GEOLOGICAL CONTROLS ON SONIC VELOCITY IN …legacy.earthsci.unimelb.edu.au/basinstudies/Publications/Wallace et... · stratigraphy of the Northern Carnarvon Basin (Hocking, 1988;

APPEA JOURNAL 2003—1

M.W. Wallace1, E. Condilis1, A. Powell1,J. Redfearn2, K. Auld2, M. Wiltshire3,G. Holdgate1 and S. Gallagher1

1School of Earth ScienceUniversity of MelbourneVictoria 30102Santos Ltd91 King William StreetAdelaide SA 50003Wiltshire Geological Services Pty Ltd13 St Andrews AvenueMount Osmond SA [email protected]

ABSTRACT

The Cenozoic carbonates of the Bounty-Talisman regioncan be divided into five major facies. From oldest toyoungest, these are: Paleocene to Eocene basinal facies,Oligocene to Miocene slope-canyon facies, Oligocene toMiocene shelf facies, Oligocene to Miocene near-shorefacies, and Pliocene-Quaternary shelf facies. Thisrepresents a shallowing-upwards cycle up to the lateMiocene, followed by a significant transgression and areturn to more open marine conditions in the Pliocene-Quaternary. The dominant geological processescontrolling sonic velocity in the Cenozoic carbonates arephysical compaction, burial calcite cementation,dolomitisation, and anhydrite/gypsum cementation. Inthe more open marine facies of the Cenozoic carbonates,compaction and burial calcite cementation have beenthe dominant geological processes that have controlledsonic velocity. Large-scale carbonate content variationsassociated with submarine canyon-fill sediments havealso produced lateral sonic velocity variations.Dolomitisation and anhydrite cementation have producedlocalised high velocity zones within the near-shore faciesof the carbonates.

KEYWORDS

North West Shelf, Carnarvon basin, Dampier sub-basin, Cenozoic, sonic velocity, carbonates, diagenesis,dolomite, anhydrite, gypsum

INTRODUCTION ANDREGIONAL GEOLOGY

Despite their enormous aerial extent and stratigraphicthickness, the Cenozoic carbonates of the North WestShelf remain relatively poorly documented. The Cenozoic

carbonates are the dominant cover sequence to thehydrocarbon-producing Mesozoic successions, and so havea significant role in source rock maturation. Furthermore,the carbonates cause considerable problems in the seismicinterpretation of structural traps because of the presenceof strong lateral sonic velocity variations (e.g. Cowley,1989; Woods, 1991). This study documents thestratigraphic framework for the carbonates of the Bounty-Talisman region (Fig. 1) in the Northern Carnarvon Basinand examines the major lithological controls on sonicvelocity in these carbonates.

While several studies have dealt with the generalstratigraphy of the Northern Carnarvon Basin (Hocking,1988; Woodside Offshore Petroleum, 1988, Romine et al,1997), relatively few studies have dealt specifically withthe Cenozoic carbonates of the region (Apthorpe, 1988;Heath and Apthorpe, 1984; Hull et al., 1998; Young et al,2001; Hull and Griffiths, 2002). Densley et al (2000) usedinterval velocities for Cretaceous sediments of theCarnarvon Basin to quantify uplift. No published studies,however, have specifically dealt with the sonic velocitybehaviour of the Cenozoic carbonates.

The offshore Cenozoic carbonate stratigraphy for theNorthern Carnarvon basin has been established by Heathand Apthorpe (1984) and is illustrated in Figure 2. Thefacies subdivisions presented in this paper have not beenproduced as lithostratigraphic units and are therefore

GEOLOGICAL CONTROLS ON SONIC VELOCITY IN THECENOZOIC CARBONATES OF THE NORTHERN CARNARVON

BASIN, NORTH WEST SHELF, WESTERN AUSTRALIA

2

2A

35

6

PG93-1060 Sable 1

Finucane 183MPB-141

95S-331

95S-321

Bounty 1Aurora 1

Calypso 1

86MPE-199

De Grey 1Alpha North 1Talisman 1

Forestier 1Forestier 2

Legendre 1Titan 1

Samson 1Spinner 1

Baleena 1

Cygnus 1

Lambert 1

Angel 1Angel 3

Angel 4

Angel 2

Wanaea 1

Cossack 1 32

4

4

34

26

7

N

Lauchie 1

20 km

Nelson Rocks 1

116 30 Eo '

19 30 So '

117 00 Eo '116 30 Eo '

19 30 So '

Australia

STUDY AREA

Pitcairn 1

Figure 1. Seismic line and well location map of the NorthernCarnarvon Basin, North West Shelf, Western Australia. Wellsshown in bold are those examined in this study.

2—APPEA JOURNAL 2003

M.W. Wallace, E. Condilis, A. Powell, J. Redfearn, K. Auld, Mike Wiltshire, G. Holdgate and S. Gallagher

SECOND PROOFSECOND PROOF—2/11 24 JANUARY 2003

not directly comparable to the units of Heath andApthorpe (1984).

METHODS

Carbonate analysis was performed on about 200samples including the wells Bounty–1, and Pitcairn–1,Finucane–1, Calypso–1, Talisman–1,2,3, 4, and 5,Aurora–1, Alpha North–1, and Sable–1 (Fig. 1) usingvolumetric analysis of gas evolved after digestion in HCl(Hülsemann, 1966). Analytical precision for carbonateanalysis was ± 1.5% at 1 σ. Elemental analyses wereperformed on powdered cuttings samples (same samplesas used for carbonate analysis) using a Phillips XL30ESEM equipped with an Oxford INCA 300 x–raymicroanalysis system. Elemental analyses were convertedto normative mineral concentrations and used as anapproximation for mineral abundance, includingdolomite, calcite, anhydrite, quartz, clay down the wells.About 100 thin sections were examined. All lithologicaland geochemical data was obtained from cuttings samplesthat were collected from the Geoscience Australia andSantos core stores. The averaging effect andrepresentative sample that cuttings provide (cuttingsbeing derived from a depth interval of several metres)proved ideal for correlation with wireline log data andseismic. Wiltshire Geological Services and Santos donateddigital wireline log data.

0

2

1

3

Dep

th k

m (

appr

oxim

ate)

Sea Floor

AGE AGE

Rosemary Nth.

Area

Withnell/

Parker Area

Rankin

Platform

Brigadier

Area

Bare

Formation

Delambre

Formation

Trealla

Formation

Mandu

Formation

Walcott Fm.

Walcott F

m.

Dockrell Fm.

With

nell Fm.

Muderong Shale

Haycock

Marl

Muderong Shale

Lambert Fm.

Pleistocene

to

Late

Miocene

M. Miocene

M.- Late Eocene

Early

Maastrichtian

to Early

Campanian

Coniacian

to Albian

Late toM. Eocene

Figure 2. Diagrammatic cross section illustrating the stratigraphy of the Northern Carnarvon Basin (after Heath and Apthorpe, 1984).

STRATIGRAPHY

We have subdivided the Cenozoic carbonates of theTalisman-Bounty area into five major facies subdivisionsthat have been defined by seismic properties, wirelinelog character (gamma and sonic velocity), petrologicdata (from thin sections of cuttings), geochemical data(carbonate analysis and elemental analysis from cuttings)and foraminiferal analysis (from cuttings). The five majorfacies, from oldest to youngest are: basinal facies, slope-canyon facies, Oligo-Miocene shelf facies, near-shorefacies, and Pliocene-Quaternary shelf facies (Fig. 3,Table 1). The Paleocene to Late Miocene successionrepresents an overall shallowing-upwards depositionalcycle. During the early Pliocene, a major transgressionoccurred, bringing more open shelf conditions and purelimestone deposition to the region.

Basinal facies

Seismically, this facies is characterised by laterallycontinuous conformable reflectors (Fig. 3). Numerousstratigraphically confined normal faults (of probablecompaction origin, Lonergan and Cartwright, 1999) arealso present in some sections of this facies. Lithologically,this facies is characterised by calcareous shales andmarls (30–60% calcite) that are frequently high in organicmatter. Fine-grained fragmental bioclasts and quartz siltare the dominant clast types, with the matrix consisting


Geological controls on sonic velocity in the Cenozoic carbonates of the Northern Carnarvon Basin, North West Shelf, Western Australia


T6

-

-

-

-

0

2800

86MPE

-19995S-32183MPB

-14195S-331PG

93-1060

BOUNTY 1 FINUCANE 1 CALYPSO 1NE SW

Mio-Pliocene Boundary

Oligo-Miocene Boundary

K-T Boundary

Tw

o W

ay T

ime s

1

2

Quartz sand

Dolomitic carbonates

Near-shore

Facies

Shelf Facies

Slope-Canyon Facies

Basinal Facies

Tw

o W

ay T

ime s

1

2

Bounty 1 Finucane 1 Calypso 1NE SW

Figure 3. Interpreted and uninterpreted seismic profile with lithological and facies data (derived from cuttings analysis) superimposed.




Table 1. Summary of characteristics for each of the carbonate facies.

Facies Lithology Seismic Age range Stratigraphic Sonic velocitycharacter equivalent

Pliocene- Uncemented Laterally continuous, Pliocene- Delambre Fm. Laterally uniformQuaternary coarse-grained, conformable reflectors Quaternary velocityShelf Facies pure limestone

Carbonate:>80 wt%

Near-shore Dolomites, quart- High amplitude Oligocene- Bare Fm. Laterally variable withFacies zose, dolomites, reflectors, small-scale Quaternary Trealla Fm. high velocity dolomtised

quartz sands, progrades, erosional Mandu Fm. zonesdolomite lime- surfacesstones, gypsumanhydriteCarbonate:0–90 wt%

Oligo- Pure skeletal Laterally continuous Oligocene- Trealla Fm. Laterally uniformMiocene limestones conformable reflectors Miocene Mandu Fm. velocityShelf Carbonate:Facies >85 wt%

Slope- Clayey lime- Large-scale prograding Oligocene- Trealla Fm. Laterally variablecanyon stones clinoforms, erosional Miocene Mandu Fm. with low velocityFacies Carbonate: surfaces abundant intervals

60–85 wt%Basinal Fine-grained Laterally continuous Paleocene- Walcott Fm. Laterally uniform lowFacies calcareous conformable reflectors Miocene Dockrell Fm. velocity

shales and Lambert Fm.marlsCarbonate:60–85 wt%

of microcrystalline clay-rich carbonate (Fig. 4A). Somelithologies have a high proportion of delicately preservedplanktonic foraminifera as clasts. The unit is relativelywell-cemented, with foraminiferal chambers being filledpredominantly by calcite cements. Coarsely crystallineeuhedral siderite is also sporadically present. Theuppermost portion of this facies is a laterally continuouscherty unit (~50 m thick), characterised by a relativelyhigh carbonate content and high sonic velocities. The ageof the basinal facies in the Bounty-Finucane area isPaleocene to mid-Eocene and is chronostratigraphicallyequivalent to the Walcott, Dockrell and Lambertformations of Heath and Apthorpe (1984). Furtherbasinward, the basinal facies becomes younger (Oligoceneand Miocene) in the upper sections (Fig. 3).

The abundance of planktonic foraminifera and theuniformly fine, clay-rich nature of the sediments indicatea deep marine basinal environment of deposition (baseof slope and deeper). A sparse low diversity bathyalbenthic foraminiferal assemblage with abundantplanktonic forms typifies this facies. C. perforatus andLenticulina occur with deeper water agglutinated formssuch as Cyclammina and Marsonella. Planktonicforaminifera such as Acarinina Globigerinahteka andSubbotina are abundant.

Slope-Canyon facies

The slope-canyon facies is characterised seismicallyby prograding clinoformal reflectors that displayconsiderable irregularity. Numerous erosional surfacesand downlaps are also present (Fig. 3). Reflectors of theslope-canyon facies largely downlap onto the underlyingbasinal facies. Lithologically, this facies is characterisedby clayey limestones (averaging 60–85 % calcite) whichincrease in carbonate content upwards. The lithologiesare highly variable both laterally and vertically, andinclude fine-grained calcareous shales, marls, packstonesand grainstones. The dominant rock types are well-sortedfragmental bioclast packstones and grainstones with avariable proportion of quartz silt (Fig. 4B). However,quite coarse-grained packstones and grainstones withabundant large foraminifera are also present (Fig. 4C).Small quantities of euhedral finely crystalline dolomiteare sporadically present in this facies. The age of thisfacies in the Bounty-Finucane area ranges from LateOligocene to mid-Miocene. Chronostratigraphically, thisfacies is partly equivalent to the Mandu and Treallaformations of Heath and Apthorpe (1984).

Based on seismic, lithological and foraminiferal data,the facies has been assigned to a deep marine slope-canyon setting. The abundant erosional structures (inseismic) and lateral lithological variability are consistent




0.2 mm

0.6 mm 0.2 mm

0.3 mm 0.4 mm

A

0.2 mm

A B

DC

FE

Figure 4. Photomicrographs of the major lithologies from the Cenozoic carbonates. A) Organic-rich marl of the basinal facies. Bounty–1,2,600 m. B) Well-sorted quartz-fragmental bioclast packstone from the slope-canyon facies. Bounty–1, 1800 m. C) Large foraminifera incuttings sample from slope-canyon facies. Also note the marly cuttings (dark) present in the sample. D) Bioclast grainstone cemented byanhydrite (A), from the Oligo-Miocene shelf facies. Bounty–1, 1,100 m. E) Uncemented quartz sand with sparse bioclasts, near-shore facies.Bounty–1, 400 m. F) Uncemented bioclast grainstone, Pliocene-Quaternary shelf facies, Bounty 1, 220 m.




is characterised by relatively high amplitude reflectorswith a variety of geometries including small-scaleprogradation and erosion. Dolomite-cemented quartzsands and calcareous sands cemented by gypsum andanhydrite occur at two major stratigraphic intervals(mid-Miocene and Late Miocene-Early Pliocene). Thequartz is generally well-rounded, and is commonly coatedwith a thin rim of dolomite cement. Gypsum and anhydritecements in the quartz sands generally have a very coarselycrystalline poikilotopic fabric. Seismically, the quartzsand-dominated lithologies are characterised by highamplitude reflectors with small-scale progradation andmounding (Fig. 3).

Underlying each of these quartz sand units arepervasively dolomitised carbonates and interbeddedlimestones containing abundant gypsum and anhydrite.Dolomite fabrics range from euhedral to anhedral andmedium to coarsely crystalline (Figs 5A-C). Finely-crystalline dolomite is more rarely present. The crystalsize of the dolomite has a tendency to increase downwardsfrom the base of the quartz sands. Coarsely crystallinegypsum and anhydrite cements (Figs 5A, C) commonlyfill secondary porosity within the dolomites and primaryporosity in the interbedded limestones. There iscommonly very little secondary porosity, however, withinthe pervasively dolomitised lithologies (Fig. 5B).Lithologically and chronostratigraphically, the quartzsands may be equivalent to the Bare Formation (Heathand Apthorpe, 1984) while the dolomites arechronostratigraphically equivalent to the Trealla andMandu formations.

We interpret this facies as being deposited in a near-shore setting. This is based on the presence of minorfinely crystalline dolomite and anhydrite, interpreted tobe of syn-sedimentary hyper-saline supra-tidal origin(Morrow, 1982). The abundant coarsely crystallinegypsum and anhydrite cements, together with the mediumto coarsely crystalline dolomites are probably derivedfrom hyper-saline refluxing brines, produced in sabkhaand possibly playa lake environments (Sears and Lucia,1980; Morrow, 1982). The quartz sand-rich lithologiesmay represent near-shore marine clastics and barriersystems.

A poorly preserved sparse coarse-grained microfaunais present in this facies. Planispiral and trochospiralrotaliids are present although these could not be assignedto a family or generic level. The presence of this faunasuggests deposition in a restricted high-energy inner-shelf environment.

Pliocene-Quaternary shelf facies

The shallowest portion of the wells Finucane–1 andBounty–1 are characterised by uncemented coarse-grained pure limestones (mostly grainstones). Clastsinclude foraminifera, echinoderms, bivalves and peloids(Fig. 4F). Seismically, the facies is characterised bylaterally continuous conformable reflectors. The age ofthis unit ranges from Pliocene to Quaternary age and the

with the presence of small (tens of metres deep and<1 km wide) submarine canyons. Lithologically, the well-sorted fragmental bioclast packstones are similar tothose from the Cenozoic canyons of the Gippsland Basin(Wallace et al, 2002).

The bimodal foraminiferal assemblage (outer shelfand bathyal forms) suggests mixing of shelf and basinforms, also consistent with a slope-canyon environment(Gallagher et al, 2001). The assemblage is dominated byplanktonic foraminifera and a diverse outer shelf toupper bathyal benthic assemblage including: Cibicidoidesperforatus, Gyroidinoides spp., Pullenia spp. andSphaeroidina bulloides. Taxa typical of upwelling dysoxicfacies such as S. bulloides, Globocassidulina subglobosa,Brizalina spp. and Cassidulina spp. are also common. Cooltemperate plankton taxa such as Catapsydrax dissimilisco-occur with warmer plankton Globoquadryina andGlobigerina praebulloides. Overall, the data suggests anouter shelf to upper slope environment with upwellingwithin canyon facies. Occasional inner shelf taxa (rareSpirillina and Elphidium ) and size sorting of tests aretypical of canyon environments where material istransported downslope from the shelf.

Oligo-Miocene shelf facies

Relatively pure limestones (> 85 % calcite) overlie themarly carbonates of the slope-canyon facies. Seismically,this facies is characterised by laterally continuousconformable reflectors that are laterally equivalent tothe underlying clinoforms of the slope-canyon facies(Fig. 3). A range of lithologies is present includingrelatively coarse-grained bioclastic wackestones,packstones and grainstones. Clast types includefragmental bioclasts, and peloids (Fig. 4D). Calcitecements are present in all lithologies, but do notcompletely fill porosity. In the uppermost portions of thisfacies, anhydrite cements are common. The upperboundary of this facies is arbitrarily taken as the firstoccurrence of abundant dolomite. The age of this faciesin the Bounty-Finucane area ranges from Late Oligoceneto mid-Miocene, becoming younger basinwards. This unitis chronostratigraphically equivalent to the Mandu andTrealla formations of Heath and Apthorpe (1984).

Foraminiferal analysis of one sample from the upperpart of this facies (Bounty–1, 1,100 m) shows the presenceof common Elphidium crispum, an inner shelf taxon,associated with abundant Operculina complanata. Otherimportant taxa present are Lepidocyclina andAmphistegina. Plankton is absent and the foraminiferaare coarse-grained suggesting high-energy inner-shelfdeposition.

Near-shore facies

Overlying and interspersed with the Oligo-Mioceneshelf facies limestones are dolomites, quartzose dolomites,dolomitic and pure quartz sands and dolomitic limestoneswith common gypsum and anhydrite (Fig. 4E). This facies




0.2 mm

0.2 mm 0.3 mm

0.3 mm 0.2 mm

0.2 mm

A B

DC

FE

A

A

G

Figure 5. Photomicrographs of the major diagenetic phases and textures from the Cenozoic carbonates. A) Medium crystalline dolomitewith secondary porosity filled by anhydrite cements (A). Crossed polars. Talisman 4, 520 m. B) Coarsely crystalline dolomite with nosecondary porosity. Crossed polars. Talisman 4, 620 m. C) Medium crystalline euhedral dolomite with secondary porosity filled by gypsum(G). Crossed polars. Talisman 4, 520 m. D) Undolomitised skeletal grainstone with primary porosity filled by anhydrite (A). Crossed polars.Talisman 4, 720 m. E) Large foraminifera in cuttings from the low velocity zone (Fig 8). Note that there is no calcite cement within theintraskeletal pores. Plane light. F) Carbonate packstone with large foraminifera cemented by equant calcite. Crossed Polars.




unit appears equivalent to the Delambre Formation ofHeath and Apthorpe (1984). The environment ofdeposition is interpreted as an open marine shelf. Thepresence of tropical large foraminifera like Cycloclypeussp. with abundant Operculina complanata in this intervalsuggests deposition in water depths shallower than 130 m,in an open marine environment. Tropical planktonicforaminifers such as Globigerinoides, Globoconella menardiiand Beella digitata are also common (comprising ca. 20%of the total assemblage). The assemblage data suggestmiddle to outer shelf deposition (palaeodepths from50 m–130 m).

DIAGENESIS

In general, sonic velocity in clastic sedimentary rocksis determined by the elastic properties and density of themix of lithotypes present, and the average efficiency ofintergrain connection in terms of seismic wavepropagation (Prasad and Dvorkin, 2001, Dvorkin andGutierrez, 2002). The main post-depositional processesaffecting intergrain connectivity are burial compactionand chemical diagenesis, both of which affect porosity.Within relatively thick and monotonous lithologies,increasing depth of burial generally results in systematicreduction in porosity, and increase in sonic velocity. Incoarse clastics, after initial near-surface porosityreduction due to packing adjustment, the rocks becomeframe-supported, and subsequent increases in burialdepth have little effect on bulk density. Subsequentporosity reduction is controlled by diagenesis.

Major porosity-altering diagenetic processes thereforehave the greatest subsequent effect on the sonic velocity(Wyllie et al, 1958). This paper therefore only examinesthe major porosity-altering diagenetic processes and themyriad of small-scale diagenetic processes like pyritecementation, quartz cementation, meteoric carbonatedissolution, and marine cementation have beenspecifically excluded. These small-scale processes maybe locally important, but are unlikely to affect the large-scale sonic velocity behaviour of these sediments. In thiscontext, the major diagenetic processes observed withinthe carbonates of the Talisman-Bounty area are: burialcalcite cementation, dolomitisation, and anhydritecementation. The interpreted timing of the majordiagenetic processes affecting sonic velocity in the studyarea is shown in Figure 6.

Burial calcite cementation

Clear calcite cementation is abundant below a depthof 1,000 m in most wells indicating a burial diageneticorigin for the cements. Calcite cements have texturesranging from equant to scalenohedral and fibrous,depending on the cement substrate (skeletal componentshaving a fibrous microstructure have fibrous intra-skeletal cements, etc). Different skeletal componentsalso display different degrees of cementation withinintraskeletal porosity. Benthonic foraminifera generally

preserve more porosity at a given depth than do someother skeletal fragments. The degree of porositypreservation and cementation in skeletal fragments isalso dependent on the size of pores. Larger intraskeletalpores retain porosity to greater depths, perhaps becausemore calcite cement is required to fill larger pores.

The intraskeletal porosity within many largeforaminifera (Lepidocyclina, Amphistegina and Operculina)from the Bounty-Talisman area is variably occluded byclear calcite cements. Many large intraskeletal pores inthe foraminifera have almost no calcite cement (Fig. 5E),while other samples are strongly cemented (Fig. 5F).

Dolomitisation

Dolomite is abundant in the near-shore facies andcuttings analysis indicates that dolomite is stronglyassociated with anhydrite and/or gypsum cement (Fig. 5Aand 5C). Dolomite is generally abundant directly beneaththe various quartz sand units in the succession (Fig. 3). Inthe Bounty–1 and Finucane–1 wells, dolomite is mostlypresent in two horizons, both occurring beneath quartzsand (Fig. 3). The lowermost horizon is around mid-Miocene in age, while the upper dolomite horizon appearsto be late Miocene in age. In the Calypso–1 and Talisman(1–4) wells, however, dolomite is more widespread.

Dolomitisation is mostly present as a pervasive (100%dolomite) dolomite replacement in cuttings samples,with few partially dolomitised cuttings being present.Dolomite cements in calcite lithologies are also notcommon. Dolomite crystal sizes range from raremicrocrystalline fabrics (20 micron crystal size) to morecommon coarsely crystalline fabrics (average crystal size100 microns). Crystal size tends to increase downwards,with medium crystalline dolomites being more abundantimmediately beneath the quartz sand units, and coarselycrystalline dolomites being more abundant in thelowermost portions of the dolomite lenses.

The dolomite texture ranges from anhedral through toeuhedral, with little apparent correlation between crystalsize and crystal shape being evident. Unlike manydiagenetic dolomites, very little secondary porosity ispresent within the dolomitised lithologies (Fig. 5B).Furthermore, what little secondary porosity was presentwithin the dolomite has commonly been completely filledby coarsely crystalline anhydrite or gypsum (Figs 5A,5C). The overall lack of secondary porosity, combinedwith the presence of anhydrite and/or gypsum cementshas produced dolomites with a very low overall porosity.

Anhydrite-Gypsum cementation

Anhydrite and gypsum cements are present in quartzsand, limestone and dolomite lithologies and fill bothprimary and secondary porosity. Anhydrite and gypsumare both spatially associated with dolomite, and have asimilar overall distribution to dolomite. Anhydrite ismore common at shallower depths and is completelyabsent below 1 km depth.




Anhydrite and gypsum cements generally have coarselycrystalline fabrics, which are commonly poikilotopic(Figs 5A, 5C). Rare finely crystalline anhydrite has beenobserved in cuttings samples and this may be derivedfrom synsedimentary nodular anhydrite. Gypsum andanhydrite cements generally directly overlie bioclastgrains (when occurring in primary porosity) and dolomitecrystals (when occurring in secondary dolomitic porosity).Clear equant calcite cements are generally not presentin gypsum or anhydrite cemented lithologies, indicatingthat the anhydrite and gypsum cements pre-date mostburial calcite cementation and are probably of relativelyearly diagenetic origin. This is also consistent with thepresence of these gypsum/anhydrite cements at shallowburial depths.

SONIC VELOCITY

In their simplest form, sonic velocity logs from theBounty-Talisman area display a relatively uniformincrease with depth of burial, down to the base of theslope-canyon facies (Fig. 7). A strong peak in sonic velocitycoincides with the cherty horizon at the top of the basinalfacies, and below this, sonic velocity drops to around3,000 m/s. Sonic velocity is also strongly correlated withcarbonate content (Fig. 8). Therefore, as a firstapproximation, sonic velocity is related to two factors;carbonate content and burial depth. Outside of thedolomitic intervals discussed in preceding sections, calciteis the dominant carbonate and measured carbonatecontent therefore represents the calcite content. Wallaceet al (2002) have previously derived an empiricalrelationship between sonic velocity, calcite content andburial depth from Cenozoic carbonates of the GippslandBasin as follows:

v = 0.013cd + 13.125c + 0.35d + 1175where v = sonic velocity (m/s) from the sonic log, c =

CaCO3 weight percent as measured from cuttings samples,and d = depth below sea bed m. Wallace et al (2002)suggested that this relationship was valid for depthsranging from 500–2,500 m, and calcite contents rangingfrom 20–80 wt%. When this Gippsland velocity equationis applied to the carbonates from Bounty–1 and Pitcairn–1(i.e. using the carbonate content and burial depth topredict sonic velocity), the predicted velocities follow

the same general trend as those measured from the soniclog (Figs 7, 8).

Predicted velocities for Bounty–1 and Pitcairn–1 are,however, significantly higher than measured velocitiesfor most of the section. In Bounty–1, the interval between1,000 and 1,500 m depth has carbonate contents that arehigher (>80%) than those encountered in the study ofWallace et al (2002). These high calcite contents mayexplain the mismatch between predicted and actualvelocities over this interval (i.e. the relationship betweencarbonate content and sonic velocity is not linear wherecarbonate content is greater than 80%).

Other factors may also explain the mismatch betweenpredicted and actual sonic velocity, including burialhistory (Wallace et al, 2002), geothermal gradient, andlithologic texture (e.g. grainsize). These observationssuggest that while sonic velocity in the Bounty-Talismanarea is, to a large extent, controlled by calcite contentand burial depth, the carbonates of the North West Shelfare not behaving exactly as the carbonates of theGippsland Basin.

Wallace et al (2002) suggested that this correlationbetween sonic velocity, calcite content and burial depthwas a product of the main processes controlling porosityocclusion in the Gippsland carbonates; pressure solutionand burial calcite cementation. As suggested by Wallaceet al (2002), both of these processes are caused by burial,which drives compaction. Below 500 m, chemicalcompaction (pressure solution and resultant calcitecementation) is the dominant diagenetic process

Near-surfacediagenesis

Early burialdiagenesis

Diagenetic Phaseor Fabric

Anhydrite and gypsumcementation

Dolomitization

Late burialdiagenesis

Depth of Burial m

Physical compaction

Pressure solution

Clear calcite cementation

0 50010 2000

Figure 6. Paragenetic diagram illustrating the interpreted timing ofthe major diagenetic processes in the Cenozoic carbonates.

2000 3000 4000 50000

500

1000

1500

2000

2500

Dep

th m

Sonic velocity m/s

0 50 100Carbonate wt %

Basinal Facies

Canyon-Slope Facies

Oligocene-MioceneShelf Facies

Quartzsands

Dolomiticcarbonates

Pliocene-QuaternaryShelf Facies

Near-shoreFacies

Predicted velocitySonic log velocityCarbonate

BOUNTY 1

Figure 7. Carbonate content from cuttings and sonic velocity forBounty 1. Also plotted is the predicted sonic velocity, derived fromthe carbonate content using the relationship of Wallace et al., 2002.




3000 4000 5000 6000

0 50 100

1600

1800

2000

2200

2400

2600

Dep

th m

Sonic velocity m/s

Carbonate wt %

Predicted velocitySonic log velocityCarbonate

Pitcairn 1

Basinalfacies

Canyon-Slopefacies

Chertmarker

LowVelocityZone

Figure 8. Carbonate content from cuttings and sonic velocity forPitcairn–1. Also plotted is the predicted sonic velocity, derivedfrom the carbonate content using the relationship of Wallace et al(2002).

controlling porosity and hence sonic velocity.Since the sonic velocity of carbonates in the Bounty-

Talisman is correlated with burial depth and carbonatecontent, it follows that pressure dissolution and burialcalcite cementation (i.e. chemical compaction processes)are probably also important processes controlling porosityocclusion in the North West Shelf carbonates. This isconsistent with petrologic observations from thesecarbonates. The sonic velocity behaviour for the Cenozoiccarbonates of the Bounty-Talisman area is. however,more complicated than that of the Gippsland carbonates.Superimposed on this depth of burial and carbonatetrend are several other complicating factors, includingthe presence of quartz-rich sediments, dolomitisation,and anhydrite-gypsum cementation. Each of these isdealt with in the following sections.

Quartz-rich sediments

Quartz-rich intervals in the stratigraphy have morevariable and lower average velocities than limestones atsimilar depths (Fig. 7). Based on petrologic observations

from cuttings samples, the lower velocities are probablycaused by the higher average porosity, and lower calcitecement contents. The variable velocities observed in thequartz sand are probably due to the erratic presence ofanhydrite/gypsum and carbonate cements.

Dolomitisation and Gypsum-Anhydrite Cements

Intervals with a relatively high dolomite content havea more variable and higher average velocity thanlimestone lithologies at equivalent depths (Figs 9, 10).For example, in Alpha North–1, dolomitised carbonateshave sonic velocities greater than 5,000 m/s at only 800 mdepth (Fig. 10). The higher sonic velocities fromdolomitised intervals may be partly due to the highermatrix velocity of dolomite relative to calcite(Schlumberger, 1989). Matrix velocities for dolomite areestimated to be around 7,000 m/s, while those for calcitevary from 6,400– 7,000 m/s (Schlumberger, 1989). Takingthe lower matrix velocity estimate for calcite, limestonewould have a 10 % lower matrix velocity than dolomite.Therefore, by itself, the higher dolomite matrix velocityis insufficient to explain the observed higher sonicvelocities for dolomitised intervals in the study area(which can be greater than 20% higher in dolomitisedintervals).

The higher average velocities of the dolomites maylargely be explained by the lower average porosity ofdolomitised intervals (from petrologic analysis). As wasdiscussed in preceding sections, there is very littlesecondary porosity in strongly dolomitised intervals, andthe little secondary porosity that was present has largelybeen occluded by gypsum and anhydrite cements. Themore variable nature of the velocity in dolomitic sectionsmay be due to the erratic presence of secondary porosityand sulfate cements.

SONIC VELOCITY AND DEPOSITIONALENVIRONMENT

The Cenozoic carbonate succession of the Bounty-Talisman area records carbonate deposition from deep-water basinal environments through to peritidalconditions. All of these environments have produced acharacteristic set of depositional and diageneticconditions that has led to a wide variety of rock types androck properties. Strong lateral sonic velocity variationsare produced where large-scale lateral changes indepositional or diagenetic conditions have occurred (e.g.Fig. 10). The two major environmental settings that haveproduced large-scale lateral sonic velocity variations inthis area are the near-shore facies and the slope facies.

In the near-shore facies, dolomitisation and closelyassociated gypsum and/or anhydrite cementation haveproduced high velocity zones. These also appear to berelated to the non-porous nature of the dolomites.Dolomitised lithologies predominantly occur under oradjacent to quartz sand-rich lithologies (Figs 3, 10). Thepresence of rare finely crystalline dolomite and anhydrite




400

600

800

1000

1200

1400

Dep

th m

2000 3000 4000 5000 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40

Sonic Velocity m/s Dolomite % Anhydrite % Quartz %

Dep

th m

Sonic Velocity m/sQuartz sands &quartzose carbonates

Dolomiticcarbonates Limestones Marls and calcareous

shales

400

600

800

1000

1200

1400

BOUNTY 1 FINUCANE 1 CALYPSO 1 ALPHA NORTH 1

2000 3000 4000 5000 6000 2000 3000 4000 5000 6000 2000 3000 4000 5000 6000 2000 3000 4000 5000 6000

Figure 9. Sonic velocity log for Finucane–1, compared with normative mineral abundance derived from cuttings analysis.

Figure 10. Well log cross section illustrating extensive lateral sonic velocity variation caused by dolomitisation. Note the very high sonicvelocities present in Alpha North–1 that correspond to an extensively dolomitised interval.




suggests that much of the dolomitisation is related tosabkhas which developed adjacent to near-shore coastalsand and barrier systems (Morrow, 1982).

Most of the dolomite is not, however, of syn-sedimentary sabkha origin as it is either medium orcoarsely crystalline. The bulk of this coarsely crystallinedolomite may have been produced by brine reflux (Searsand Lucia, 1980). In this scenario, dense hyper-salinebrines produced on the sabkhas may have descendedinto the underlying shelfal limestones to produce earlydiagenetic dolomitisation. The coarsely crystallineanhydrite/gypsum cements may be similarly related tobrine reflux.

It appears probable that these episodes ofdolomitisation on the shelf are related to major regressiveevents. The greater intensity of dolomitisation onstructural highs (e.g. Alpha North–1, and Calypso–1,Fig. 10) may be due to repeated exposure and sabkhadevelopment during various regressive phases in theCenozoic.

For different reasons, the slope facies has producedsimilarly problematic lateral sonic velocity variations. Inthis environment, submarine canyons are important incontrolling large-scale lithological changes (c.f. GippslandBasin, Holdgate et al, 2000; Wallace et al, 2002). Theslope-canyon facies has a highly variable sonic velocityand lithological character. This is almost certainly due tothe large differences in environmental conditions withinand between individual canyons. Canyon fill sedimentsare typically high-energy mass flow deposits derivedfrom the outer shelf. The slope sediments betweenindividual canyons are more typically low energysediments deposited by suspension settling. This intimatemixture of high and low energy environments would beexpected to produce carbonate-rich and carbonate-poorlithologies that, during burial diagenesis, produce highand low velocity zones (Fig. 8).

CONCLUSION

The Cenozoic carbonates of the Bounty–Talismanregion can be divided into five major facies. From oldestto youngest, these are: Paleocene to Eocene basinalfacies, Oligocene to Miocene slope-canyon facies,Oligocene to Miocene shelf facies, Oligocene to Miocenenear-shore facies, and Pliocene-Quaternary shelf facies.Overall, this represents a shallowing upwards up to thelate Miocene, followed by a significant transgression anda return to more open marine conditions in the Pliocene-Quaternary.

The dominant geological processes controlling sonicvelocity in the Cenozoic carbonates are compaction,burial calcite cementation, dolomitisation, and anhydrite-gypsum cementation. In the more open marine facies ofthe Cenozoic carbonates, physical compaction and burialcalcite cementation have been the dominant geologicalprocesses that have controlled sonic velocity. In suchconditions, sonic velocity is largely a function burialdepth and carbonate content. Lateral variations in sonic

velocity can be caused by laterally varying carbonatecontent that are commonly associated with submarinecanyons.

Dolomitisation and anhydrite cementation haveproduced localised high velocity zones within the near-shore facies of the carbonates. These diagenetic processeshave probably been initiated by extensive sabkhadevelopment with associated brine reflux in the near-shore carbonate systems. Dolomitisation and sabkhadevelopment appear to be associated with regressivephases on the shelf and are better developed on structuralhighs.

ACKNOWLEDGEMENTS

We are most grateful to Santos Ltd for technical helpand for permission to publish. Wiltshire Geological Ser-vices kindly donated the digital log data for the study.We thank the reviewers David Dewhurst and AbbasKhaksar for their constructive reviews. This study formspart of an ARC-Linkage project on the Cenozoic carbon-ates of the North West Shelf.

REFERENCES

APTHORPE, M., 1988—Cainozoic depositional history ofthe North West Shelf. In: Purcell, P.G. and Purcell, R.R.(eds), The North West Shelf, Australia. Proceedings,PESA Symposium, Perth, 55–84.

COWLEY, R.C., 1989—Depth conversion problems ofthe Skua Field. Exploration Geophysics, 20, 297–300.

DENSLEY, M.R., HILLIS, R.R. AND REDFEARN, J.E.P.,2000—Quantification of uplift in the Carnarvon Basinbased on interval velocities. Australian Journal of EarthScience, 47(1), 111–22.

DVORKIN, J. AND GUTIERREZ, M., 2002—Grain size,porosity and elasticity. Petrophysics, 43,185–96.

GALLAGHER, S.J., SMITH, A.J., JONASSON, K.,WALLACE, M.W., HOLDGATE, G.R., DANIELS, J. ANDTAYLOR, D., 2001—The Miocene palaeoenvironmentaland palaeoceanographic evolution of the Gippsland Basin,southeast Australia: a record of Southern Ocean change.Palaeogeography, Palaeoclimatology, Palaeoecology, 172,53–80.

HEATH, R.S. AND APTHORPE, M.C., 1984—Newformation names for the Late Cretaceous and Tertiarysequence of the southern North West Shelf. GSWA,Record, 1984/7, 35.

HOCKING, R.M., 1988—Regional geology of the northernCarnarvon Basin. In: Purcell, P.G. and Purcell, R.R.(eds), The North West Shelf, Australia. Proceedings,PESA Symposium, Perth, 97–114.




HOLDGATE, G.R., WALLACE, M.W., DANIELS, J.,GALLAGHER, S.J., KEENE, J.B. AND SMITH, A.J.,2000—Controls on Seaspray Group sonic velocities in theGippsland Basin—a multidisciplinary approach to thecanyon seismic velocity problem. APPEA Journal, 40 (1),295–313.

HULL, J.N.F., SMITH, S.A. AND YOUNG, H.C., 1998—Sequence stratigraphic interpretation of carbonatewireline log motifs: an example from the North WestShelf of Australia. APPEA Journal, 38 (1), 188–98.

HULL, J.N.F. AND GRIFFITHS, C.M., 2002—Sequencestratigraphic evolution of the Albian to Recent section ofthe Dampier Sub-basin, North West Shelf Australia. InKeep, M. and Moss, S.J. (eds), The Sedimentary Basins ofWestern Australia 3: Proceedings, PESA Symposium,Perth, 617–39.

HÜLSEMANN, J., 1966—On the routine analysis ofcarbonates in unconsolidated sediments. Journal ofSedimentary Petrology, 36, 622–5.

LONERGAN, L. AND CARTWRIGHT, J.A., 1999—Polygonal faults and their influence on deep-watersandstone reservoir geometries, Alba Field, UnitedKingdom, North Sea. AAPG Bulletin, 83, 410–32.

MORROW, D.W., 1982—Diagenesis 2 Dolomite—Part 2.Dolomitisation models and ancient dolostones.Geoscience Canada, 9, 95–107.

PRASAD, M. AND DVORKIN, J., 2001—Velocity toporosity transform in marine sediments. Petrophysics,42,5, 429–36.

ROMINE, K.K., DURRANT, J.M., CATHRO, D.L. ANDBERNADEL, G., 1997—Petroleum play elementpredicution for the Cretaceous-Tertiary basin phase,

Northern Carnarvon Basin. APPEA Journal, 37 (1),315–39.

SCHLUMBERGER, 1989—Log Interpretation Principles/Applications. Schlumberger Educational Services,Houston, Texas, various pagings.

SEARS, S.O. AND LUCIA, F.J., 1980—Dolomitisation ofnorthern Michigan Niagara reefs by brine refluxion andfreshwater/seawater mixing. In: Zenger, D.H., Dunham,J.B. and Ethington, R.L. (eds), Concepts and Models ofDolomitisation. SEPM, Special Publication No. 28, Tulsa,215–35.

WALLACE, M.W., HOLDGATE, G.R., DANIELS, J.,GALLAGHER, S.J. AND SMITH, A.J., 2002—Sonicvelocity, submarine canyons, and burial diagenesis inOligocene-Holocene cool—water carbonates, GippslandBasin, southeast Australia. AAPG Bulletin, 86 (9),1,593–607.

WOODS, E.P., 1991—Eocene seismic velocity anomaliesfrom the Timor Sea. Exploration Geophysics, 22, 461–4.

WOODSIDE OFFSHORE PETROLEUM, 1988—A reviewof the petroleum geology and hydrocarbon potential ofthe Barrow-Dampier sub-basin and environs. In: Purcell,P.G. and Purcell, R.R. (eds), The North West Shelf,Australia. Proceedings, PESA Symposium, Perth, 115–28.

WYLLIE, M.R.J., GREGORY, A.R. AND GARDNER,G.H.F., 1958—An experimental investigation of factorsaffecting elastic wave velocities in porous media.Geophysics, 23, 459–93.

YOUNG, H.C., LEMON, N.M. AND HULL, J.N.F., 2001—The Middle Cretaceous to Recent sequence stratigraphicevolution of the Exmouth-Barrow margin, WesternAustralia. APPEA Journal, 41 (1), 381–413.

Authors' biographies over page.




Malcolm Wallace obtained aPhD in 1988 from the Univer-sity of Tasmania on sedimentol-ogy and diagenesis of Devonianreefs, Canning Basin WA. Heworked as a research fellow atthe University of Adelaide onthe sedimentology andgeochemistry of the Late Prot-erozoic Acraman impact ejectahorizon (1988-91). He is pres-ently a senior lecturer in sedi-mentology at the School of Earth

Sciences, University of Melbourne. His current researchinterests include the stratigraphy, sedimentology and diagen-esis of Tertiary successions of the Australian basins, togetherwith diagenesis and mineralisation in sedimentary carbonatesuccessions worldwide.

THE AUTHORS

Michael Wiltshire has a de-gree in Geology and Geophysicsfrom Sydney University, and hasworked in the Australian petro-leum exploration industry since1964. He established WiltshireGeological Services as an inde-pendent consultancy in 1969; thatcompany has since operated con-tinuously, providing technical anddata services to the industry.Michael’s current interests lie inrock physics, and the use of the

WGS data resource to guide development of interpretivetechniques integrating geology, petrophysics and seismic studies.

Kerri Auld B.Sc (Hons) MBA isa commercial analyst, formerly asenior geologist in the CarnarvonBasin exploration team at Santosworking in the Dampier Sub ba-sin. Kerri obtained an honoursdegree in Geology and Geophys-ics from the National PetroleumCentre for Geology and Geo-physics (NCPGG) in 1994. In1995, Kerri joined Santos and hasgained eight years experience inoffshore Western Australian

acreage. Kerri has completed a MBA (Advanced) at the Univer-sity of Adelaide and currently works in the Commercial andPlanning Department. Kerri is a member of the PESA and AAPGorganisations.

Elizabeth Condilis studiedat the University of Melbourneand completed her under-graduate degree with honourswithin the School of Earth Sci-ence in 2000. In 2001 sheworked as a consultant at anenvironmental andgeotechnical company, Dou-glas Partners Pty Ltd. She iscompleting her first year in aPhD at the University ofMelbourne and is also work-

ing at the Department of Primary Industries within thePetroleum and Energy Division.

Anne Powell completed a BSc(Hons) degree at the Universityof Melbourne in 2002. HerHonours project focussed onsonic velocity controls within theCenozoic carbonates of theDampier Sub-Basin, North WestShelf. In early 2003 she joinedthe Basins Studies Group withinthe Petroleum DevelopmentBranch of the Department ofPrimary Industries in Melbourne.Member: PESA

Jim Redfearn graduated with aBA in Natural Sciences fromCambridge University (1976) anda MSc from the University Col-lege of North Wales (1977). Jimcommenced his career in seismicdata processing with SSL beforemoving on to interpretation geo-physics with South Australian Oiland Gas Corporation, Adelaidethen Amerada Hess, London. Hejoined Santos in 1988 as a seniorExplorationist and has worked in

the Cooper Basin, new ventures and offshore exploration. Heis now senior staff geophysicist in the Asset Development teamof the WA Business Unit. Member: PESA and SEG.




Continued from previous page.

Stephen Gallagher has a PhD(University College, Dublin) onLower Carboniferous stratigra-phy, biostratigraphy and foramin-ifera in 1992. He is thepalaeontology lecturer at theSchool of Earth Sciences,Melbourne University. His re-search interests include the bio-stratigraphy of macro and micro-fossils in carbonate sediments.His previous employment in-cludes part-time consultant for

Conodate Ltd, Ireland (1990-94) and as a micropaleontologistfor the Geochem Group Ltd, UK from 1991 to 1993.

Guy Holdgate has a PhD inGippsland Basin coal geology(Monash University, 1997) and aMSc in geology (Victoria Univer-sity of Wellington, 1972). Since1990 he has consulted to a num-ber of companies on a variety ofAustralian and New Zealand ex-ploration ventures in oil and gas,coal and minerals. He is the prin-cipal of Guy Holdgate and Asso-ciates Pty Ltd and a director ofTyers Petroleum Pty Ltd. Since

1998 he has been a research associate at the School of EarthSciences, Melbourne University on carbonates in the offshoreGippsland Basin. He has worked as a consulting geologist(1970-71); GSV basin studies (1973-1977); and between 1977and 1990 he was Coal Geology Head with the Exploration andGeological Division, State Electricity Commission, Victoria.

Documents

GEOLOGICAL CONTROLS ON SONIC VELOCITY IN …legacy.earthsci.unimelb.edu.au/basinstudies/Publications/Wallace et... · stratigraphy of the Northern Carnarvon Basin (Hocking, 1988;