Geraldine Ford and EastFord fields were developed using data from BellCanyon outcrops and subsurface data (Barton andDutton, 1999; Dutton et al., 1999; Dutton et al.,2003). Stratigraphic relations in outcrop indicatethat the Bell Canyon sandstones were deposited ina basin-floor setting by a system of leveed channelsthat have attached lobes and overbank splays thatfilled topographically low interchannel areas (Bartonand Dutton, 1999, 2007). Water depths were be-tween 1000 and 2000 ft (300 and 600 m) duringdeposition of the Bell Canyon Formation (Keranset al., 1992).

Maximumburial depth of the Bell Canyon sand-stones at the East Ford and Geraldine Ford fieldswas approximately 4800 ft (1450 m). Maximumburial was reached by the Early Jurassic, and upliftduring the Jurassic and again in the late Tertiaryremoved about 2000 ft (600 m) of overburden inthis area (Williamson, 1978), leaving the Ramsey

sandstones at a depth of approximately 2755 ft

(840 m) in both fields. The present geothermalgradient is 1.0jF/100 ft (18.2jC/km) (Woodruffet al., 1984). The maximum burial temperature ofthe Ramsey sandstones at the East Ford and Geral-dine Ford fields was approximately 125jF (50jC),assuming a mean annual surface temperature of77jF (25jC) and a geothermal gradient of 1.0jF/100 ft (18.2jC/km) in the Jurassic.

The oxygen isotopic composition (d18O) of for-mation water in the Sullivan oil field, which pro-duces from theBell Canyon Formation 2mi (3 km)south of the East Ford field, is �2.0x (relative tostandardmean oceanwater, SMOW) at a depth of2700 ft (820 m) (Williamson, 1978). Formationwater in the Geraldine Ford and East Ford fieldsis assumed to have a similar isotopic composition,but no data were available. Chloride concentrationof formation water in the two fields ranges from40,000 ppm in the southwest to 70,000 ppm in thenortheast (Ruggiero, 1985; Dutton and Flanders,2001a). A regional trend of decreasing salinity to thesouthwest results from late Tertiary tilting of theDelaware Basin and flow of meteoric water intoDelaware sandstones in outcrop (McNeal, 1965).

METHODS

Vertical and areal distribution of calcite-cementedintervals in the Geraldine Ford field was mappedfrom descriptions of 3615 ft (1102 m) of Ramseysandstone core from 70 wells in the field (Duttonet al., 1999).Calcite-cemented zones have a distinctwhite color in the core, and identification of ce-mented zones was confirmed by hydrochloric acid.These datawere supplemented by core descriptionsby Ruggiero (1985) from 13 additional wells. Be-causemanywells have only a gamma-ray log,whichcould not be used to identify cemented intervals,calcite distribution was mapped from core dataalone.

Only one corewas available from theEast Fordfield, so calcite-cemented zones in other wells inthe East Ford field were identified on interval tran-sit time logs as prominent high-velocity deflections.Core and log descriptions were supplemented byporosity and permeability analyses in the Ramsey

Figure 2. Typical log from the 24 East Ford unit well (fromDutton et al., 2003; reprinted with premission from the AAPG).SH1 = laminated siltstone.


interval. A total of 4752 core analyses of sandstonesamples from 153 cored wells from the GeraldineFord field and 369 sandstone samples from12 coredwells from the East Ford field identified calcite-cemented zones that have lowporosity and perme-ability. TheEast Ford andGeraldine Ford fields havebeen unitized and named the East Ford unit (EFU)and the Ford Geraldine unit (FGU).Wells locatedin the units are called EFU and FGU, respectively.

The composition of Bell Canyon sandstone andsiltstone from the East Ford and Geraldine Fordfieldswas determined by standard point-count anal-ysis (200 points) of 68 thin sections and by catho-doluminescent petrography.Chips used tomake thethin sections were end trims of core-analysis plugsso that petrographic parameters could be comparedwith porosity and permeability. Thin sections werestained for K-feldspar and carbonates. Authigenicmineralswere categorizedby location, that is,wheth-er they filled primary pores or replaced frameworkgrains. Framework grain size and sorting were esti-mated by grain-size point counts (50 points). Cal-cite cement was analyzed for stable isotope ratiosof carbon and oxygen.

RESULTS

Petrography of Ramsey Sandstones

The Ramsey sandstones in the East Ford and Ge-raldine Ford fields are very fine grained and havea narrow range of grain sizes. The average grainsize is 0.092 mm (3.45 f), and the range is 0.056–0.106mm (4.1–3.3 f). Sorting is good tomoderate,averaging 0.45 f standard deviation and rangingfrom 0.33 to 0.62 f (sorting of Folk, 1974).

The Ramsey sandstones (Figure 3) are arkosesthat have an average composition of quartz = 65%,feldspar = 29%, and rock fragments = 6%. Plagio-clase andpotassiumfeldspar are approximately equalin abundance. The most common lithic grains aremetamorphic, plutonic, and carbonate rock frag-ments. Detrital carbonate rock fragments and fos-sils (mostly crinoid and fusulinid fragments) togetherconstitute 1% of the present rock volume. No de-trital clay-sized matrix in these Bell Canyon sand-

stones is observed, reflecting the interpreted eoliansediment source (Fischer and Sarnthein, 1988).

Cements and replacive minerals constitute be-tween 1 and 31% of the sandstone volume, calciteand chlorite being the most abundant authigenicminerals. Calcite has an average volume of 7% inthe Ramsey sandstone and ranges from 0 to 30%(Figure 3A, B). Calcite is not uniformly distrib-uted but occurs as concretions, defined byMcBrideet al. (1995, p. 1044) as ‘‘discrete and tightly ce-mented bodies embedded in weakly cemented oruncemented host sandstones.’’ Calcite-cementedzones range from 2 to 16 in. (5 to 40 cm) in thick-ness. Cemented zones contain an average of 18.5%authigenic calcite, and the ‘‘uncemented’’ sand-stones contain an average of 3.6% authigenic calcite.

Most of the calcite fills intergranular pores, butan average of 1.5% calcite occurs as grain replace-ment. Some feldspar grains are only partly replacedby calcite (Figure 3C), and it is clear that partialgrain replacement has occurred. In other cases, re-placement has been complete, and the original de-trital grain cannot be identified. These sand-grain-size areas of large, single crystals of clear calciteare interpreted to be replaced grains, probably re-placed feldspars (Figure 3D). An alternate inter-pretation is that they are detrital grains of coarselycrystalline sparry limestone or fossils. However, itis unlikely that the limestone source area containedsuch large calcite crystals (R. G. Loucks, 2005, per-sonal communication). These grains are not echi-noid fragments because echinoid fragments inthese sandstones can be distinguished by their gran-ular microtexture (‘‘small pores filled with ‘dirt’’’;Scholle, 1978, p. 95).Microprobe analysis of calcitecement in the Delaware Mountain Group sand-stones (mainly Bell Canyon sandstones) by Haysand Tieh (1992) indicates a mean composition ofCa0.961Mg0.013Fe0.003Mn0.023.

Chlorite (average volume = 1%) forms rimsaround detrital grains, extending into pores and porethroats. Other authigenic minerals include perva-sive but volumetrically minor authigenic quartzand feldspar (both K-feldspar and Na-plagioclase)overgrowths (average <1%) and local anhydrite ce-ment. Quartz generally appears to have overgrownand included chlorite.

Dutton 769

Porositydeterminedby thin-sectionpoint countsaverages 20% (Figure 3A), consisting of 18% pri-mary porosity and 2% secondary porosity. Sec-ondary pores are interpreted as forming mainly byfeldspar dissolution; some partly dissolved feldsparswere observed. Core-analysis porosity in the Ram-sey sandstones from the Geraldine Ford and EastFord fields averages 22.0% and ranges from 2.6 to36.1%. Sandstone permeability ranges from 0.01 to408 md. Average permeability is 39 md, and geo-metric mean permeability is 17 md.

The Trap and Ford siltstones, which overlieand underlie the Ramsey sandstone, respectively

(Figure 2), have an average grain size of 0.058 mm(4.1 f) and an average composition of Q68F26R6.Detrital fossils and carbonate rock fragments inthe siltstones are about as abundant as in the sur-rounding sandstones, having a combined averagevolume of 1%. The average volume of authigeniccalcite in the siltstones is 2.4%, ranging from 0 to7%. The DelawareMountain Group siltstones con-tain an average of 3% total organic carbon (TOC),and TOC values range from 0.5 to 12% (Hays andTieh, 1992).

On the basis of petrographic evidence, the rela-tive order of major events in the diagenetic history

Figure 3. Petrographic photos. (A) Porous Ramsey 1 sandstone from 41R East Ford unit well, 2766.2 ft (843.1 m). (B) Ramsey 1sandstone completely cemented by calcite from 41R East Ford unit, 2766.8 ft (843.3 m). (C) Potassium feldspar grain partly replacedby calcite in Ramsey 1 sandstone from 94 Ford Geraldine unit, 2671 ft (814.1 m). In photos A–C, porosity is stained blue; C = calcite,K = potassium feldspar. (D) Cathodoluminescence photo of calcite cement (orange) in Ramsey 1 sandstone from 41R East Ford unit,2766.8 ft (843.3 m). The large area of calcite cement labeled RF is interpreted to be a feldspar grain that was replaced by calcite.Photo by R. M. Reed.


of the Ramsey sandstones is interpreted to be(1) precipitation of chlorite rims, (2) formation of

quartz overgrowths, and (3) precipitation of cal-cite cement. The same basic paragenetic sequencewas observed by Williamson (1978) in the BellCanyon sandstones in nearby fields. Calcite cementhas a complex relationship with chlorite, appear-ing both to include and be overgrown by clay flakes;this relationship suggests that some chlorite pre-cipitated after calcite.

Hays and Tieh (1992) emphasized the dissolu-tion of early calcite cement as a major event in thecreation and preservation of porosity in the Dela-ware sandstones. However, examination of sam-ples from the East Ford and Geraldine Ford fieldsfor this study suggests that although some calcitecement dissolution occurred, dissolution was notwidespread or extensive.

Isotopic Composition of Calcite Cement

Isotopic composition of calcite cement in the EastFord field was measured in eight samples. All data

are reported as per-mil deviation from the Peedeebelemnite (PDB) standard. The d13C compositionof the calcite has a very narrow range from �1.82to �2.95x (PDB); d18Ocomposition ranges from�4.55 to �6.28x (PDB) (Table 1). Most of thesamples were taken from highly cemented sand-stones (12 to 30% authigenic calcite), but one sam-ple contained 2% detrital carbonate rock frag-

ments and no authigenic calcite, as determined bypoint counts (error <1% [Folk, 1974]). The samplewithout calcite cement has a d13C compositionof �1.82x (PDB) and a d18O composition of�6.15x (PDB). None of the samples containedfossil fragments, as determined by point counts. Iso-topic measurements of calcite cement from othernearby oil fields have a d13C composition rangingfrom �1.1 to �2.9x (PDB) and a d18O compo-sition from �4.5 to �7.8x (PDB) (Williamson,1978).

Effect of Calcite Cement on Porosity and Permeability

In these sandstones that have little variation ingrain size and contain no detrital clay, the volumeof calcite cement is the most important control onpermeability. Statistically significant inverse rela-tionships exist between the volume of calcite ce-ment and both porosity and permeability in the

Ramsey sandstone (Dutton and Flanders, 2001b)(Figure 4). In the calcite-cemented zones (definedas samples that have greater than 10% authigeniccalcite), average permeability is 8.3md, geometric

Table 1. Isotopic Composition of Calcite Cement (relative to

PDB) in Ramsey Sandstone*

Depth (ft) d13C (x) d18O (x)

Authigenic

Calcite (%)

Detrital

CRF (%)

2738.1 �2.94 �4.55 26.5 0.5

2753.3 �2.70 �5.79 11.5 0.5

2757.3 �2.83 �5.77 12.0 0.5

2757.4 �2.95 �5.61 21.0 1.0

2760.8 �2.86 �5.96 12.0 0.5

2763.2 �1.82 �6.15 0 2.0

2766.8 �2.66 �4.91 30.0 1.5

2767.2 �2.95 �6.28 24.0 0.5

*All samples are from the 41R EFU well. Calcite and carbonate rock fragment(CRF) volume are from thin-section point counts.

Figure 4. Calcite cement volume is the main control onpermeability in Ramsey sandstones from the 41R East Ford unitwell (from Dutton and Flanders, 2001b; reprinted by permissionof the AAPG Southwest Section). Permeability is unstressed per-meability to air.

Dutton 771

mean permeability is 0.9 md, and average core-analysis porosity is 13.7%. Sandstones containingless than 10% authigenic calcite have an averagepermeability of 54 md, a geometric mean perme-ability of 33 md, and an average core-analysis po-rosity of 22.1%.

The calcite-cemented zones have sharpmarginsat the top and bottom (Figure 5). Closely spacedpermeability measurements were taken on theslabbed core face from the 41R EFU well at ap-proximately 1-in. (2.5-cm) intervals above, within,

and below a 5-in.-thick (13-cm-thick) cementedzone at 2757 ft (840 m) (Figures 5, 6) using adevice thatmeasures permeability by an unsteady-state pulse-decay method. The lowest permeabil-ity, and presumably the highest volume of calcitecement, occurs from 2757.5 to 2757.6 ft (840.49to 840.52 m) (Figure 6). Permeability increasesslightly in the 2 in. (5 cm) above and below thisdepth (Figure 6), but the absence of oil fluores-cence in the entire 5-in.-thick (13-cm-thick) zone(Figure 5) suggests that most porosity is occluded

Figure 5. Core of Ramsey 1 sandstone from the 41R East Ford unit well photographed in ultraviolet light. The cemented zonesappear black in ultraviolet light because they have no oil-filled porosity, whereas uncemented sandstones have bright-yellowfluorescence because the pores are saturated with oil. Depths are marked in feet.


by calcite cement. Moderate fluorescence in 1-in.-wide (2.5-cm-wide) zones above and below thecompletely cemented layer (Figure 5) suggests thatthere is a thin transition zone in which porosityis only partly filled by calcite cement. Thin sectionsfrom depths of 2757.3 ft (840.43m) and 2757.4 ft(840.46 m) contain remnants of intergranular cal-cite cement that showevidence of havingundergonedissolution. The gradual increase in permeability atthe top of the cemented zone (Figure 6) may becaused by partial dissolution of calcite cement, andthe upper boundary may have been sharper in thepast. Carbonate-cemented zones at theNorth ColesLevee field in California have very sharp boundaries(Boles and Ramseyer, 1987); the transition from to-tally cemented to porous sandstone occurs within0.5 in. (13 mm) or less.

Anexcellent relationship betweenporosity andpermeability exists in sandstones from the 41REFU well (Figure 7). Bryant et al. (1993a, b) de-veloped a model for calculating permeability fromgrain size and porosity in simple granular porousmedia containing equal-thickness cement. To testthe applicability of the model to the Ramsey sand-stones, permeability values measured on core plugs

from the 41REFUwell were comparedwith perme-ability calculated from the Bryant-Finney modelusing the modified power-law equation of Jenningsand Lucia (2003):

lnðkÞ ¼ lnðaÞ þ b lnðf� f0Þ þ 2 lnðrÞ

where a, b, and f0 are constants having values of85.9, 3.39, and 0.0221, respectively, and k, f, and rare given in units of millidarcys, fraction of bulkvolume, and micrometers, respectively.

Measured permeability in these sandstones isabout1.5 orders ofmagnitude lower thanpredicted

permeability calculated by the Bryant-Finneymodel(Figure 7). Actual permeabilitymay be lower thanpredicted permeability because of the presence ofchlorite rims around detrital grains. Although theyhave little effect on porosity, chlorite rims extendinto pore throats and reduce permeability. In ad-dition, calcite cement tends to completely fill poreswhere it occurs and be absent from other pores, soit is not an equal-thickness cement. The Bryant-Finney model correctly predicts the shape of the

Figure 6. Plot of permeability versus depth across a calcite-cemented layer in the 41R East Ford unit well (from Dutton andFlanders, 2001b; used with permission from the AAPG SouthwestSection). Permeability was measured at 1-in. (2.5-cm) intervalsdirectly on the slabbed core face using a device that measurespermeability by an unsteady-state pulse-decay method. Perme-ability has been mathematically converted from air-permeabilitymeasurements to equivalent liquid-permeability values, or Klin-kenberg permeability. Core interval is shown in Figure 5.

Figure 7. Plot of measured porosity versus permeability fromcore analysis data from the 41R East Ford unit well. Predictedpermeability was calculated from the Bryant-Finney model(Bryant et al., 1993a, b) using the modified power-law equationof Jennings and Lucia (2003). See text for details.

Dutton 773

porosity-permeability curve in the 41R EFU sand-stones but overestimates permeability.

Calcite Cement Distribution in Geraldine Ford Field

Calcite-cemented zones can occur anywhere withinthe Ramsey sandstone section in the Geraldine Fordfield, but they are somewhat more common nearthe top and base of the Ramsey interval (Figure 8)(Dutton et al., 1999). It was not possible to deter-mine the lateral extent of the cemented intervalsobserved in the core or to correlate any of the ce-mented zones between wells with certainty. Mostof these zones of calcite-cemented sandstone areassumed to be concretions that extend no morethan a few meters laterally. Although concretionsof this size are not barriers, they can reduce av-erage reservoir permeability,make flowpathsmoretortuous, change breakthrough time and location,andmodify sweep efficiency in a reservoir (Duttonet al., 2002). Cemented zones were not observedin cores of the overlying Trap siltstone or in theunderlying Ford siltstone.

Median thickness of the 234 calcite-cementedzones in 70 cores in theGeraldine Ford field is 0.6 ft(0.2 m), and 75% of the cemented zones are lessthan 1 ft (0.3 m) thick. An average of 6% of thetotal reservoir interval in the Geraldine Ford fieldis tightly cemented by calcite (206 ft [62.8 m] ofcalcite-cemented sandstone/3615 ft [1101.9m] ofsandstone core). The total volume of authigeniccalcitewithin the reservoir is estimated to be about4.4%, assuming18.5%calcite in the cemented zonesand 3.6% in the ‘‘uncemented’’ zones.

Geographic distribution of calcite cement in theGeraldine Ford field was mapped by calculatingthe percentage of calcite-cemented sandstone ineach cored well. Most of the areas that have a highpercentage (>15%) of calcite-cemented sandstoneare found along the margins of the field, where thesandstone pinches out into siltstone (Figure 9). To-

tal thickness of calcite-cemented sandstone zonesalso shows a similar distribution, with the thickestcemented zones at the field margins. The trendshown in Figure 9 is thus not simply an artifactof thinning of the Ramsey sandstone toward the

pinch-out into siltstone. Calcite cement is mostcommon in what are interpreted to be overbank-splay and lobe facies at the margins of the sand-stone body, and less abundant in the channel faciesin the center (Figure 9).

The presence of thin layers of tightly calcite-cemented sandstone (Figure 8) results in highlyvariable, streakypermeability distribution through-out the Ramsey sandstone (Dutton et al., 1999).Calcite-cemented zones are common near the topof the Ramsey sandstone (�1 ft [0.3 m] below thesandstone-siltstone contact), but in some wells, un-usually high-permeability zones (>100md) occurabove these calcite layers, at the top of the sand-stone. Because samples were not available fromthese high-permeability zones, the reason for thehigh permeability— greater than normal feldspardissolution, calcite cement dissolution, or grain-sizevariation, for example— could not be determined.Aswill bediscussedbelow,extensive feldspardissolu-tion where acidic formation fluids entered the sand-stones may explain the high-permeability zones.

Calcite Cement Distribution in East Ford Field

Distribution of calcite cement in the East Ford fieldwas interpreted from geophysical logs taken in20 wells that were calibrated to a core taken in the41R EFU well (Figures 5, 10). In this core, four cal-cite layers in the lower part of the Ramsey 1 sand-stone are spaced about3 ft (1m)apart.TheRamsey2sandstone contains one cemented layer that is 14 in.(36 cm) thick (Figure 10). Calcite-cemented zoneswere mapped throughout the East Ford field usingcore analysis data and sonic and resistivity logs toidentify cemented, low-permeability intervals. Thistechnique can be used because calcite is the pre-dominant control on permeability in these sand-stones (Figure 4). The cemented thicknesses maybe somewhat overestimated because cementedzones appear thicker on the logs than they actually

are (Figure 10) because of log resolution.

Vertical DistributionIn the 41R EFU well, calcite-cemented zones oc-cur near the top and base of the sandstone body,


Figure 8. West–east cross section AA0 of the north end of Geraldine Ford field (modified from Dutton et al., 1999). Calcite-cemented intervals were identified in cores. Location ofcross section is shown in Figure 9. GR = gamma ray.

Dutton

775

Figure 9. Map of the percentage of the Ramsey sandstone interval (both Ramsey 1 and Ramsey 2) that is cemented by calcite inGeraldine Ford field (modified from Dutton et al., 1999). Calcite-cemented zones were identified in cores from the 73 wells shown onthis map. Sandstone pinch-out was identified using geophysical logs from 305 wells in the field (not all shown). See Dutton et al.(1999) for the interpretation of sandstone facies.


that is, at the base of the Ramsey 1 sandstone andat the top of the Ramsey 2 sandstone (Figure 10).These two zones contain �21% intergranular cal-cite cement and showgreatest deflection in neutronand density curves (Figure 10). Other cementedlayers in the Ramsey 1 sandstone contain lowervolumes of calcite cement (8.5 to 15%) and havesmaller log deflections.

Similar calcite-cemented zones are commonnear the top and base of the Ramsey sandstone

throughout the East Ford field. A 1- to 2-ft-thick(0.3- to 0.6-m-thick) calcite-cemented zone wasobserved in most wells just below the top of theRamsey 2 sandstone (labeled 1 on Figure 11), andanother cemented zone occurs just above the baseof the Ramsey 1 sandstone (labeled 7 on Figure 11).In most wells, including the 41R EFU well, theselayers are not right at the sandstone-siltstone con-tact, but about 6 in. (15 cm) into the sandstone(Figure 10).

Figure 10. Gamma ray, neu-tron, and density logs fromthe Ramsey producing intervalcored in the 41R East Ford unitwell. Calcite-cemented sand-stone intervals were identifiedin the core. Production andtemperature logs indicated thatthe gas production essentiallyall occurred from the intervallabeled ‘‘Zone of gas effect’’(modified from Dutton andFlanders, 2004). SH1 = lami-nated siltstone.

Figure 11. West–east cross section BB0 of the south end of East Ford field (EFU) (modified from Dutton and Flanders, 2004).Four calcite-cemented layers (labeled 4, 5, 6, and 7) in the lower Ramsey 1 sandstone can be correlated in the 40, 41, and 41R EFU wells.The layer at the base of the Ramsey 1 sandstone (labeled 7) and another near the top of the Ramsey 2 sandstone (labeled 1) occurin all five wells. Location of cross section is shown in Figure 12.

Dutton 777

Lateral Distribution and ExtentMaps of the percentage of calcite-cementedsandstone in the Ramsey 1 and Ramsey 2 intervals(Figures 12, 13, respectively) show variations in ce-ment volume across the East Ford field. In general,the percentage of calcite-cemented sandstone islower in the Ramsey 1 than in the Ramsey 2 sand-stone. In both sandstones, areas that have higherpercentages of calcite-cemented sandstone (>20%)tend to occur along the margins of the sandstonebody, in overbank splay and lobedeposits.Areas thathave the lowest percentage of calcite-cemented sand-stone (<10%) occur where sandstone is thickest,in the interpreted channel facies.

Inmany field studies, the lateral extent of calcite-cemented zones cannot bedetermined fromwidelyspaced subsurface well data. Cemented zones thatappear to correlate between wells may not be con-tinuous layers but instead are discontinuous, strata-bound concretions. Production data from the EastFord field, however, indicate that some calcite inthe south part of the field occurs in laterally con-tinuous layers and not as isolated concretions. The41R EFU well was drilled 5 yr after the begin-ning of CO2 flooding in the EFU as an offset tothe 41 EFU well (Figures 11, 12). Sonic and neu-tron logs from the 41REFUwell showed a gas effectin the lower 8 to 10 ft (2.4 to 3 m) of the Ramsey 1sandstone, in the same interval where the calcite-cemented layers occur (Figure 10). When the wellwas first completed, it produced a high gas vol-ume (750 mcf gas/day) that contained a high con-centration of CO2 (>90%) (Dutton and Flanders,2004). Production and temperature logs confirmedthe gas effect by indicating that inflow to the wellbore was essentially all occurring in the bottom10 ft (3 m) of the Ramsey 1 sandstone. The CO2,most likely derived from the 40EFU injectorwell,was trapped below the calcite layers. TheCO2wasinjected in the 40 EFUwell both above and belowthe calcite-cemented layers, but producing 39 and41 EFU wells had no perforations in the Ramsey 1sandstone below the calcite layers. The CO2 in-jected below the calcite-cemented layers in the40 EFU well became trapped, resulting in signifi-cantly higher reservoir pressure. TheCO2 that wasproduced in the 41R EFU well probably repre-

sents banked-up energy that gave a first flush ofCO2 when the well was completed (Dutton andFlanders, 2004).Thesedatademonstrate that calcite-cemented layers can cause effective horizontal per-meability barriers within a field. Calcite-cementedzones identified on logs cannot be assumed to besmall, discontinuous concretions; perforations aboveand below calcite layers may be necessary to con-tact the entire reservoir volume.

The trapping ofCO2 suggests that one ormoreof the calcite layers is laterally continuous between40 EFU and 41R EFU. Spikes on the 41R EFUsonic log in the lower part of the Ramsey 1 sand-stone appear to correlate to those on the 40 EFUand 41 EFU sonic logs (Figure 11), further evidencesuggesting the lateral continuity of the cement lay-ers. Because the distance between 40 EFU well and41 and 41R EFUwells is about 1000 ft (300m), thefour calcite layers observed in the 41R EFU coreare interpreted as having a lateral extent of at leastthat distance. Most of these layers apparentlydo not extend to the 39 EFU and 38 EFU wells(Figure 11). However, cemented zones near thebase of the Ramsey 1 sandstone and near the topof the Ramsey 2 sandstone both appear to be con-tinuous across the entire south part of the field.The 24 EFU and 28 EFU wells (Figure 12) containthree calcite-cemented layers in the bottom of theRamsey 1 sandstone that are interpreted as corre-lating between the two wells, a distance of about735 ft (225 m).

DISCUSSION

Several studies of calcite cement in shallow-marineand deep-water sandstones have concluded that thesource of the calcium carbonate was internal tothe sandstones, and distribution of cement reflectsthe original distribution of skeletal debris (see ref-erences summarized in Bjørkum and Walderhaug,1990;Hendry et al., 1996;Morad, 1998;Walderhaugand Bjørkum, 1998). Other studies of calcite ce-ment in fluvial, deltaic, and deep-water settings haveconcluded that internal calcite sources were lack-ing, and calcium carbonate was transported intothe sandstones from outside (e.g., Sullivan and


Figure 12. Map of the percentage of theRamsey 1 sandstone in East Ford field that iscemented by calcite. Only wells showing valuesfor calcite percent penetrated the entire Ramsey 1interval with sonic or resistivity logs, or both.Other wells in the field were used to map sand thick-ness. Cross section BB0 is shown in Figure 11.

Dutton 779

Figure 13. Map of the percentage of theRamsey 2 sandstone in East Ford field thatis cemented by calcite. Only wells showing val-ues for calcite percent penetrated the entireRamsey 2 interval with sonic or resistivity logs,or both. Other wells in the field were usedto map sand thickness.


McBride, 1991;Moraes and Surdam, 1993; McBrideet al., 1995; Anjos et al., 2000; Dutton et al., 2000;McBride and Milliken, 2006).

Calcite cement in the East Ford and GeraldineFord fields is most abundant at the margins of theRamsey sandstone bodies, at the top, base, andsides of the sandstones where they pinch out intosiltstone, although calcite-cemented intervals occurthroughout the sandstone bodies (Figures 8, 11).Distribution of calcite cement in the Ramsey sand-stone bodies probably reflects the source of cal-ciumcarbonate, but twodifferent hypotheses couldexplain this cementation pattern: (1) the cementdistribution reflects the original distribution of de-trital carbonates and skeletal material within thesandstone, which acted as the source of calcium car-bonate, or (2) calcium carbonate was imported fromoutside the sandstone, and calcite preferentiallyprecipitated near the margins of the sand bodies,particularly in zones of higher permeability.

Internal Versus External Source of Calcite Cement

Isotopic data suggest that biogenic carbonate wasthe source of calcite cement. The d13C composi-tion of the calcite cement in the East Ford field,�1.82 to �2.95x (PDB) (Table 1), indicates thatthe major source of carbon was skeletal carbonatedebris (Gross, 1964), with aminor contribution of13C-depleted carbon derived from oxidation ordecarboxylation of organic matter (Curtis, 1978).The location of the skeletal debris that sourced thecement, however, whether inside or outside thesandstone, is unclear. The Ramsey sandstones cur-rently contain only small volumes of detrital calciumcarbonate as both calcite fossil fragments (mainlyechinoid fragments, fusulinid tests, and molluskfragments) and carbonate rock fragments (coarselycrystalline limestone and micrite). The averagevolume of carbonate rock fragments currently inthe sandstones is 1%, and the average volume offossil fragments is less than 0.1%. Sandstones alsocontain an average of 2% secondary pores.

Thin sections taken at 1-ft (0.3-m) intervalsfrom the Ramsey sandstone in the 41R EFU corecontain an average volume of 7% authigenic cal-cite, consisting of 6% primary-pore-filling cement

and 1% grain replacement. Even if all the second-ary pores were originally calcium carbonate thatdissolved and reprecipitated as calcite cement,the total internal source material does not ap-pear to be sufficient to account for the volume ofauthigenic calcite in the sandstones. Furthermore,partly dissolved feldspars indicate that much ofthe secondary porosity was probably derived fromfeldspar dissolution and not dissolution of skeletaldebris.

It is unknown, however, how much calciumcarbonatemay have beenpresent in the sandstonesat the time of deposition but was subsequentlycompletely dissolved, leaving only calcite cementto indicate its former presence. If dissolution hap-pened relatively early in the burial history, grainrearrangement and sediment compaction couldhavedestroyed any secondary pores that developed. Noaragonitic fossil debris is present in the sandstonesnow, although fragments of sponges, scleractiniancorals, and phylloid algae could have been derivedfrom the carbonate shelf. These basin-floor tur-bidites were deposited in water depths of 300 to600m (Kerans et al., 1992), so it is unlikely that theRamsey sandstoneswere exposed tometeoricwaterafter deposition, even during periods of sea-levellowstand.As a result, any aragonite grains thatweredeposited probably avoided early dissolution andwere carried into the subsurface as the Ramsey sand-stones were buried by younger deposits. Aragonitegrains may have been unstable in the subsurfaceBell Canyon pore fluids, which have high Ca/Mgratios (McNeal and Mooney, 1968; Williamson,1978). Therefore, if aragonite grainswere originallypresent in the Ramsey sandstones, their dissolutioncould have been a source of calcite cement.

Although the former presence of sufficient ara-gonite skeletal debris to account for the volume ofcalcite cement in these sandstones cannot be ruledout, there is no evidence to support this interpre-tation. Williamson (1978) noted that carbonaterock fragments and fossils are more abundant inoutcropping Ramsey sandstones, which are closeto the shelf margin, than in subsurface reservoirsandstones farther out in the basin. In the basin-floor position of the East Ford and Geraldine Fordfields, it is unlikely that carbonate debris from the

Dutton 781

surrounding shelves would have been significantlymore abundant than currently observed. Further-more, the small volume of calcite fossils in the sand-stones suggests that aragonitic skeletal debris wasprobably not abundant either. Even if there had

originally been an equal volume of aragonite fossilfragments like the calcite fossils that are still pre-sent, itwould account for less than 1%of additionalinternal volumeof calciumcarbonate.Hendry et al.(1996) interpreted aragonitic fossil debris to be thesource of calcite cement in Cretaceous turbiditesin theNorth Sea, but those sandstones still containevidence of abundant bioclasts, including remain-ing calcite bioclasts and micritized outlines offormer aragonitic skeletal fragments.

Themost likely interpretation, therefore, is thatthe source of calcium carbonate for calcite cementin theRamsey sandstoneswas external, coming fromorganic-rich siltstones or limestones in the Bell Can-yon Formation or other carbonate units in the Dela-ware Basin.

Depth and Timing of Calcite Cementation

Isotopic data and textural observations from thinsections were used to interpret conditions of calcitecementation. Textural data indicate that the cal-cite precipitated at or nearmaximumburial depth.The average intergranular volume (IGV) of calcite-cemented sandstones is 26% (range of 23 to 29%),essentially the same as the 27% IGV of poorly ce-mented sandstones (range of 22 to 33%). The factthat the Ramsey sandstones have been compactedto an IGV of 26% suggests that they reachedmaxi-mum burial of approximately 4800 ft (1.5 km)prior to calcite cementation (Paxton et al., 2002).Precipitation of minor quartz cement prior to cal-cite also indicates that calcite cementation did notoccur shortly after deposition.

The d18O of the calcite cement ranges from�4.55 to �6.28x (PDB) (Table 1; Figure 14).The isotopic data do not provide a unique solutionwith which to interpret conditions at the time ofcementation; at least two different scenarios of tem-perature andwater composition are consistentwiththe isotopic data.

The calcite cement is in equilibrium with thepresent water composition of �2x (SMOW) andtemperature of 82jF (28jC) in the Ramsey sand-stoneatSullivan field (Williamson,1978) (Figure14).Williamson (1978) interpreted calcite cement inthe Ramsey sandstone as having precipitated fromwater similar in isotopic composition and tempera-ture to that of the sodiumchloride formation brinespresent in the Bell Canyon Formation today. Thisinterpretation implies that the calcite precipitatedafter the late Tertiary tilting of theDelaware Basin

and the development of the current hydrodynamicsetting at 5 to 10 Ma (McNeal, 1965; Bein andDutton, 1993).

Alternatively, the calcite could have precipitatedfrom Permian seawater at a temperature of 104jF(40jC), assuming that the seawater compositionwas approximately 0x (SMOW) (Land andLynch,1996). In this scenario, precipitation may have oc-curred from formationwaters during theLate Perm-ian and Mesozoic burial of the sediments. If thecomposition of Permian seawater in the DelawareBasin was isotopically depleted compared with that

Figure 14. Possible water temperatures and d18O composi-tions that could have precipitated Ramsey sandstone calcitecement. The equation relating temperature, d18O-water, andd18O-mineral is 103 ln a = 2.78 � 106 T � 2 � 2.89 (O’Neilet al., 1969), where a is the fractionation factor and T is temper-ature. Present water composition and temperature are shownfor Ramsey sandstone in Sullivan field (data from Williamson,1978), located 2mi (3 km) south of East Ford field. PDB = Peedeebelemnite, SMOW = standard mean ocean water.


of today’s oceans and had d18O of �1.4 to �3.0x(SMOW) (Given andLohmann, 1985; Korte et al.,2005), the calcite could have precipitated at lowertemperatures of 77 to 95jF (25 to 35jC).

The second scenario is more likely because ofthe timing of oilmaturation andmigration into thefields. Tightly calcite-cemented intervals in the 41REFU sandstone (Figure 5) show no hydrocarbonfluorescence, suggesting that the calcite cementprecipitated before oilmigrated into the reservoirs.Hays and Tieh (1992) calculated that the upperDelawareMountainGroup began generating oil inthemiddle Eocene (�50Ma).Organicmaturationand generation of organic acidswould have occurredwell before the late Tertiary uplift and tilting of theDelaware Basin, so it is unlikely that the calciteprecipitated in the current hydrodynamic setting.Calcite precipitation is interpreted as occurring nearmaximumburial depth during theMesozoic, priorto oil generation and migration in the middle Eo-cene (�50 Ma).

Precipitation of Calcite Cement

Calcite cement is located preferentially near themargins of the sandstone bodies in both the EastFord and Geraldine Ford fields. Because internalsources of calcite were apparently insufficient toaccount for the volume of cement in the sand-stones, necessary components for the cement areinterpreted as having been transported in from theoutside. Field-scale maps of cement distribution(Figures 9, 12, 13) suggest that considerable cal-cite precipitated near the point of entry into thesandstone bodies.

Skeletal carbonate debris and carbonate rockfragments in organic-rich basinal siltstones and lime-stones are the most likely external sources of neces-sary components for the cement. Organic acidsgenerated during thermal maturation of organicmatter probably provided a source of acid for car-bonate dissolution and export into the sandstones.Organic-rich siltstones in the Delaware MountainGroup contain an average of 3% TOC, ranging upto 12% TOC (Hays and Tieh, 1992). Hays andTieh (1992) determined that the organic matterwas oxygen-rich type II and III kerogen that could

have generated a significant volumeof organic acid.In many formations, the volume of organic acidsthat can be generated cannot account for the ob-served volumeof secondary pores (Lundegard et al.,1984; Lundegard and Land, 1986), but in theDela-ware Basin, the organic-rich siltstones and lime-stones may have generated organic acid sufficientto keep the dissolved carbonate in solution andtransport it out of the siltstones and limestonesand into sandstones.When calcium carbonate wastransported into the sandstones, abundant reactivefeldspar buffered the acid, allowing calcite to pre-cipitate (Milliken and Land, 1991, 1993).

Relatively 13C-enriched carbon in the Ramseysandstone calcite cement (�1.82 to �2.95x [PDB])may result from derivation of carbon primarily fromCO�2

3 from dissolved inorganic carbonate rock frag-ments and skeletal debris and not from organic mat-ter (K. L.Milliken, 2006, personal communication).If protons released by dissociation of organic acidsdissolve detrital carbonates, the carbon isotopic com-position of the resulting inorganic carbon pool willbe determined by dissolving carbonate minerals, solong as the acetate ion is not destroyed (Lundegardet al., 1984, p. 402).

Controls on Calcite Distribution

The greater volume of calcite cement near the mar-gins of the Ramsey sandstone bodies may reflectthe abundance of feldspars in these sandstones.The feldspars effectively bufferedmuch of the acid-ic formation waters near where they entered thesandstone, resulting in calcite precipitation nearthe sandstone-siltstonemargins. Zones of high per-meability at the top of the Ramsey sandstone mayresult from extensive feldspar dissolution whereacidic formation waters went into the sandstone.Occurrence of calcite cement layers near the topand base of the Ramsey sandstone in the East Fordfield (labeled 1 and 7 in Figure 11) fits this model.Furthermore, layers at the top and base of the sand-

stone body have higher volumes of calcite cementthan do cemented zones that are farther from thesandstone-siltstone contact (Figure 15). Turbiditesandstones of the Pendencia Formation, Brazil, alsoshowperipheral calcite cementation at the contact

Dutton 783

with interbedded shales thatwere themajor sourceof dissolved carbonate (Anjos et al., 2000).

Other laterally extensive calcite cement layersin the lower part of the Ramsey 1 sandstone in theEast Ford field (labeled 4, 5, and 6 in Figure 11) areharder to explain by this mechanism. These layersdonotoccur at themargin of the sandstonebodybutare located about 0.3mi (0.5 km) from the southernpinch-out of the Ramsey 1 sandstone (Figure 12).They are in the lower part of the Ramsey 1 sand-stone (Figures 10, 11) but not at the base. Each

layer is spaced about 3 ft (1 m) apart from theother vertically, similar to the 3-ft (1-m) verticalspacing of calcite-cemented zones in turbidites ofthe Namorado Sandstone in Brazil (Sombra et al.,1998). The lateral extent and continuity of thesecalcite-cemented zones suggest that they are local-ized by a stratigraphic control that is also laterallyextensive, such as coarser-than-average sedimentlayers or within certain zones of the Bouma se-quence. One hypothesis to explain the location ofthese laterally extensive calcite layers is that theyare located at the base of turbidite pulses, in thecoarsest grained sandstones. Another possibility isthat the shape and hydrodynamics of aragoniteskeletal grainsmay have concentrated them in over-bank and splay facies. Recognizing sedimentary

structures in the Ramsey sandstones is difficultbecause they are very well sorted and contain nodetrital clay (Figure 5), but a plot of grain size ver-sus depth in the 41R EFU core defines the coarsestgrained sandstones and upward-fining packages(Figure 15). The calcite layers tend to be locatedat, or somewhat above, coarser grained layers, nearthe base of upward-fining packages. These zonesmay have had slightly higher initial porosity andpermeability and, thus, transmitted larger volumesof formation water.

However, no statistically significant correla-tion between calcite cement and grain size is seen.The interpretation that calcite-cemented layers co-incide with the base of turbidite deposits is thusuncertain. Unlike the channel lags in the BrushyCanyon sandstone (Beaubouef et al., 1999), thesecemented layers in the Ramsey sandstone do notcontain increased volumes of bioclastic debris. Noevidence is proven that these layers originally con-tained abundant aragonitic material that was thesource of the calcite, but neither can it be disproved.Coarser grained intervals of the Pendencia For-mation, Brazil, are also preferentially cemented bycalcite (Anjos et al., 2000). The coarser and finergrained Pendencia sandstones have equivalent de-trital compositions, suggesting that selective advec-tion in the coarser laminae carried dissolved car-bonate into zones of originally higher permeability(Anjos et al., 2000).

The prediction of calcite cement distributionin sandstone reservoirs remains an elusive goal. Insandstones where the source of calcite cement isinterpreted to have been from the outside, cementmay preferentially precipitate near sandstonemar-gins, as it did in the Ramsey sandstone. However,precipitation near the immediate point of entryinto the sandstone bodies may occur only wherefeldspar is sufficiently abundant to buffer most ofthe acidic formationwaters nearwhere they enteredthe sandstone, allowing calcite to precipitate there(Milliken and Land, 1991, 1993). Distribution ofother calcite-cemented zones that are not at themargins of the sandstonebodiesmay relate to grain-size variations that exert a control on fluid flow(Sullivan andMcBride, 1991;McBride et al., 1995;Dutton et al., 2000; McBride and Milliken, 2006).

Figure 15. Plot of calcite volume and grain size versus depthin the 41R East Ford unit core, East Ford field. An additionalcalcite-cemented layer at 2764 ft (842.4 m) is clearly visible onthe core photos (Figure 5) but was not sampled by a thinsection; this layer is indicated by the dashed line.


In sandstones where the source of calcite cementwas the dissolution of internal aragonite skeletalgrains, the prediction of cement distribution is trans-ferred to the prediction of distribution and con-centration of bioclasts.

CONCLUSIONS

Reservoir quality of deep-water turbidite depositsin the Upper Permian (Guadalupian) Bell CanyonFormation, Delaware Basin, is strongly controlledby distribution of calcite cement. In these sand-stones that have little variation in grain size andcontain no detrital clay, volume of calcite cementis the most important control on porosity and per-meability. The presence of thin layers of tightlycalcite-cemented sandstone results in highly var-iable, streaky permeability distribution through-out the Ramsey sandstone. Most areas that have ahigh percentage (>15%) of calcite-cemented sand-stone occur along the margins of sandstone bodies,where the sandstone pinches out into siltstone.Production data from awell at the south end of theEast Ford field indicate that at least some calcite inthis field occurs in laterally continuous layers andnot as isolated concretions. Four calcite layers areinterpreted as having a continuous lateral extent of

at least 1000 ft (300 m) because injected CO2 wastrapped below the layers. One or more of the thincalcite layers formed a permeability barrier withinthe reservoir, and perforations both above and be-low the cemented layer were required to contactthe entire reservoir volume.

Because internal sources of calcite were appar-ently insufficient to account for the volume of ce-ment in the sandstones, necessary components forthe cement are interpreted as having been trans-ported into the sandstone from outside. Skeletalcarbonate debris and carbonate rock fragments inorganic-rich basinal siltstones and limestones arethe most likely external sources of necessary com-ponents for the cement. Organic acids generatedduring thermalmaturation of organicmatter prob-ably provided a source of acid for carbonate dis-solution and export into the sandstones. Feldsparsin the sandstones effectively buffered much of the

acidic formation waters near where they enteredthe sandstone, resulting in calcite precipitation nearthe sandstone-siltstone margins. Other calcite-cemented layers may coincide with the base of tur-bidite deposits.

REFERENCES CITED

Anjos, S. M. C., L. F. De Ros, R. S. Souza, C. M. A. Silva,and C. L. Sombra, 2000, Depositional and diageneticcontrols on the reservoir quality of Lower CretaceousPendencia sandstones, Potiguar rift basin, Brazil: AAPGBulletin, v. 84, p. 1719–1742.

Barton, M. D., and S. P. Dutton, 1999, Outcrop analysisof a sand-rich, basin-floor turbidite system, Permian BellCanyon Formation, west Texas, in Transactions, GulfCoast Section SEPM 19th Annual Bob F. Perkins Re-search Conference, December 5–8, Houston, p. 53–64.

Barton, M. D., and S. P. Dutton, 2007, A sand-rich, basinfloor turbidite system, Willow Mountain, Texas, USA,in T. H. Nilsen, R. D. Shew, G. S. Steffens, and J. R. J.Studlick, eds., Atlas of deep-water outcrops: AAPG Stud-ies in Geology 56, p. 418–421.

Beaubouef, R. T., C. Rossen, F. B. Zelt, M. D. Sullivan, D. C.Mohrig, and D. C. Jennette, 1999, Deep-water sand-stones, Brushy Canyon Formation, west Texas: AAPGContinuing Education Course Notes Series 40, 48 p.

Bein, A., and A. R. Dutton, 1993, Origin, distribution, andmovement of brine in the Permian Basin (U.S.A.): Amod-el for displacement of connate brine: Geological Societyof America Bulletin, v. 105, p. 695–707.

Bjørkum, P. A., and O. Walderhaug, 1990, Geometrical ar-rangement of calcite cementation within shallowmarinesandstones: Earth-Science Reviews, v. 29, p. 145–161.

Boles, J. R., 1998, Carbonate cementation in Tertiary sand-stones, San Joaquin Basin, California, in S. Morad, ed.,Carbonate cementation in sandstones: International As-sociation of Sedimentologists Special Publication 26,p. 261–283.

Boles, J. R., and K. Ramseyer, 1987, Diagenetic carbonate inMiocene sandstone reservoir, San Joaquin Basin, Cali-fornia: AAPG Bulletin, v. 71, p. 1475–1487.

Brown,A.A., andR.G. Loucks, 1993a, Influence of sedimenttype and depositional processes on stratal patterns inthe Permian basin-margin Lamar Limestone, McKittrickCanyon, Texas, inR.G. Loucks andR. Sarg, eds., Carbon-ate sequence stratigraphy: Recent advances and applica-tions: AAPG Memoir 57, p. 133–156.

Brown,A.A., andR.G. Loucks, 1993b, Toe of slope, inD.G.Bebout and C. Kerans, eds., Guide to the Permian Reefgeology trail,McKittrickCanyon,GuadalupeMountainsNational Park, west Texas: University of Texas at Austin,Bureau of Economic Geology Guidebook 26, p. 5–13.

Bryant, S. C., C. A. Cade, and D. W. Mellor, 1993a, Perme-ability prediction from geological models: AAPG Bulle-tin, v. 77, p. 1338–1350.

Dutton 785

Bryant, S. C., D. W. Mellor, and C. A. Cade, 1993b, Phys-ically representative network models of transport in po-rous media: American Institute of Chemical EngineersJournal, v. 39, no. 3, p. 387–396.

Curtis, C. D., 1978, Possible links between sandstone dia-genesis and depth-related geochemical reactions occur-ring in enclosing mudstones: Geological Society (London)Journal, v. 135, p. 107–117.

Dutton, S. P., and W. A. Flanders, 2001a, Application ofadvanced reservoir characterization, simulation, and pro-duction optimization strategies to maximize recoveryin slope and basin clastic reservoirs, west Texas (Dela-ware Basin), Class III: University of Texas at Austin, Bu-reau of Economic Geology, final report prepared forU.S. Department of Energy, Assistant Secretary for Fos-sil Energy, under contract no. DE-FC22-95BC14936,170 p.

Dutton, S. P., and W. A. Flanders, 2001b, Deposition anddiagenesis of turbidite sandstones in East Ford field, BellCanyon Formation,DelawareBasin, Texas:AAPGSouth-west Section, Annual Meeting Papers and Abstracts,Dallas, March 10–13, 12 p.

Dutton, S. P., and W. A. Flanders, 2004, Evidence of reser-voir compartmentalization by calcite cement layers indeep-water sandstones, Bell Canyon Formation, Dela-ware Basin, Texas, in J. M. Cubitt, W. A. England, andS. Larter, eds., Understanding petroleum reservoirs: To-wards an integrated reservoir engineering and geochem-ical approach: Geological Society (London) Special Pub-lication 237, p. 279–282.

Dutton, S. P.,M.D.Barton,G. B.Asquith,M.A.Malik,A.G.Cole, J. Gogas, J. I. Guzman, and S. J. Clift, 1999, Geo-logic and engineering characterization of turbidite res-ervoirs, Ford Geraldine unit, Bell Canyon Formation,west Texas: University of Texas at Austin, Bureau ofEconomic Geology Report of Investigations 255, 88 p.

Dutton, S. P., B. J. Willis, C. D. White, and J. P. Bhatta-charya, 2000,Outcrop characterization of reservoir qual-ity and interwell-scale cement distribution in a tide-influenced delta, Frontier Formation, Wyoming, USA:Clay Minerals, v. 35, p. 95–105.

Dutton, S. P., C. D. White, B. J. Willis, and D. Novakovic,2002, Calcite cement distribution and its effect on fluidflow in a deltaic sandstone, Frontier Formation, Wyo-ming, USA: AAPG Bulletin, v. 86, p. 2007–2021.

Dutton, S. P., W. A. Flanders, and M. D. Barton, 2003,Reservoir characterization of a Permian deep-water sand-stone, East Ford field, Delaware Basin, Texas: AAPGBulletin, v. 87, p. 609–627.

Fischer, A. G., and M. Sarnthein, 1988, Airborne silts anddune-derived sands in the Permian of theDelawareBasin:Journal of Sedimentary Petrology, v. 58, p. 637–643.

Folk, R. L., 1974, Petrology of sedimentary rocks: Austin,Texas, Hemphill, 182 p.

Gardner, M. H., 1992, Sequence stratigraphy of eolian-derivedturbidites: Patterns of deep water sedimentation along anarid carbonate platform, Permian (Guadalupian) Dela-ware Mountain Group, west Texas, in D. H. Mruk andB. C. Curran, eds., Permian Basin exploration and pro-duction strategies: Applications of sequence stratigraphic

and reservoir characterization concepts: West Texas Geo-logical Society Publication 92–91, p. 7–12.

Given, R. K., and K. C. Lohmann, 1985, Derivation of theoriginal isotopic composition of Permian marine ce-ments: Journal of Sedimentary Petrology, v. 55, p. 430–439.

Gross, M. G., 1964, Variations in the O18/O16 and C13/C12

ratios of diagenetically altered limestones in the Ber-muda Islands: Journal of Geology, v. 72, p. 170–194.

Hays, P. D., and T. T. Tieh, 1992, Organic geochemistryand diagenesis of the Delaware Mountain Group westTexas and southeast New Mexico: AAPG SouthwestSection, Publication SWS 92–90, p. 155–175.

Hendry, J. P., N. H. Trewin, and A. E. Fallick, 1996, Low-Mg calcite marine cement in Cretaceous turbidites: Or-igin, spatial distribution and relationship to seawaterchemistry: Sedimentology, v. 43, p. 877–900.

Jennings Jr., J. W., and F. J. Lucia, 2003, Predicting per-meability from well logs in carbonates with a link togeology for interwell mapping: Society of Petroleum En-gineers Reservoir Evaluation and Engineering, v. 6, no. 4,p. 215–225.

Kerans, C., W. M. Fitchen, M. H. Gardner, M. D. Sonnen-feld, S. W. Tinker, and B. R. Wardlaw, 1992, Styles ofsequence development within uppermost Leonardianthrough Guadalupian strata of the Guadalupe Moun-tains, Texas and New Mexico, in D. H. Mruk and B. C.Curran, eds., Permian Basin exploration and productionstrategies: Applications of sequence stratigraphic and res-ervoir characterization concepts: West Texas GeologicalSociety Publication 92–91, p. 1–6.

Korte, C., T. Jasper, H. W. Kozur, and J. Veizer, 2005, d18Oand d13C of Permian brachiopods: A record of seawaterevolution and continental glaciation: Palaeogeography,Palaeoclimatology, and Palaeoecology, v. 224, p. 333–351.

Land, L. S., and F. L. Lynch, 1996, d18O values of mudrocks:More evidence for an 18O-buffered ocean: Geochimicaet Cosmochimica Acta, v. 60, p. 3347–3352.

Lundegard, P. D., and L. S. Land, 1986, Carbon dioxide andorganic acids: Their role in porosity enhancement andcementation, Paleogene of the Texas Gulf Coast, inD. L.Gautier, ed., Roles of organic matter in sediment diagen-esis: SEPM Special Publication 38, p. 129–146.

Lundegard, P. D., L. S. Land, and W. E. Galloway, 1984,Problem of secondary porosity: Frio Formation (Oligo-cene), Texas Gulf Coast: Geology, v. 12, p. 399–402.

Macaulay, C. I., A. E. Fallick, R. S. Haszeldine, and G. E.McAulay, 2000, Oil migration makes the difference:Regional distribution of carbonate cement d13C in north-ern North Sea Tertiary sandstones: Clay Minerals, v. 35,p. 69–76.

McBride, E. F., and K. L. Milliken, 2006, Giant calcite-cemented concretions, Dakota Formation, central Kansas,USA: Sedimentology, v. 53, p. 1161–1179.

McBride, E. F., K. L. Milliken, W. Cavazza, U. Cibin, D.Fontana, M. D. Picard, and G. G. Zuffa, 1995, Hetero-geneous distribution of calcite cement at the outcropscale in Tertiary sandstones, northern Apennines, Italy:AAPG Bulletin, v. 79, p. 1044–1063.


McNeal, R. P., 1965, Hydrodynamics of Permian Basin:AAPG Memoir 4, p. 308–326.

McNeal, R. P., and T. D. Mooney, 1968, Relationships of oilcomposition and stratigraphy of Delaware reservoirs, inBasins of the southwest, v. 2: Wichita Falls, North TexasGeological Society, p. 68–75.

Meissner, F. F., 1972, Cyclic sedimentation in Middle Perm-ian strata of the Permian Basin, west Texas and NewMexico, in Cyclic sedimentation in the Permian Basin,2d ed.:Midland,West TexasGeological Society, p. 203–232.

Milliken, K. L., and L. S. Land, 1991, Reverse weathering,the carbonate-feldspar system, and porosity evolutionduring burial of sandstones (abs.): AAPG Bulletin, v. 75,p. 636.

Milliken, K. L., and L. S. Land, 1993, The origin and fate ofsilt sized carbonate in subsurface Miocene–Oligocenemudstones, south TexasGulf Coast: Sedimentology, v. 40,p. 107–124.

Morad, S., 1998, Carbonate cementation in sandstones: Dis-tribution patterns and geochemical evolution, in S.Morad,ed., Carbonate cementation in sandstones: InternationalAssociation of Sedimentologists Special Publication 26,p. 1–26.

Moraes, M. A. S., and R. C. Surdam, 1993, Diagenetic hetero-geneity and reservoir quality: fluvial, deltaic, and tur-biditic sandstone reservoirs, Potiguar and Reconcavo riftbasins, Brazil: AAPG Bulletin, v. 77, p. 1142–1158.

O’Neil, J. R., R.N.Clayton, andT. K.Mayeda, 1969,Oxygenisotope fractionation in divalent metal carbonates: Jour-nal of Chemical Physics, v. 51, p. 5547–5558.

Paxton, S. T., J. O. Szabo, J. M. Ajdukiewicz, and R. E.Klimentidis, 2002, Construction of an intergranularvolume compaction curve for evaluating and predictingcompaction and porosity loss in rigid-grain sandstonereservoirs: AAPG Bulletin, v. 86, p. 2047–2067.

Pittaway, K. R., and R. J. Rosato, 1991, The Ford Geraldineunit CO2 flood-update 1990: Society of Petroleum En-gineers Reservoir Engineering, v. 6, no. 4, p. 410–414.

Ruggiero, R. W., 1985, Depositional history and performanceof a Bell Canyon sandstone reservoir, Ford-Geraldinefield, west Texas: Master’s thesis, University of Texas atAustin, Austin, Texas, 242 p.

Scholle, P. A., 1978, A color illustrated guide to carbonaterock constituents, textures, cements, andporosities:AAPGMemoir 27, 241 p.

Silva, C. M. A., E. B. Rodrigues, and S. M. C. Anjos, 1998,Composition and origin of carbonate concretions in res-

ervoirs of Marlim field, Campos Basin, Brazil (abs.):AAPG Bulletin, v. 82, p. 1967.

Slatt, R. M., and G. L. Hopkins, 1990, Scaling geologic res-ervoir description to engineering needs: Journal of Petro-leum Technology, v. 42, no. 2, p. 202–210.

Sombra, C. L., A. B. Barreto Jr., R. K. Romeu,W. L. Lanzarini,H. Lopes, S. Pesco, andG. Tavares, 1998, Stochastic mod-eling of calcite concretions in Cretaceous turbiditic sand-stones of the Albacora oil field, Campos Basin, Brazil(abs.): AAPG Bulletin, v. 82, p. 1969.

Souza, R. S., and C. M. A. Silva, 1998, Origin and timingof carbonate cementation of the Namorado Sandstone(Cretaceous), Albacora field, Brazil: Implications for oilrecovery, in S. Morad, ed., Carbonate cementation insandstones: Distribution patterns and geochemical evo-lution: Special Publication of the International Associa-tion of Sedimentologists, 1998, v. 26, p. 309–325.

Stewart, R. N. T., R. S. Haszeldine, A. E. Fallick, M.Wilkinson, and C. I. Macaulay, 2000, Regional distribu-tion of diagenetic carbonate cement in Palaeocene deep-water sandstones: North Sea: ClayMinerals, v. 35, p. 119–133.

Sullivan, K. B., and E. F. McBride, 1991, Diagenesis of sand-stones at shale contacts and diagenetic heterogeneity,Frio Formation, Texas:AAPGBulletin, v. 75, p. 121–138.

Walderhaug, O., and P. A. Bjørkum, 1998, Calcite growthin shallow marine sandstones: Growth mechanisms andgeometry, in S. Morad, ed., Carbonate cementation insandstones: International Association of Sedimentolo-gists Special Publication 26, p. 179–192.

Watson, R. S., N. H. Trewin, and A. E. Fallick, 1995, Theformation of carbonate cements in the Forth and Bal-moral fields, northern North Sea: A case for biodegrada-tion, carbonate cementation and oil leakage during earlyburial, in A. J. Hartley and D. J. Prosser, eds., Char-acterization of deep marine clastic systems: GeologicalSociety (London) Special Publication 94, p. 177–200.

Williamson, C. R., 1978, Depositional processes, diagenesisand reservoir properties of Permian deep-sea sandstones,Bell Canyon Formation, Texas-New Mexico: Ph.D. dis-sertation, University of Texas at Austin, Austin, Texas,262 p.

Woodruff, C. M., C. Gever, and D. R. Wuerch, 1984, Geo-thermal gradientmap of Texas (and generalized tectonicfeatures): University of Texas at Austin, Bureau ofEconomic Geology map prepared for U.S. Depart-ment of Energy under DOE contract no. DE-AS07-79ID12057, scale 1:1,013,760, 1 sheet.

Dutton 787

Calcite cement in Permian deep-water sandstones, Delaware ... · stone in the upper Bell Canyon...

Documents

Transcript of Calcite cement in Permian deep-water sandstones, Delaware ... · stone in the upper Bell Canyon...