Timing and genesis of base-metal mineralisation in black shales of the Upper Permian Ravnefjeld Formation, Wegener Halvø, East Greenland

ARTICLE

Mikael Pedersen Æ Jesper K. Nielsen

Adrian J. Boyce Æ Anthony E. Fallick

Timing and genesis of base-metal mineralisation in black shalesof the Upper Permian Ravnefjeld Formation, Wegener Halvø,East Greenland

Received: 10 September 2001 /Accepted: 7 March 2002 / Published online: 7 June 2002� Springer-Verlag 2002

Abstract Bituminous mud shales of the Upper PermianRavnefjeld Formation (Zechstein 1 equivalent) aremineralised with zinc, lead and copper within a ca.50 km2 area on Wegener Halvø in central East Green-land. The occurrence of base-metal sulphides in shalenodules cemented prior to compaction indicates an earlycommencement of base-metal mineralisation. In othercases, post-compactional sulphide textures are observed.Homogeneous lead isotope signatures of galena andsphalerite from the shales (206Pb/204Pb: 18.440–18.466;207Pb/204Pb: 16.554–16.586; 208Pb/204Pb: 38.240–38.326)suggest that all base metals were introduced during asingle hydrothermal event. Therefore, post-compac-tional textures are believed to result from recrystallisa-tion of early diagenetic sulphides during deep burial inthe Upper Cretaceous to Tertiary. Lead isotope signa-tures of galena hosted in Upper Permian carbonatebuild-ups are relatively heterogeneous compared tothose of the shale-hosted sulphides. The observed rela-tions indicate a shared lead source for the two types ofmineralisation, but different degrees of homogenisationduring mineralisation. This suggests that lead was

introduced to the carbonate rocks and black shalesduring two separate events. d34S of base-metal sulphidesin the Ravnefjeld Formation lie between –12 and –4&,whereas synsedimentary and early diagenetic pyrite inunmineralised shales in general have d34S between –47and –16.5&. Early diagenetic pyrite in the WegenerHalvø area in general has d34S 15 to 20& higher than thesame pyrite morphotype in Triaselv in the western partof the basin. This relatively high d34S can be explainedby extensive microbial sulphate reduction within per-sistent euxinic (super-anoxic) bottom waters underwhich supply of isotopically light seawater sulphate (anddisproportionation of intermediate sulphur compounds)was restricted. The sulphur in the base-metal sulphides isbelieved to represent sulphide-dominated pore water,enriched in 34S due to preferential removal of 32S bysulphate-reducing bacteria and precipitation of diage-netic pyrite in the near-seafloor environment. We sug-gest that the sulphide-dominated pore water wastrapped in the shale formation prior to introduction ofbase-metal-bearing fluids through fractures in the un-derlying carbonates, and that sulphide precipitationtook place when the two fluids met. d34S values of car-bonate-hosted base-metal sulphides fall within the samerange as the shale-hosted ones. The relationship betweenbarite and sulphides and evidence for pre-mineralisationentrapment of liquid hydrocarbons in the carbonatessuggest that the sulphide in this case is derived by in-situthermochemical sulphate reduction (TSR). Measuredfractionation between sulphide and sulphate rangesfrom 18.5 to 24.4&, suggesting temperatures of TSRaround 70 to 100 �C. Vitrinite reflectance measurementsin mineralised shale samples are all between 1.7 and2.0%, except for samples taken close to a Tertiary dykegiving ca. 3.0%. Vitrinite reflectance data are compa-rable to previously published data from unmineralisedshale samples in the area and could not be proven tocorrelate with the degree of mineralisation. This indi-cates that any early hydrothermal effect has been over-printed later, probably during deep burial in the LateCretaceous to Early Tertiary as previously proposed.

Mineralium Deposita (2003) 38: 108–123DOI 10.1007/s00126-002-0283-6

Editorial handling: J. Menuge

M. Pedersen (&)Geological Survey of Denmark and Greenland,Thoravej 8, 2400 Copenhagen NV, DenmarkE-mail: [email protected].: +45-33326575

J.K. NielsenGeological Institute, University of Copenhagen,Øster Voldgade 10, 1350 Copenhagen K, Denmark

A.J. Boyce Æ A.E. FallickScottish Universities Environmental Research Centre,Glasgow G75 0QF, Scotland, UK

Present address: M. PedersenNansensgade 75A, 4. tv.,1366 Copenhagen K, Denmark

Present address: J.K. NielsenDepartment of Geology,University of Tromsø, Dramsveien 201,9037 Tromsø, Norway

Keywords Sulphur isotopes Æ Lead isotopes ÆBase metals Æ Diagenetic mineralisation Æ Anoxia

Introduction

For centuries, Upper Permian transgressive sequences inlarge parts of Europe have been known to be highlyprospective with respect to base metals. In Poland andGermany, important Cu deposits are found togetherwith Pb and Zn in the Kupferschiefer (e.g. Jowett et al.1987a, 1987b; Vaughan et al. 1989; Puttmann et al. 1990;Oszczepalski 1999; Bechtel et al. 2000), and in easternEngland the time-equivalent Marl Slate and overlyingcarbonates of the Cadeby Formation contain abundantoccurrences of Pb, Zn, Cu, Fe, Ba and F (Turner et al.1978; Vaughan and Turner 1980; Harwood and Smith1986). Since the first discoveries of copper, lead andsilver on Wegener Halvø, East Greenland, by membersof the Danish Three-Year Expedition to East Greenlandbetween 1930 and 1934 (Eklund 1944; Koch 1955),Upper Permian carbonate rocks in central East Green-land have been known to be pervasively mineralised withbase metals, Ag and Ba. The Upper Permian sedimen-tary rocks later attracted much attention during explo-ration by Nordisk Mineselskab A/S, who located severalprospects. In 1968, a Kupferschiefer-type Zn–Pb–(Cu)occurrence was discovered in the bituminous mud shalesof the Upper Permian Ravnefjeld Formation in theWegener Halvø area near the eastern margin of theJameson Land basin. The find was reported in internalcompany reports (Lehnert-Thiel 1968; Lehnert-Thieland Walser 1968), and the results of subsequent follow-up programmes in 1969 and 1979 were published byThomassen (1973) and Harpøth et al. (1986). Interest inthe occurrence ceased, because too little high-grade orewas found. A few years later, however, the oil industrybecame interested in the Ravnefjeld Formation due to itspotential as a hydrocarbon source rock. Intense oil-related studies have since been carried out, leading to adetailed knowledge concerning the distribution, sedi-mentology, geochemistry, maturation patterns and pal-aeontology of the Ravnefjeld Formation (Surlyk et al.1984a, 1984b, 1986; Karlsen et al. 1988; Christiansenet al. 1990; Piasecki and Stemmerik 1991; Christiansenet al. 1992a, 1993a, 1993b).

The shales of the Ravnefjeld Formation were de-posited in a basin which was probably linked to thelarger depositional basin of Central Europe, where theKupferschiefer and Marl Slate were deposited (Maync1961). The widespread base-metal enrichment in UpperPermian units in many parts of the basin is striking, andis most likely a consequence of the coincidence of anumber of important factors common to all the areas ofmineralisation. Upper Permian sequences in all casesdefine the onset of marine sedimentation following along period of continental basin infilling, creating astratigraphic framework where excellent base-metalsource rocks are in contact with potential metal ‘traps’.

The importance of thick, continental sedimentary pilesas sources of base metals in East Greenland has beenshown by Pedersen (1997), and a similar conclusion wasreached for the Kupferschiefer deposits by Wedepohlet al. (1978) and Vaughan et al. (1989). The black shalesof the Kupferschiefer, Marl Slate and Ravnefjeld For-mation constitute a regionally important reducing en-vironment, where ascending metal-bearing brines fromthe red, continental clastic and volcanic rocks were likelyto precipitate base-metal sulphides during hydrothermalevents, with or without simultaneous introduction ofsulphur from Upper Permian or overlying Triassicevaporites.

The timing of mineralisation lies at the heart of thecontroversies concerning the genesis of black shale-hosted ores (Jowett et al. 1987a; Jowett 1991; Bechtelet al. 1999). However, during the last 20–30 years,diagenetic models have become prevalent and the dis-cussions since then have centred mainly on early versuslate diagenetic timing of mineralisation. The minerali-sation in the Ravnefjeld Formation was originallythought to be synsedimentary, due to the similarities tothe Kupferschiefer deposits and the then prevailinggenetic models. In this paper we will show that a syn-sedimentary origin for the mineralisation in theRavnefjeld Formation is unlikely, and will present newpetrographic observations as well as lead and sulphurisotopic data in support of an early diagenetic origin.

Geology and mineralisation

Basin evolution

The Jameson Land basin forms the southern part of theEast Greenland depositional basin, a more than 400-km-long, N-S-elongated basin situated within the Caledo-nian mountain chain (Fig. 1). Basin formation was ini-tiated in the Middle Devonian due to sinistral wrenchfaulting and extensional collapse of the Caledonianorogen (Larsen and Bengaard 1991). Post-Caledoniancrustal thinning in the Jameson Land area has beenextensive, and seismic investigations have revealed thatas much as 17 km of sedimentary rocks intruded byTertiary dykes and sills lies on a thin (6–8 km) crystal-line basement (Christiansen et al. 1992b; Larsen andMarcussen 1992).

The fill in the Jameson Land basin is interpreted to bedominated by up to 13 km of Devonian to LowerPermian continental clastic sediments and volcanic rocks(Larsen and Marcussen 1992). These have been de-formed during tectonic events in the Devonian and inmid-Permian times (Surlyk et al. 1984b). During theLate Permian, regional sea-level rise, combined withsubsidence related to thermal contraction of the crust,led to the establishment of marine conditions in the EastGreenland basin (Surlyk 1990). Two major transgres-sions can be discerned, the first of which resulted indeposition of carbonate platform sediments which were

109

subsequently karstified during subaerial exposure(Surlyk et al. 1986; Scholle et al. 1993). The secondmajor transgression led to growth of carbonate build-ups (Wegener Halvø Formation) on palaeotopographichighs along the basin margins, and deposition of blackshales of the Ravnefjeld Formation in the deeper partsof the basin and in inter-reef depressions (Surlyk et al.1986). Marine conditions continued into the lowermostTriassic, after which extensive uplift of the basin marginsled to rapid deposition of alluvial fan sediments on thebasin margin, and flood plain sediments in the basininterior (Clemmensen 1980). When uplift of the bor-derland ceased, the alluvial fans stopped growing andthe depositional environment changed into a valley withaeolian sedimentation in the north and west, and gyps-iferous playa lake conditions to the east. The whole arealater changed into a lake, which was periodically open tothe sea. By Late Triassic/Early Jurassic times, the cli-matic conditions changed from arid to humid (Surlyk

et al. 1981). The dominant red sediments characteristicof the Triassic were succeeded by deposition of black,organic-rich mudstones and sheet sandstones in a largewave- and storm-dominated anoxic lake with periodi-cally propagating delta fronts along the margins (Surlyket al. 1973; Dam and Surlyk 1992, 1993; Dam et al.1995). During the Early Jurassic, the sea transgressedthe southern part of the basin and marine conditionsprevailed during the remainder of the Jurassic andCretaceous.

During the Palaeocene, renewed rifting eventually ledto successful seafloor spreading and flood basalt vol-canism. The thick sedimentary package in the JamesonLand basin was during this period overlain by a north-wards-thinning series of flood basalts (Larsen andMarcussen 1992). These were later eroded away duringcrustal uplift and glaciation in the area, but thick floodbasalt units are still exposed south of Jameson Land,and minor occurrences can be found as much as 400 kmnorth of Jameson Land (Larsen et al. 1989). Seismicinvestigations have furthermore revealed sills up to300 m thick to be present in lower parts of the sedi-mentary sequence in the Jameson Land basin (Larsenand Marcussen 1992). Later in the Tertiary, a line ofintrusive alkaline centres was emplaced along the coastof East Greenland (Nielsen 1987), transecting the north-western part of the Jameson Land basin where vein-typelead-zinc and barium occurrences as well as a largeClimax-type Mo deposit were formed (Harpøth et al.1986; Schønwandt 1988).

Ravnefjeld Formation

Exposures of the Ravnefjeld Formation can be foundover an area of 80·400 km in East Greenland (Fig. 1).The formation consists of calcareous mud shales, whichcan be divided into two laminated and three bioturbatedunits (Piasecki and Stemmerik 1991). The shales inter-finger with the time-equivalent carbonate build-ups ofthe Wegener Halvø Formation (Surlyk et al. 1984a).

On Wegener Halvø, the shales of the RavnefjeldFormation are found in two different settings. Basinal-type shale sequences up to 60 m thick are found alongthe north-western coast of the peninsula, with the lam-inated facies being 12–20 m thick (Piasecki and Stem-merik 1991). In the central part, however, the shalesoften lie in karstic troughs cut down into the carbonatebuild-ups of the Karstryggen Formation. The shale se-quences in these troughs are of variable thickness andcontain abundant carbonate-rich horizons, representingresedimented build-up material. Similar horizons arealso found in the proximal basinal settings as far as 1 kmfrom the build-ups (Christiansen et al. 1993b).

The two laminated units in the Ravnefjeld Formationare very rich in organic material (average TOC: 3.8%)and sulphur (average TS: 2.0%), in contrast to thebioturbated units (TOC: 0.5%; TS: 1–2%; Piasecki andStemmerik 1991; Christiansen et al. 1993b). The

Fig. 1 Map of central East Greenland showing outcrops of UpperPermian rocks (modified after Piasecki and Stemmerik 1991). Thesquare encloses the Wegener Halvø area enlarged on Figs. 3 and 9

110

laminated units are interpreted as having been depositedunder regional anoxic episodes during maximum flood-ing at water depths in excess of 125 m (Surlyk et al.1986). It has been suggested that rapid growth of thefringing carbonate build-ups during times of maximumflooding, together with a stratified water column, createda ‘silled basin’ with restricted water circulation (Piaseckiand Stemmerik 1991).

Geology and general mineralisation patternsin the Wegener Halvø area

The overall geology of the Wegener Halvø area ischaracterised by a system of tilted, fault-boundedblocks. The stratigraphy is relatively simple, with theUpper Permian sediments lying unconformably onfolded and peneplaned Devonian and locally LowerPermian siliciclastic and volcanic rocks, overlain by aTriassic sequence consisting of shales, sandstones,evaporites, mudstones and minor limestones.

In Upper Permian times, the Wegener Halvø areaconsisted of a system of north-westward-tilted faultblocks on which carbonates of the Karstryggen andWegener Halvø formations were laid down (Fig. 2;Larsen et al. 1998). The Karstryggen Formation wasdeposited in a hypersaline environment during the firsttransgressive event and was later eroded during subae-rial exposure (Surlyk et al. 1986). When the sea trans-gressed the area again, reef-carbonates belonging to theWegener Halvø Formation grew preferentially on pal-aeotopographic highs. In deeper water in the north-western part of the area and in karstic troughs, mudshales of the Ravnefjeld Formation were deposited. TheUpper Permian sequence is in most of the area overlainby carbonate-rich sandstones of the Schuchert DalFormation.

Mineralisations are found in large parts of the stra-tigraphy in the Wegener Halvø–Devondal area (Fig. 3),with a general vertical zonation pattern going fromsulphide-bearing quartz veins in the Devonian rocks to

stratabound and stratiform sulphide mineralisation inthe Upper Permian and Triassic strata.

Stratabound occurrences of copper and barium withminor lead, zinc and fluorite are widespread in the car-bonate build-ups of the Wegener Halvø Formation onWegener Halvø and in Devondal (Harpøth et al. 1986).Mineralisation in the Ravnefjeld Formation geographi-cally coincides with these occurrences, but consistsmainly of lead and zinc (see below), with only minorcopper (increasing towards the north-west) and no bar-ite. The richest zones of mineralisation in both theWegener Halvø and Ravnefjeld Formations are confinedto the vicinity of the N–S-orientated Vimmelskaftetlineament (Fig. 3; Pedersen 1997). This lineament can betraced for 12 km and coincides in all localities withnarrow shale basins cut down into the underlying car-bonates. No faulting of Upper Permian rocks has beenobserved along the lineament, but a ca. 1 m thick Ter-tiary dyke is intruded along its entire length. Extensivemineralisation in the Ravnefjeld Formation is found inthe northern end of the lineament on Lille Ravnefjeld,whereas the Wegener Halvø Formation is stronglymineralised further south on Quensel Bjerg (Fig. 3). Nomineralisation has been found in the Schuchert DalFormation.

Minor but widespread occurrences of Cu, Ag, Pb andZn exist in almost all Triassic formations in the easternpart of the Jameson Land basin, including the WegenerHalvø area. Most of these are interpreted as diageneticdue to occurrences of ore minerals in sandstone cementsas well as in desiccation cracks (Harpøth et al. 1986).

Petrography of sulphides in theRavnefjeld Formation

Mineralisation within the mud shales of the RavnefjeldFormation is found in an area of about 50 km2 onWegener Halvø (Fig. 3; Nielsen and Pedersen 1998).Mineralisation is in most places confined to the lower-most metre of the shale sequence, immediately above the

Fig. 2 Schematic E-W crosssection through Wegener Halvøshowing the general distribu-tion of Upper Permian carbon-ate build-ups and shales(modified after Stemmerik et al.1997)

111

carbonates of the Wegener Halvø Formation. Aroundthe Vimmelskaftet lineament, however, significant con-centrations of galena and sphalerite can be found over ashale interval of several tens of metres within a deep,narrow shale basin.

Galena and sphalerite are the most common base-metal sulphides in the shales, with minor chalcopyriteoccurring especially in the north-western part of theWegener Halvø area. Sulphides are mainly concentratedin thin, carbonate-rich debris beds of resedimented ma-terial from adjacent build-ups where they replace calcitefossils and cement. Also, relatively carbonate-deficient,bituminous mud shales are found to contain galena andsphalerite. Mineralised layers are usually (but not al-ways) strongly lithified and stand out relative to thesurrounding shales. Mostly, this lithification is caused bycementation with low-Fe calcite, but also fine crystalline

silica, high-Fe calcite and saddle dolomite can beobserved occasionally. Galena and sphalerite mostlyreplace the low-Fe calcite cement and appear to be thelast phases to crystallise. The scarce occurrence of high-Fe calcite suggests that only insignificant amounts ofiron were introduced along with the base metals, whichis in accordance with the general scarcity of non-framboidal pyrite.

Disseminated aggregates (<5 mm) of sulphides areoccasionally found in calcareous nodules which have beencemented prior to compaction (Fig. 4). These aggregatesare noticeable in having highly irregular outlines andbeing either mono- or polysulphidic. In contrast to thecarbonate-rich replacement ores, pyrite is relativelyabundant in the nodules. In some samples, pyrite is theonly sulphide mineral but in others, composite sulphidesare found in which sphalerite, galena, euhedral pyrite andchalcopyrite can coexist within single aggregates. In suchcases, chalcopyrite is the latest crystallising phase,whereas the age relation between galena and sphalerite isimpossible to determine, as is their relation to non-framboidal pyrite, although the base-metal sulphidesengulf the small pyrite framboids (<10 lm). In yet othernodules, sphalerite, galena and chalcopyrite occur indisseminated, monomineralic aggregates.

Although apparently finely disseminated, many of thesulphides found in shale nodules are related to verticalmicrofractures (<1 mm), but they often occur up toseveral centimetres from the fractures. However, casesalso exist where no evidence of microfractures is seen,but where sulphides are finely disseminated throughoutnodules which have obviously been lithified during earlyburial history (Fig. 4). This strongly indicates an earlyintroduction of sulphides. By contrast, base-metalsulphides in fossil-rich beds mostly appear to havecrystallised late. No definitive evidence of post-comp-actional crystallisation has, however, been recorded insamples which are unaffected by Tertiary dykes.

The upper part of the Ravnefjeld Formation in theWegener Halvø area is generally unmineralised withrespect to base metals. In these shales abundant, smallframboidal pyrite (<10 lm) constitutes the only metalsulphide.

In the study area at Triaselv at the western marginof the basin, no trace of base-metal enrichment in theRavnefjeld Formation has been observed; pyrite is theonly metal sulphide present. The pyritic grains aregenerally larger than in the upper, unmineralised partof the formation in the Wegener Halvø area (<35 lm),but also here display the characteristic framboidal andeuhedral forms with the classical growth transitionbetween them, which is well known from both modernand ancient marine sediments (Sawlowicz 1993; Wilkinet al. 1996). Pyritisation of iron phyllosilicates andfossils are observable with restricted compaction de-formation inside carbonate concretions, and thereforeassigned to an initial formation during the depositionand earliest diagenesis (this study; Nielsen and Peder-sen 1998).

Fig. 3 Generalised geological map of the Wegener Halvø area withthe location of mineralised samples (>500 ppm combined Cu, Pband Zn) marked by circles

112

Samples and analytical methods

Mineralised Upper Permian and Triassic samples from WegenerHalvø have been examined primarily with respect to isotopic (Pband S) composition. In addition, samples of diagenetic pyrite fromunmineralised shale samples at Triaselv and Lille Cirkusbjerg(Figs. 1 and 3) have been analysed for S-isotope ratios in order toestablish the ‘background’ d34S values in relation to which themineralisation sulphur should be discussed. New vitrinite reflec-tance data from mineralised shale samples at Wegener Halvø arealso reported.

Lead isotopes were measured in a total of 21 galena and onesphalerite samples from the Ravnefjeld and Wegener Halvø For-mations (Table 1). The galena was sampled using a binocular mi-croscope during drilling of single grains. The samples weredissolved in HCl+HNO3 and the Pb separated using standard HBrion-exchange chemistry. Nine of the galena samples were measuredin a Finnigan MAT-261 multiple-collector mass spectrometer,whereas the other samples were analysed by Geochron Laborato-ries using a VG 54 multi-collector mass spectrometer. Fractiona-tion correction was in both laboratories determined from multipleruns of the NBS 981 standard to be close to 0.12%/amu.

Sulphur isotopic composition of sulphides from the RavnefjeldFormation shales was measured at the Scottish Universities Envi-ronmental Research Centre (SUERC; Tables 2 and 3). Sampleswere run by either conventional or the in-situ laser combustionmethod, depending on grain size. The laser combustion techniqueof Kelley and Fallick (1990) was in most cases preferred in order todifferentiate the d34S of the various sulphide phases due to the fine-grained, disseminated nature of the shale-hosted sulphides. Pol-ished blocks were inserted into a laser port, which was evacuatedand subsequently filled with an excess of oxygen gas (Fallick et al.1992). Typically, a spot area of approximately 100 lm diameter ofthe respective sulphide minerals was combusted using a SpectronLasers 902Q CW Nd:YAG laser (1 W power), operating in TEM00

mode (Fallick et al. 1992). The SO2 gas produced by the laser

combustion was purified in a vacuum line, which operates similarlyto a conventional sulphur extraction line (Kelley and Fallick 1990).Determination of the S-isotope composition of the purified SO2 gaswas carried out online by a VG Sira II gas mass spectrometer. Rawgas d34S data were corrected for mineral-dependent fractionationfactors following Kelley and Fallick (1990), and the results aregiven in Tables 2 and 3. Conventional S-isotope analyses of sul-phates and sulphides were performed following the standardtechnique of Coleman and Moore (1978) and Robinson andKusakabe (1975) respectively. Coarse-grained sulphides were hand-picked for this method, whereas framboidal pyrite was extractedfrom shale samples by treating the crushed samples with dichlo-romethane and methanol for an hour, and subsequently withboiling 10% HCl for another hour. Reproducibility of all analyticalresults was controlled through replicate measurements of the in-ternational standards NBS-123 (+17.1&) and IAEA-S-3 (–31&),as well as SUERC’s internal laboratory standard CP-1 (–4.6&). AllS-isotope compositions were calculated relative to Canon DiabloTroilite (V-CDT), and are reported in standard notation. The an-alytical precision, based on replicate analyses of the standards, wasaround ±0.2&.

Maturity of organic matter in 12 mineralised samples was pe-trographically determined by random reflectance measurements onthe vitrinite (Table 4). The equipment was a Leitz MPV-SP system,which was calibrated against a standard of 1.667% Ro. Between 32and 115 measurements were made on each sample in monochro-matic light and oil immersion.

Lead isotope data

Galena from the Ravnefjeld Formation plots inside anarrow field (averages: 206Pb/204Pb=18.447; 207Pb/204Pb=15.569; 208Pb/204Pb=38.278; n=12; Fig. 5). The dataplot inside the general field defined by East Jameson Landgalena lead (Jensen 1994; Pedersen 2000), close to the

Fig. 4A–C Cross sectionthrough mineralised shale nod-ule. A The bending of internallayering as well as B the pres-ence of well-preserved shellfragment in the nodule suggestthat lithification took placeduring early diagenesis. C Oneof many sulphide aggregateswhich are found disseminatedin the shale nodule. Introduc-tion of mineralising fluids intothe nodule has been impossibleafter cementation, and sulp-hides are assumed to predatecementation and thereforecompaction. Cp Chalcopyrite,Gn galena, Sp sphalerite

113

upper end of the feldspar lead trend defined by Pedersen(2000). The feldspar lead trend is interpreted by Pedersen(2000) to represent a mixing trend between more radio-

genic lead derived from Caledonian granite feldspar andless radiogenic lead from pre-Caledonian basement inJameson Land. In this model, the Pb in the Ravnefjeld

Table 1 Lead-isotope ratios ofgalena and sphalerite from theRavnefjeld and Wegener HalvøFormations. Ga Galena, Spsphalerite, * analyses fromJensen (1994)

Formation Sample No. Mineral 206Pb/204Pb 207Pb/204Pb 208Pb/204Pb

Wegener Halvø(Upper Permian)

403260 Ga 18.454 15.617 38.421403267 Ga 18.407 15.586 38.285403272 Ga 18.398 15.562 38.215428258 Ga 18.468 15.592 38.3687903808* Ga 18.405 15.568 38.2317903812* Ga 18.435 15.586 38.2997903813* Ga 18.570 15.583 38.3677903814* Ga 18.459 15.569 38.2827901914* Ga 18.461 15.578 38.2977901915/A* Ga 18.465 15.572 38.308

Ravnefjeld (Upper Permian) 428240 Ga 18.465 15.569 38.288428243 Ga 18.454 15.574 38.296428253 Ga 18.446 15.567 38.266428263 Ga 18.466 15.570 38.307428277 g Ga 18.445 15.564 38.256428277 s Sp 18.465 15.586 38.326428291 Ga 18.443 15.554 38.240428294 Ga 18.440 15.560 38.259428901 Ga 18.432 15.564 38.2457901931* Ga 18.432 15.572 38.2697901932/A* Ga 18.467 15.582 38.3157901933* Ga 18.433 15.571 38.268

Table 2 Sulphur-isotope ratiosof base-metal sulphides andpyrite from mineralised samplesof the Ravnefjeld and WegenerHalvø Formations. Ga Galena,Sp sphalerite, Py pyrite, Babarite, C conventional analysis,L laser analysis

Formation Sample No. Mineral d34Scdt(&)

Analysismethod

Wegener Halvø (Upper Permian) 403258 Ga –5.6 C403260 Ga –4.8 C403272 Ga –9.7 C403270 Ba +13.7 C428285 Ba +14.7 C

Ravnefjeld (Upper Permian) 428240 Ga –10.9 L428269 Ga –10.2 L428269 Ga –10.4 L428273 Ga –9.7 L428273 Ga –9.2 L428277 Ga –7.3 L428291 Ga –8.1 L428291 Ga –7.2 L428901 Ga –4.9 L428901 Ga –6.1 L428901 Ga –5.5 L428901 Ga –5.9 L428254 Sp –9.3 L428268 Sp –12.2 C428273 Sp –8.9 L428273 Sp –4.4 L428273 Sp –4.6 L428277 Sp –11.1 L428277 Sp –10.2 L428291 Sp –10.6 L428901 Sp –4.0 L428901 Sp –5.1 L428901 Cp –4.9 L428240 Py –13.5 L428243 Py –15.7 C428250 Py –10.0 C428256 Py –18.8 L428288 Py +1.7 L428901 Py –6.0 L

Pingo Dal (Lower Triassic) 403291 Ga –8.9 L

114

Formation appears to have been derived mainly fromleaching of Caledonian granites or sedimentary productsderived from breakdown of such granites.

The lead isotopic homogeneity of galena from theRavnefjeld Formation on Wegener Halvø indicates thatlead was introduced from a well-mixed hydrothermalsystem, probably during a single mineralising event. Theslightly larger spread in lead-isotope ratios from galena inthe carbonates of the Wegener Halvø Formation (Fig. 6)excludes the possibility that the carbonate-hosted leadwas exclusively derived by remobilisation of lead in theblack shales, as suggested by Harpøth et al. (1986). Thedata show that the two mineralisation types have derivedmost of their lead from either a common source or fromsources of similar composition, but that the carbonate-

hosted lead stems from a less homogenised hydrothermalcell than is the case with the shale-hosted lead. The dis-crepancy in lead-isotope patterns suggests that at leastpart of the lead was introduced to the Wegener HalvøFormation in either a separate event or within one ormore separate phases of the same event which causedmineralisation of the Ravnefjeld Formation.

Sulphur isotope data

Unmineralised horizons

Early diagenetic pyrite in unmineralised RavnefjeldFormation shales was analysed in samples from the

Table 3 Sulphur-isotope ratiosof framboidal pyrite from theRavnefjeld Formation atTriaselv and Lille Cirkusbjerg.C Conventional analysis, Llaser analysis

Formation Locality SampleNo.

d34Scdt(&)

Analysismethod

Ravnefjeld(Upper Permian)

Triaselv(GGU core 303102)

1-1 –42.6 L2 –34.0 L2 –34.4 L

219 –47.0 L40 –26.1 L40 –41.8 L40 –39.3 L40 –44.1 L27 –37.4 L25 –18.7 L

201 –43.0 L30 –41.3 L

215 –31.4 C12 –41.7 C11 –35.3 C

Lille Cirkusbjerg(GGU core 303109)

151 –25.2 C154 –29.3 C157 –25.0 C160 –23.6 C163 –28.5 C167 –21.4 C170 –21.7 C174 –20.3 C175 –22.1 C177 –21.6 C180 –17.6 C181 –16.5 C182 –17.4 C185 –19.4 C189 –16.9 C193 –21.4 C

Table 4 Vitrinite reflectance,base metals and TOCmeasurements of mineralisedRavnefjeld Formation samplesfrom Wegener Halvø

Sample No. Vit. Reflectance TOC Zn Pb Cu(%) (%) (ppm) (ppm) (ppm)

428240a 3.00 1.78 3,300 6,650 154428240b 2.98 – – – –428246 3.02 2.87 87 1,490 261428253 1.72 1.23 483 1,210 11428256 1.81 0.95 383 39 9428266 1.85 1.25 10,600 5,390 994428279 1.76 0.65 141 23 8428282 1.69 1.22 18 42 222428287 2.03 1.10 1,950 2,120 40428291 1.88 3.98 1,200 2,630 232428294 1.65 1.72 1,460 7,340 28428901 1.67 1.94 2,260 6,150 463

115

upper part of the formation at Lille Cirkusbjerg (Fig. 3),as well as in samples from Triaselv in western JamesonLand (Fig. 1) where the sediments are slightly morecoarse grained and thermally immature (Christiansenet al. 1990, 1993b). d34S ranges from –47.0 to –16.5& inthese samples (Table 3). This is within the same generalrange as sulphides from the time-equivalent Kupfers-chiefer shales of Poland (Jowett et al. 1991; Bechtel et al.2001), but with a tendency for a smaller degree of 34Sdepletion when mean values are taken into consider-ation. The sulphur in pyrite from the Lille Cirkusbjergis distinctly heavier (–29.3 to –16.5&; mean=–21.7±3.9&, n=16) than the sulphur found in Triaselvsamples (–47.0 to –18.7&; mean=–37.2±7.5&, n=15;Fig. 7; Nielsen et al. 1999). In spite of the relativelyisotopically heavy diagenetic pyrite at Lille Cirkusbjerg,the extent of fractionation between these and contem-poraneous Upper Permian seawater (Fig. 7; ca. 11&), asestimated from gypsum d34S in this basin and elsewhere(Claypool et al. 1980; Clemmensen et al. 1985;Stemmerik et al. 1988; Kampschulte et al. 1998), is be-tween 28 and 40&, which is consistent with open-sys-tem, bacterial sulphate reduction. This deviates from thegeneral fractionation range of 37 to 58& in Triaselv.The isotopic difference between the two areas suggests amore closed-basin environment in the Wegener Halvøarea during deposition of the Ravnefjeld Formation,involving a more persistent development of euxinicbottom water. This is consistent with the stratified watercolumn/silled basin model for the deposition of theRavnefjeld Formation of Piasecki and Stemmerik(1991). In such environments, the system imposes a limiton the reoxidation and disproportionation reactions(Jørgensen 1990; Canfield and Thamdrup 1994), whichare necessary to obtain the more extensive isotopicfractionations seen in the Triaselv area (Nielsen et al.1999). There is an abundance of framboidal pyrite inLille Cirkusbjerg, and a lack of overprinting of theseframboids, which suggests that the relatively isotopicallyheavy values here were achieved by one-time fractiona-tion of seawater sulphate. One-time fractionation is apossibility at the transition to euxinic bottom waters,resulting in small framboidal pyrite which, due to theirdensity, fall through the water column relatively fast(Wilkin et al. 1996). A minor amount of small euhedra,derived under sulphate depletion in the pore water(Wilkin and Barnes 1997), might well have contributedto some 32S depletion.

Mineralised horizons

Sulphur isotopic compositions of galena, sphalerite andchalcopyrite from mineralised Ravnefjeld Formationshales define a relatively narrow range from –12.2 to–4.0& with a mean of –7.9& (±2.6&, n=23; Fig. 7).These negative numbers contrast the generally positived34S values of sulphides in base-metal occurrences inWest Jameson Land (8.3±3.8&), as pointed out by

Fig. 5 Lead isotopic composition of galena from the RavnefjeldFormation (open circles) plotted against the feldspar lead trend ofPedersen (2000), and the general East Jameson Land ore lead fieldof Jensen (1994) and Pedersen (2000). The orogene model growthcurve is from the ‘plumbotectonic’ model of Zartman and Doe(1981)

Fig. 6 Close-up of lead isotope diagrams in Fig. 5. Note the well-defined field of galena from the Ravnefjeld Formation as comparedto galena from the Wegener Halvø Formation

116

Pedersen and Fougt (1998). On a more detailed sul-phur-isotope diagram (Fig. 8), two modes are suggest-ed; one at –11 to –10&, and the other from –6 to –4&.The latter is partly caused by seven analyses of varioussulphides from a single sample, which all group in theinterval from –6.1 to –4.0&. The mode between –11and –10&, therefore, represents the majority of

samples. No systematic sulphur-isotope variation existsbetween galena and sphalerite, and attempts to calcu-late isotopic equilibrium temperatures from co-existingmineral pairs in all cases give unrealistic results, indi-cating equilibrium was not established if these pairswere co-precipitated, or that they were precipitated atdifferent times.

Pyrite in Ravnefjeld Formation shales from minera-lised horizons is significantly isotopically heavier thanpyrite from framboidal pyrite in unmineralised shales,but generally lighter than the base-metal sulphides. Thetotal range is from –18.8 to +1.7& (mean=–10.4±6.8&; n=6; Fig. 7; Table 2). Pyrite with theisotopically heaviest sulphur (often sub- to euhedralgrains) is found in mineralised samples, whereas thesample giving –18.8& is a pyritic framboid seam whichis clearly of early diagenetic origin. The sulphur-isotopecomposition suggests formation of the pyrite within thepore waters where sulphate depletion prevailed, reflect-ing a sequential precipitation of pyrite (Sælen et al. 1993;Wilkin and Barnes 1997). This is consistent with theregular observation of euhedra formation confined tothe pore waters of modern marine environments (Wilkinand Barnes 1997).

Three samples of galena from the carbonate reefsunderlying the black shales, and one sample of galenafrom overlying Triassic sandstone were analysed. Thecarbonate-hosted galena range between d34S –9.7 and–4.8&, and the Triassic-hosted galena gives –8.9&.Also, two samples of vein barite from a mineralisedzone in the carbonates of the Wegener Halvø Forma-tion were analysed, giving d34S values of +13.7 and+14.7&.

Vitrinite reflectance data

Vitrinite reflectance was measured in 12 mineralisedshale samples. Carbonate-rich shales and packstones ingeneral contained too little vitrinite for this method, andonly samples rich in organic matter gave satisfactoryresults. Due to the relatively high reflectance of vitrinitein all samples, bireflectance was observed, giving rise toa large spread. However, a more important explanationfor the spread is that all samples were taken on thesurface, meaning that surficial oxidation may have takenplace, which can also be seen by the alteration of pyriteto goethite in some samples.

All analysed samples have Ro above 1.6%, which isconsistent with earlier published maturity data from thearea (Karlsen et al. 1988; Christiansen et al. 1990). Onlytwo samples gave values around 3.0%. These sampleswere taken very close to the Vimmelskaftet lineament,and the high maturity is most likely caused by the Ter-tiary dyke emplaced in the lineament. No correlationexists between vitrinite reflectance and grade of miner-alisation or various base-metal ratios (Table 4). Fur-thermore, no systematic geographical patterns can beobserved (Fig. 9).

Fig. 7 Summary histogram showing the sulphur isotopic compo-sition of base-metal sulphides in the Ravnefjeld Formation incomparison with pyrite from mineralised and unmineralised shales,and Upper Permian and Triassic evaporites. (evaporite data arefrom Clemmensen et al. 1985, and Stemmerik et al. 1988)

117

Discussion

The unimodal sulphur and lead isotopic homogeneity forbase-metal sulphides supports the hypothesis that a sin-gle event was responsible for mineralisation in the Rav-nefjeld Formation on Wegener Halvø. Importantly,strong petrographic evidence suggests that at least somemetals were introduced to the sediment prior to deepburial and compaction, as demonstrated above. Thus, ifall metals were introduced during one event, this musthave taken place relatively early in the history of theshales, probably in the Upper Permian. This implies thatthe post-compactional sulphide textures, which can beobserved in some samples, may well be caused by remo-bilisation of earlier formed sulphides. This is not unlikely,because all post-compactional textures have been foundwithin the Vimmelskaftet lineament, which was ther-mally affected during dyke emplacement in the Tertiary.

The relatively narrow, d34S range could be explainedby introduction of sulphide from the same well-mixedhydrothermal system which carried the lead. This,however, seems unlikely, since reduced sulphur cannotbe transported in considerable amounts together withbase metals under hydrothermal conditions (Seward andBarnes 1997). The slightly higher d34S found in the base-metal sulphides in the Ravnefjeld Formation as com-pared to early diagenetic pyrite can be explained by oneof two models.

1. Thermochemical reduction of sulphate introduced tothe non-compacted shales from beneath in the samefluids which carried the metals, and in-situ precipi-tation of base-metal sulphides. Assuming that sea-water sulphate played a role, and taking the generald34S values of Upper Permian gypsum from the basininto consideration (ca. 11&), the fractionation wouldbe around 19&, which corresponds well with TSR ataround 100 �C (Machel et al. 1995; Machel 2001).

2. Incorporation of a residual sulphide component ininterstitial pore water in the shales shortly after de-position. During consumption of sulphate in bacte-riogenic systems closed with respect to sulphate, the

d34S of the residual sulphate increases as a result ofthe preferential incorporation of 32S into the productsulphide. Under such circumstances, the d34S of dis-solved pore-water sulphide also increases.

Coleman and Raiswell (1981) demonstrated that d34Svalues of pyrite increased from –26 to –2& during

Fig. 8 Detailed histogramshowing the distribution of d34Svalues of base-metal sulphidesfrom the Ravnefjeld Formation

Fig. 9 Distribution of vitrinite reflectance measurements onWegener Halvø

118

progressive concretionary growth in Jurassic shales fromEngland. Similarly, Melezhik et al. (1998) demonstrateda positive correlation between the degree of diagenesisand d34S values (–8 to +24&) of pyrite and pyrrhotitein a sulphide concretion from black shales in the Pech-enga Greenstone Belt. These d34S ranges are significantlyhigher than the range observed in this study, whichcould be an argument for the TSR model. On the otherhand, the size of the d34S range is the result of a numberof factors. It could be argued that, if most pore-watersulphur was in a reduced state when the metalliferousfluids entered the shales, the S isotopes would be more orless evenly distributed, and the differentiation of sulphurwhich entered the single ore mineral grains would beminimised. Residual pore-water-dominated regimes arethe result of exhaustion of reactive iron in euxinic sedi-ments, where active bacterial reduction persists duringearly diagenesis (Raiswell 1997). The dominance of iron-lean carbonate in the Ravnefjeld Formation prior to oresulphide deposition indicates that such a regime waslikely in these sediments.

The mineralisation in the Ravnefjeld Formation wasinterpreted by Harpøth et al. (1986) to be of primary,sedimentary origin. Given the discussion above, and thesubstantially higher d34S in galena and sphalerite com-pared to diagenetic pyrite, we believe this model is nowuntenable. The fact that the mineralisation is confined tothe lowermost metres of the formation in most localitiesfurthermore suggests that metals were introduced frombeneath after deposition of the shales, which may not beexpected in a synsedimentary model.

An important aspect of the mineralisation history inthe Wegener Halvø area is the relationship of the shale-hosted mineralisation to other sulphide occurrences inthe area. In an attempt to elucidate this relationship,samples of galena from the carbonate reefs underlyingthe black shales, and from an overlying Triassic sand-stone were analysed. These results (d34S –9.7 to –4.8&)fall within the general range defined by base-metalsulphides from the Ravnefjeld Formation. This couldsuggest remobilisation of sulphide from the shales,which would not be unlikely considering the regionalheating event which has affected the area (see below).The widespread occurrence of vein barite in connectionwith the carbonate-hosted sulphides, however, indicatesthat the sulphides represent a dissolved sulphate pre-cursor which has been reduced locally in the carbonates.Stemmerik (1991), through microscopic studies, foundevidence that liquid hydrocarbons were trapped in thecarbonate build-ups at the time when the hydrothermalsolutions were introduced. This is ascribed to deepburial in the Late Cretaceous to Early Tertiary, and isheld responsible for migration of liquid hydrocarbonsfrom the Ravnefjeld Formation and into the carbonatebuild-ups of the Wegener Halvø Formation. Regionalheating of the area by volcanic activity in the EarlyTertiary probably generated hydrothermal activity(Stemmerik 1991). Stemmerik et al. (1997) suggestedthat in-situ thermochemical reduction of dissolved

sulphate in the presence of these organic compounds wasa likely mechanism leading to sulphide precipitation.Some of the vein barite was analysed in this study to testthis. The d34S values of the barite (+13.7 and +14.7&)are slightly heavier than Upper Permian and Triassicevaporitic gypsum in the basin (Clemmensen et al. 1985;Stemmerik et al. 1988). Because there is negligible frac-tionation of d34S between the mineralising solution andprecipitating sulphate minerals (Ohmoto and Goldhaber1997), the d34S values of these barite samples reflect thed34S of the sulphate in the hydrothermal brine. Thismeans that a fractionation between 18.5 and 24.4& hastaken place during reduction, which would be in ac-cordance with low-temperature (�100�C) thermochem-ical sulphate reduction (TSR; Machel et al. 1995;Machel 2001). A temperature range of TSR around 70to 100�C is suggested using the equation of Kiyosu andKrouse (1990).

In a regional maturity study of the RavnefjeldFormation, Christiansen et al. (1990) found the entireWegener Halvø area to be post-mature in contrast toneighbouring areas. They ascribed this high maturationlevel to deep burial caused by extensive block-faulting inthe area, and Mathiesen et al. (1995) later modelled thetime of deepest burial to be Early Tertiary. We re-inves-tigated this subject in order to evaluate a possible hydro-thermal effect of an early diagenetic mineralisation eventon the black shales, as it is known from recent hydro-thermal systems that widespread alteration of organicmatter accompanies hydrothermal activity on and nearthe seafloor (Simoneit et al. 1987; Simoneit 1995).

Most mineralised samples were taken within thelowermost few metres of the shale sequence, and shouldbe more affected by upwelling hydrothermal fluids thanshales higher in the sequence. The general overlap be-tween our Ro values (Fig. 9) and the Ro values ofChristiansen et al. (1990) therefore suggest a regionalburial event to be responsible for the high maturation,rather than localised input of hydrothermal fluids. Thissupports the conclusions of Christiansen et al. (1990).Any thermal effect caused by the hydrothermal fluidshas been overprinted by this Early Tertiary late event.

The data presented in this study could comply with amodel in which ca. 100�C sulphate- and metal-bearingwater encountered the Upper Permian sediments frombeneath shortly after shale deposition, leading to TSR inboth the carbonates and the shales. This, however,contradicts the petrographic evidence in favour of a LatePermian age for the mineralisation in the RavnefjeldFormation, and a Late Cretaceous to Early Tertiary agefor the carbonate-hosted occurrences. It could be arguedthat an early hydrothermal event could lead to genera-tion of hydrocarbons earlier than proposed by Stem-merik (1991). This would eventually migrate out into theadjacent carbonate build-ups. Cu- and Ba-dominatedmineralisation would form in the relatively oxidisingenvironment in the build-ups, with sulphide formationbeing the result of in-situ TSR utilising the trappedhydrocarbons. In the lower part of the Ravnefjeld

119

Formation shales, TSR would on the other hand resultfrom consumption of organic matter leading to Pb- andZn-dominated mineralisation. We do not, however, as-sume that the model of Stemmerik (1991) has to bechanged. Separate introductions of the base metals inthe Wegener Halvø and Ravnefjeld formations mayseem to immediately contradict evidence gained fromfield observations, including striking geographic coinci-dence between the two types of mineralisation, andconfinement of the zone of most intense mineralisationto the Vimmelskaftet lineament in both cases. The dis-crepancy, however, may be explained by consideringsome simple factors. Firstly, both carbonate build-upsand black shales are excellent traps for base metals, ifthe correct conditions are present. As the two rock typesare exposed in the same general area, the geographicoverlap of mineralisation is not remarkable. The im-portance of the Vimmelskaftet lineament during twoseparate hydrothermal events is also likely as other, in-dependent field evidence shows that it has been an im-portant structural element in at least two periods(Pedersen 1997). The coincidence of the lineament withthe central axis of carbonate troughs in several localitiessuggests that the lineament was already present as a zoneof weakness during Upper Permian subaerial erosionprior to subsidence and shale deposition. No faulting ofthe Upper Permian strata has been observed along themain lineament, but the occurrence of a synsedimentaryfault in the shales close to the lineament in one localityindicates some movement during the Upper Permian.Upper Permian synsedimentary faulting was also no-ticed on the northern fault-block of Wegener Halvø bySeidler (2000). Furthermore, the occurrence of a basalticdyke along the entire exposed part of the lineament (ca.12 km) suggests that it has been a significant structure inthe Tertiary as well, along which dyke intrusion couldtake place. We therefore believe this lineament had along-lived influence on the development of this area,from at least Permian to Tertiary times, and we specu-late that hydrothermal fluids could have moved alongthe lineament at any time prior to emplacement of theTertiary dyke.

Mineralisation model

We have argued that the base metals in the WegenerHalvø and Ravnefjeld formations were likely introducedduring two separate mineralising events. The main hy-drothermal cell was in both cases focused in the Vim-melskaftet lineament, but the solutions were redistributedto a larger area. We envisage the following model.

Stage 1a

Deposition of Ravnefjeld shale in a semi-closed euxinicbasin. Debris flow incursion of carbonate detritus frommarginal build-ups adds to organic matter preserved inthe sequence.

Stage 1b

Early diagenetic framboidal pyrite deposition occursutilising bacteriogenic sulphide from an open systemsulfuretum shortly after sediment is deposited. d34S av-erages around–22& for pyrite of theWegenerHalvø area.

Stage 2

Early pyrite deposition leads to depletion of reactiveiron in the sediments, whilst bacterial sulphate reductioncontinues, using abundant organic matter and pore-water sulphate. Early, low-Fe calcite cements indicate alack of mobile iron. A pore-water sulphide reservoirwith relatively high d34S (averaging –8&) builds up inthe sediment as the bacterial reduction system becomesmore closed with respect to sulphate.

Stage 3

Incursion of hydrothermal fluids bearing Zn, Pb andsome Fe from below, preceding the carbonate concre-tion cementation episode, and prior to full compaction.Close association of sulphides with the Vimmelskaftetlineament suggests this was the active, controllingstructure. Hydrothermal fluids either (1) mix with pore-water sulphide to precipitate the ore sulphides, utilisingfractures in the carbonates of the Wegener Halvø For-mation (Fig. 10), or (2) precipitate sulphides from dis-solved metals and sulphate being reduced bythermochemical sulphate reduction by reaction with theorganic material in the lower part of the RavnefjeldFormation. The absence of a reducing agent within thecarbonates most likely prevented precipitation of base-metal sulphides within the carbonate build-ups at thistime. Sulphides were, however, precipitated in the lowerpart of the Ravnefjeld Formation when the upwellingbrines encountered sulphide-dominated pore watertrapped within the uncompacted shales and carbonatedebris flows. A higher permeability/porosity of thesecarbonate debris flows (shelly horizons) could have en-hanced storage of pore-water sulphide, via local organicmatter decomposition, and allowed incursion of thehydrothermal fluids within these horizons, simulta-neously with the general lithification of these sediments.

The thickening of the ore horizons around the Vim-melskaftet lineament probably indicates tectonic per-meability creation near the lineament, as opposed to thedominant sediment permeability which allowed the ‘fa-vourable horizon’ mode of sulphide deposition awayfrom the structure.

Stage 4

Carbonate-hosted sulphide derived from the thermo-chemical sulphate reduction route at this later stage.

120

Deep burial in the Late Cretaceous to Early Tertiary ledto migration of liquid hydrocarbons from the RavnefjeldFormation into the carbonate build-ups of the WegenerHalvø Formation. Regional heating of the area byTertiary volcanic activity probably facilitated hydro-thermal activity. Metalliferous fluids probably alsocontaining sulphate from the Triassic evaporites wereagain focussed along the Vimmelskaftet lineament andother major faults. Copper and barium were the majormetals in these Tertiary fluids, with lead and zinc beingof minor importance. When the hydrothermal solutionsencountered the trapped hydrocarbons, in-situ thermo-chemical sulphate reduction took place and causedprecipitation of chalcopyrite, while remaining sulphateprecipitated as barite.

Stage 5

Tertiary dyke emplacement caused local recrystallisationof sulphides, especially around the Vimmelskaftet lin-eament.

Conclusion

Lead-isotope evidence indicates that the metals in theRavnefjeld Formation shales were introduced during asingle event. The occurrence of base-metal sulphideswithin shale nodules formed prior to sediment compactionshows that mineralisation commenced close to the sea-water/sediment interface. The sulphur-isotope data areconsistent with the sulphide being derived from the localpore water after utilisation of the isotopically lightestbacteriogenic sulphur by precipitation of diagenetic py-rite just below the seafloor.

Different lead-isotope patterns in galena from theRavnefjeld Formation and the underlying WegenerHalvø Formation suggest that the two types of miner-alisation belong to two separate events, and the dis-tinctive metal characteristics (Pb+Zn versus Cu+Ba)may reflect this. The metal enrichment of the carbonatesprobably took place during a regional heating event inthe Tertiary, when in-situ thermochemical sulphate re-duction and sulphide precipitation took place whenheated metal- and sulphate-bearing brines encounteredliquid hydrocarbons trapped in the carbonate build-ups.Thus, the Wegener Halvø area is a fine example of howmultiple types of mineralisation can form in the samearea, during several events where the prevailing condi-tions vary.

Acknowledgements The major part of our research was supportedthrough the TUPOLAR project ‘Resources of the sedimentarybasins of North and East Greenland’, funded by The DanishNatural Science Research Council. The work benefited from inte-gration with contemporary studies of the Geological Survey ofDenmark and Greenland and the Geological Institute, Universityof Copenhagen. SUERC is funded by NERC and the ScottishUniversities consortium. AJB is funded by NERC support of theIsotope Community Support Facility at SUERC. The authors aregrateful to Sven M. Jensen for access to unpublished lead-isotopedata and for constructive comments on the manuscript, which ispublished with the permission of the Geological Survey of Den-mark and Greenland. We also thank Julian Menuge and AchimBechtel for their careful reviews.

References

Bechtel A, Elliot WC, Wampler JM, Oszczelpalski S (1999) Claymineralogy, crystallinity, and K-Ar ages of illites within thePolish Zechstein basin: implications for the age of the Kup-ferschiefer mineralization. Econ Geol 94:261–272

Bechtel A, Gratzer R, Puttmann W, Oszczepalski S (2000) Geo-chemical and isotopic composition of organic matter in theKupferschiefer of the Polish Zechstein Basin: relation to ma-turity and base metal mineralization. Int J Earth Sci 89:72–89

Bechtel A, Sun YZ, Puttmann W, Hoernes S, Hoefs J (2001) Iso-topic evidence for multi-stage base metal enrichment in the

Fig. 10 A Schematic section showing the structural framework ofthe Wegener Halvø block in Upper Permian time. B The proposedpathways for hydrothermal fluids. The mineralisation event ispossibly caused by subsidence of the Wegener Halvø block in theUpper Permian. This has focused overpressured, metal-bearingbasinal brines up along the Vimmelskaftet lineament from wherethey were redistributed out into fractures in carbonate build-ups ofthe Wegener Halvø Formation. Base-metal sulphides were precip-itated when the upwelling fluids encountered sulphide-dominatedpore water in the lower part of the Ravnefjeld Formation

121

Kupferschiefer from the Sangerhausen Basin, Germany. ChemGeol 176:31–49

Canfield DE, Thamdrup B (1994) The production of 34S-depletedsulfide during bacterial disproportionation of elemental sulfur.Science 266:1973–1975

Christiansen FG, Piasecki S, Stemmerik L (1990) Thermal ma-turity of the Upper Permian succession in the Wegener Halvøarea, East Greenland. Rapp Grønlands Geol Unders 148:109–114

Christiansen FG, Dam G, Piasecki S, Stemmerik L (1992a) A re-view of Upper Palaeozoic and Mesozoic source rocks fromonshore East Greenland. In: Spencer AM (ed) Generation,accumulation and production of Europe’s hydrocarbons II.Springer, Berlin Heidelberg New York, Spec Publ Eur AssocPetrol Geosci 2:151–161

Christiansen FG, Larsen HC, Marcussen C, Hansen K, Krabbe H,Larsen LM, Piasecki S, Stemmerik L, Watt WS (1992b) Upliftstudy of the Jameson Land basin, East Greenland. Norsk GeolTidsskr 72:291–294

Christiansen FG, LarsenHC,MarcussenC, Piasecki S, Stemmerik L(1993a) Late Paleozoic plays in East Greenland. In: Parker JR(ed) Petroleum geology of northwest Europe. Proc 4thConfGeolSoc Lond, pp 657–666

Christiansen FG, Piasecki S, Stemmerik L, Telnæs N (1993b) De-positional environment and organic geochemistry of the UpperPermian Ravnefjeld Formation source rock in East Greenland.AAPG Bull 77:1519–1537

Claypool GE, Holser WT, Kaplan IR, Sakai H, Zak I (1980) Theage curve of sulfur and oxygen isotopes in marine sulfate andtheir mutual interpretation. Chem Geol 28:199–260

Clemmensen LB (1980) Triassic rift sedimentation and palaeoge-ography of central East Greenland. Bull Grønlands Geol Un-ders 136:1–72

Clemmensen L, Holser WT, Winter D (1985) Stable isotope studythrough the Permian-Triassic boundary in East Greenland.Geol Soc Denmark Bull 33:253–260

Coleman ML, Moore MP (1978) Direct reduction of sulphate tosulphur dioxide for isotopic analysis. Anal Chem 50:1594–1595

Coleman ML, Raiswell R (1981) Carbon, oxygen and sulphurisotope variations in concretions from the Upper Lias of N.E.England. Geochim Cosmochim Acta 45:329–340

Dam G, Surlyk F (1992) Forced regressions in a large wave-dom-inated and storm-dominated anoxic lake, Rhaetian-SinemurianKap Stewart Formation, East Greenland. Geology 20:749–752

Dam G, Surlyk F (1993) Cyclic sedimentation in a large wave- andstorm-dominated anoxic lake; Kap Stewart Formation (Rhae-tian-Sinemurian), Jameson Land, East Greenland. In: Posa-mentier HW, Summerhayes CP, Haq BU, Allen GP (eds)Sequence stratigraphy and facies associations. Spec Publ IntAssoc Sediment 18:419–438

Dam G, Surlyk F, Mathiesen A, Christiansen FG (1995) Explo-ration significance of lacustrine forced regressions of theRhaetian-Sinemurian Kap Stewart Formation, Jameson Land,East Greenland. In: Steel RJ et al. (eds) Sequence stratigraphyof the northwest European margin. NPF Spec Publ 5. Elsevier,Amsterdam, pp 511–527

Eklund O (1944) Malmgeologisk forskning pa Nordostgronland.Tekn Tidsskr Stockholm 14:393–401

Fallick AE, McConville P, Boyce AJ, Burgess R, Kelley SP (1992)Laser microprobe stable isotope measurements on geologicalmaterials: Some experimental considerations (with special ref-erence to d34S in sulphides). Chem Geol 101:53–61

Harpøth O, Pedersen JL, Schønwandt HK, Thomassen B (1986)The mineral occurrences of central East Greenland. MeddrGrønland Geosci 17:1–139

Harwood GM, Smith FW (1986) Mineralization in Upper Permiancarbonates at outcrop in eastern England. In: Harwood GM,Smith DB (eds) The English Zechstein and related topics. GeolSoc Spec Publ 22:103–111

Jensen SM (1994) Lead isotope systematics of ore deposits andmineral occurrences in East and North-East Greenland. PhDThesis, University of Aarhus

Jørgensen BB (1990) A thiosulfate shunt in the sulfur cycle ofmarine sediments. Science 249:152–154

Jowett EC (1991) The evolution of ideas about the genesis ofstratiform copper-silver deposits. Econ Geol Monogr 8:117–132

Jowett EC, Pearce GW, Rydzewski A (1987a) A Mid-Triassic pa-leomagnetic age of the Kupferschiefer mineralization in Poland,based on a revised apparent polar wander path for Europe andRussia. J Geophys Res 92 B1:581–598

Jowett EC, Rydzewski A, Jowett R (1987b) The KupferschieferCu-Ag ore deposits in Poland: a re-appraisal of the evidence oftheir origin and presentation of a new genetic model. Can JEarth Sci 24:2016–2037

Jowett EC, Roth T, Rydzewski A, Oszczepalski S (1991) ‘‘Back-ground’’ d34S values of Kupferschiefer sulphides in Poland:pyrite-marcasite nodules. Miner Deposita 26:89–98

Kampschulte A, Buh D, Strauss H (1998) The sulfur and strontiumisotopic composition of Permian evaporites from the Zechsteinbasin, northern Germany. Geol Rundsch 87:192–199

Karlsen DA, Leythaeuser D, Schaefer RG (1988) Light hydrocar-bon redistribution in a shallow core from the Ravnefjeld For-mation on the Wegener Halvø, East Greenland. Org Geochem13:393–398

Kelley SP, Fallick AE (1990) High precision spatially resolvedanalysis of d34S in sulphides using a laser extraction technique.Geochim Cosmochim Acta 54:883–888

Kiyosu Y, Krouse HR (1990) The role of organic acid in the abi-ogenic reduction of sulfate and the sulfur isotope effect. Geo-chem J 24:21–27

Koch L (1955) Report on the expeditions to central East Greenland1926–39. Meddr Grønland 143(2):1–642

Larsen P-H, Bengaard H-J (1991) Devonian basin initiation in EastGreenland: a result of sinistral wrench faulting and Caledonianextensional collapse. J Geol Soc Lond 148:355–368

Larsen HC, Marcussen C (1992) Sill-intrusion, flood basalt em-placement and deep crustal structure of the Scoresby Sund re-gion, East Greenland. In: Storey BC, Alabaster T, PankhurstRJ (eds) Magmatism and the causes of continental break-up.Geol Soc Spec Publ 68:365–386

Larsen LM, Watt WS, Watt M (1989) Geology and petrology ofthe Lower Tertiary plateau basalts of the Scoresby Sund region,East Greenland. Bull Grønlands Geol Unders 157:1–164

Larsen M, Piasecki S, Preuss T, Seidler L, Stemmerik L, TherkelsenJ, Vosgerau H (1998) Petroleum geological activities in EastGreenland in 1997. Geol Greenland Surv Bull 180:35–42

Lehnert-Thiel K (1968) Mitteilungen uber einen permischen Kup-ferschiefer Ostgronlands, 72� nordlicher Breite. NordiskMineselskab A/S, Internal Note 3(68) (in GEUS archive)

Lehnert-Thiel K, Walser P (1968) Montangeologischen Berichtuber die Gebiete der Wegener Halbinsel, das nord-westlicheGebiet des Canning-Landes und den SE-Teil der Traill Ø.Nordisk Mineselskab A/S, Internal Rep 5(68) (in GEUS ar-chive)

Machel HG (2001) Bacterial and thermochemical sulfate reductionin diagenetic settings – old and new insights. Sediment Geol140:143–175

Machel HG, Krouse HR, Sassen R (1995) Product and distin-guishing criteria of bacterial and thermochemical sulfate re-duction. Appl Geochem 10:373–389

Mathiesen A, Christiansen FG, Bidstrup T, Marcussen C, Dam G,Piasecki S, Stemmerik L (1995) Modelling of hydrocarbongeneration in the Jameson Land basin, East Greenland. FirstBreak 13:329–341

Maync W (1961) The Permian of Greenland. In: Raasch GO (ed)Geology of the Arctic 1. University of Toronto Press, pp 214–223

Melezhik VA, Grinenko LN, Fallick AE (1998) 2000-Ma sulphideconcretions from the ‘Productive’ Formation of the PechengaGreenstone Belt, NW Russia: genetic history based on mor-phological and isotopic evidence. Chem Geol 148:61–94

Nielsen JK, Pedersen M (1998) Hydrothermal activity in the UpperPermian Ravnefjeld Formation of central East Greenland – astudy of sulphide morphotypes. Geol Greenland Surv Bull180:81–87

122

Nielsen JK, Shen Y, Canfield DE (1999) Morphology and sulphurisotope composition of early diagenetic pyrite. In: ArmannssonH (ed) Geochemistry of the Earth’s surface. Balkema, Rotter-dam, pp 335–337

Nielsen TFD (1987) Tertiary alkaline magmatism in East Green-land: a review. In: Fitton JG, Upton BGJ (eds) Alkaline igne-ous rocks. Geol Soc Spec Publ 30:489–515

Ohmoto H, Goldhaber MB (1997) Sulfur and carbon isotopes. In:Barnes HL (ed) Geochemistry of hydrothermal ore deposits.Wiley, New York, pp 517–611

Oszczepalski S (1999) Origin of the polymetallic mineralization inPoland. Miner Deposita 34:599–613

Pedersen M (1997) Investigation of ore potential in black shales ofthe Upper Permian Ravnefjeld Formation in Scoresby Landand on Traill Ø, central East Greenland. Danmarks GrønlandsGeol Unders Rapp 1997(124):1–17

Pedersen M (2000) Lead isotope signatures of sedimentary rocks asa tool for tracing ore lead sources: a study of base metal andbarite occurrences in Jameson Land Basin, central EastGreenland. Trans Inst Mineral Metall B, Appl Earth Sci109:49–59

Pedersen M, Fougt H (1998) Mineralisation diversity in an area ofintense crustal thinning: S-isotope evidence from base metaloccurrences in East Greenland. In: Abstr Vol Annu MeetMineral Deposits Studies Group, St. Andrews, Scotland, 14–15December 1998. St. Andrews University

Piasecki S, Stemmerik L (1991) Late Permian anoxia in central EastGreenland. In: Tyson RV, Pearson TH (eds) Modern and an-cient continental shelf anoxia. Geol Soc Spec Publ 58:275–290

Puttmann W, Heppenheimer H, Diedel R (1990) Accumulation ofcopper in the Permian Kupferschiefer: a result of post-deposi-tional redox reactions. Org Geochem 16:1145–1156

Raiswell R (1997) A geochemical framework for the application ofstable sulphur isotopes to fossil pyritization. J Geol Soc Lond154:343–356

Robinson BW, Kusakabe M (1975) Qualitative preparation ofsulphur dioxide for 34S/32S analyses from sulphides by com-bustion with cuprous oxide. Anal Chem 47:1179–1181

Sælen G, Raiswell R, Talbot MR, Skei JM, Bottrell SH (1993)Heavy sedimentary sulfur isotopes as indicators of super-anoxicbottom-water conditions. Geology 21:1091–1094

Sawlowicz Z (1993) Pyrite framboids and their development: a newconceptual mechanism. Geol Rundsch 82:148–156

Scholle PA, Stemmerik L, Ulmer-Scholle D, Di Liegro G, HenkFH (1993) Palaeokarst-influenced depositional and diageneticpatterns in Upper Permian carbonates and evaporites, Kars-tryggen area, central East Greenland. Sedimentology 40:895–918

Schønwandt HK (1988) Geology and geotectonic setting of cra-tonic porphyry molybdenum deposits in the North Atlanticregion. In: Boissonnas J, Omenetto P (eds) Mineral depositswithin the European Community. Springer, Berlin HeidelbergNew York, pp 210–229

Seidler L (2000) Incised submarine canyons governing new evi-dence of Early Triassic rifting in East Greenland. PalaeogeogrPalaeoclimatol Palaeoecol 161:267–293

Seward TM, Barnes HL (1997) Metal transport by hydrothermalore fluids. In: Barnes HL (ed) Geochemistry of hydrothermalore deposits, 3rd edn. Wiley, New York, pp 435–486

Simoneit BRT (1995) Organic matter transformations in hydro-thermal systems – an overview. In: Pasava J, Krubek B, Zak K(eds) Mineral deposits: from their origin to their environmentalimpacts. In: Proc 3rd Biennial SGA Meet, Prague, 28–31 Au-gust 1995. Balkema, Rotterdam, pp 13–19

Simoneit BRT, Grimalt JO, Hayes JM, Hartman H (1987) Lowtemperature hydrothermal maturation of organic matter insediments from the Atlantis II Deep, Red Sea. Geochim Cos-mochim Acta 51:879–894

Stemmerik L (1991) Reservoir evaluation of Upper Permian build-ups in the Jameson Land basin, East Greenland. Rapp Grøn-lands Geol Unders 149:1–23

Stemmerik L, Rouse JE, Spiro B (1988) S-isotope studies of shal-low water, laminated gypsum and associated evaporites, UpperPermian, East Greenland. Sediment Geol 58:37–46

Stemmerik L, Jensen SM, Pedersen M (1997) Hydrocarbon-asso-ciated mineralisation of Upper Permian carbonate build-ups,Wegener Halvø, East Greenland. In: Hendry JP, Carey P,Parnell J, Ruffell A, Worden R (eds) Ext Abstr Vol Int ConfFluid Evolution, Migration and Interaction in SedimentaryBasins and Orogenic Belts, Geofluids II ’97, Belfast, 10–14March 1997. Queen’s University of Belfast, pp 485–487

Surlyk F (1990) Timing, style and sedimentary evolution of LatePalaeozoic-Mesozoic extensional basins of East Greenland. In:Hardman RFP, Brooks J (eds) Tectonic events responsible forBritain’s oil and gas reserves. Geol Soc Spec Publ 55:107–125

Surlyk F, Callomon JH, Bromley RG, Birkelund T (1973) Stra-tigraphy of the Jurassic – Lower Cretaceous sediments ofJameson Land and Scoresby Land, East Greenland. BullGrønlands Geol Unders 105:1–76

Surlyk F, Clemmensen L, Larsen HC (1981) Post-Palaeozoic evo-lution of the East Greenland continental margin. In: Kerr JW,Fergusson AJ, Machan LC (ed) Geology of the North Atlanticborderlands. Can Soc Petrol Geol Mem 7:611–645

Surlyk F, Hurst JM, Marcussen C, Piasecki S, Rolle F, Scholle PA,Stemmerik L, Thomsen E (1984a) Oil geological studies in theJameson Land basin, East Greenland. Rapp Grønlands GeolUnders 120:85–90

Surlyk F, Piasecki S, Rolle F, Stemmerik L, Thomsen E, Wrang P(1984b) The Permian basin of East Greenland. In: Spencer AM,Johnsen SO, Moerk A, Nysaether E, Songstad P, Spinnangr A(eds) Petroleum geology of the north European margin. Gra-ham and Trotman, London, pp 303–315

Surlyk F, Hurst JM, Piasecki S, Rolle F, Scholle PA, Stemmerik L,Thomsen E (1986) The Permian of the western margin of theGreenland Sea – a future exploration target. In: Halbouty MT(ed) Future petroleum provinces of the world. AAPG Mem40:629–659

Thomassen B (1973) Heavy metal content of the Posidonia shaleson Wegener Halvø. Rapp Grønlands Geol Unders 58:33–35

Turner P, Vaughan DJ, Whitehouse KI (1978) Dolomitization andthe mineralization of the Marl Slate (N.E. England). MinerDeposita 13:245–258

Vaughan DJ, Turner P (1980) Diagenesis, magnetization andmineralization of the Marl Slate. Contrib Sediment 9:73–90

Vaughan DJ, Sweeney M, Friedrich G, Diedel R, Haranczyk C(1989) The Kupferschiefer: an overview with an appraisal of thedifferent types of mineralization. Econ Geol 84:1003–1027

Wedepohl KH, Delevaux MH, Doe BR (1978) The potential sourceof lead in the Permian Kupferschiefer bed of Europe and someselected Paleozoic mineral deposits in the Federal Republic ofGermany. Contrib Mineral Petrol 65:273–281

Wilkin RT, Barnes HL (1997) Pyrite formation in an anoxic estu-arine basin. Am J Sci 297:620–650

Wilkin RT, Barnes HL, Brantley SL (1996) The size distribution offramboidal pyrite in modern sediments: an indicator of redoxconditions. Geochim Cosmochim Acta 60:3897–3912

Zartman RE, Doe BR (1981) Plumbotectonics – the model. Tec-tonophys 75:135–162

123