Lead isotope constraints on the origin of Cenozoic orogenic gold systems in the Southern Alps and northwestern Otago, South Island, New Zealand

Abstract

Lead isotopic compositions were determined for sulphides from Pliocene-Pleistocene gold-bearing veins in the Alpine Schist and from Miocene gold-bearing veins and vein breccias from the Shotover-Macetown area in the northwest Otago Schist belt. The lead isotopic signatures are consistent with derivation of Pb in the vein minerals predominantly from metasedimentary rocks that underlie the region. Differences in Pb isotopic signatures between deposits are interpreted to result from lateral and vertical lithological variability within the source rock mass. The host rocks also contain metabasic rocks with N-MORB, E-MORB or within-plate basalt chemistry. However, the observed Pb isotopic signatures in the gold-bearing veins preclude incorporation of significant amounts of Pb from the metabasites. The Pb isotopic signatures of lamprophyre dikes that were intruded into the Otago Schist coeval with Miocene gold mineralisation are distinctly more radiogenic than those of the hydrothermal veins. Thus, although the lamprophyre dikes were emplaced into similar extensional structural sites to the gold-bearing veins, there was no genetic relationship between lamprophyre dikes and gold mineralisation.

Keywords:

• Alpine Schist
• gold
• greywacke
• lamprophyre
• lead isotopes
• orogenic
• Otago Schist
• Southern Alps

Introduction

The sources of fluid and metals for formation of orogenic gold deposits have been widely debated (Spooner 1993; Bierlein & Crowe 2000; Goldfarb et al. 2005). Definition of fluid source(s) is problematic because of the extensive fluid-rock interaction that occurs during passage through host rocks for the deposits (Bierlein & Crowe 2000; Goldfarb et al. 2005). Determination of the source of metals is probably more significant than fluid source for mineral exploration, as some authors suggest that distinctive rock types or igneous bodies contribute most metals for gold deposits (Henley et al. 1976; Rock & Groves 1988; Spooner 1993; Bierlein et al. 2006). Verification of this concept would make the geological environs of these metal-yielding rocks more prospective. However, most orogenic gold deposits occur in relatively old terranes (Bierlein & Crowe 2000; Goldfarb et al. 2001 , 2005), where the context of formation of the deposits has been obscured by erosion and tectonic overprinting. This incomplete record inhibits deduction of the relationships between gold deposits and their metal sources.

The active collisional tectonic zone of the South Island of New Zealand, the Southern Alps, provides an opportunity to examine and test some hypotheses of the origin of metals in orogenic gold deposits, as the zone hosts small but widespread young (Miocene and Pliocene-Pleistocene) gold deposits (Fig. 1). These deposits show little post-formational geological overprinting, other than minor exhumation, that might obscure the tectonic context in which they formed. The host rocks for the mineralised systems are relatively uniform in composition, with distinctive end-member rock types. Hence, this young orogenic belt is ideal to examine metal sources in relation to known available rock types. In addition, there is no evidence for plutonic activity beneath the mineralised part of the orogen; the influence of intrusion-derived metals that is commonly invoked for some orogenic gold deposits (Rock & Groves 1988; Spooner 1993; Goldfarb et al. 2005) can therefore be discounted in the Southern Alps.

Fig. 1  Location map of the South Island showing the topography of the Southern Alps (DEM from geographx.co.nz), the current plate tectonic setting and the locations of gold-bearing vein systems along the mountains. Specific mineralised localities mentioned in this study are indicated and described in Table 1 and the text (MDFZ: Main Divide Fault Zone).

There are three well-defined potential sources for metals in gold deposits in the Southern Alps. Metagreywacke forms the bulk of the orogen, and metamorphism and devolatilisation of this rock in deeper levels of the orogen has been suggested as a source for the metals and fluid involved in gold mineralisation (Pitcairn et al. 2006). Metamorphism of similar metagreywackes has also been suggested as a source for metals during Mesozoic orogenic mineralisation in the nearby Otago Schist belt (Fig. 2) (e.g., Paterson 1986; Mortensen et al. 2010). In contrast, metabasic rocks have been considered by some workers to be potentially fertile sources for metals in orogenic mineralising systems (Bierlein et al. 2006) including the Otago Schist (Henley et al. 1976; Bierlein & Craw 2009). Metabasic rocks are a minor but significant component of the Southern Alps (Fig. 2). Alternatively, lamprophyre dikes have been suggested as contributors of metals to gold deposits because of a close spatial relationship in many orogenic gold systems (e.g., Rock & Groves 1988; Rock et al. 1989). An extensive swarm of lamprophyre dikes occurs near the southern end of the Southern Alps and were emplaced coevally with some gold deposits, so a genetic relationship has to be considered.

Fig. 2  Geological map showing the principal lithotectonic terranes in the South Island, focusing on the rocks which underlie the Southern Alps (Fig. 1). A belt of schists containing abundant metabasites and scattered ultramafic pods is shown extending from Aspiring Terrane to the Pounamu belt (after Cooper 1976), although this continuity is not well established.

Tracing of the source(s) of gold in epigenetic occurrences is difficult; however, the isotopic composition of Pb contained in sulphide minerals within the gold-bearing veins provides a useful proxy for gold in metal mobilising systems (e.g., Tosdal et al. 1999; Mortensen et al. 2008). In this paper, we have determined the Pb isotopic composition of sulphides associated with gold in several gold-bearing deposits through the Southern Alps, representing a range of different structural settings and age. We use these isotopic compositions to distinguish between the principal potential contributing rock types for metals in mineralising systems in the Southern Alps.

Regional geology

Otago Schist

The South Island is mainly underlain by a broad Permian-through-middle-Cretaceous accretionary wedge, the Torlesse composite terrane (Fig. 2), which accumulated during prolonged subduction under the margin of Gondwana (Mackinnon 1983). The accretionary wedge consists primarily of turbiditic quartzofeldspathic greywackes, with rare metabasite horizons (Mackinnon 1983). The accretionary wedge comprises an older (Permian to Late Triassic) part on the South Island called the Rakaia Terrane, and a younger (Triassic to mid-Cretaceous) part on the North Island and northeast South Island called the Pahau Terrane (Fig. 2) (e.g., Mackinnon 1983; Mortimer 2004). The Rakaia Terrane was tectonically juxtaposed against a package of more volcanogenic greywackes of the Caples Terrane to the south in the Jurassic, resulting in the Otago Schist metamorphic belt at and near the boundary (Fig. 2) (Turnbull 1979; Mortimer 1993). A slice of metasedimentary rocks that contains abundant metabasites (locally up to 10%), together with metacherts and rare ultramafic pods, is also incorporated into the Otago Schist belt in this area. Metagreywacke that makes up the bulk of this package display Rakaia Terrane geochemical affinities (Mortimer 1993). However, the package is sufficiently different from the other portions of the Rakaia Terrane in the Otago Schist belt (in terms of lithology as well as structural and metamorphic history) that it has been mapped as the separate Aspiring Terrane (Fig. 2) (Craw 1984 , 1998; Norris & Craw 1987). Metabasite horizons within the Aspiring Terrane typically range from 1 to 100 m in thickness and are traceable along-strike for tens of kilometres. There is a close spatial association between these metabasites and metachert horizons (1–100 m thick) and ultramafic pods (metre scale).

The Otago Schist was pervasively recrystallised and foliated during Jurassic-Cretaceous deformation and metamorphism that accompanied the juxtaposition of the various lithotectonic assemblages within the accretionary wedge. The foliation is generally flat-lying or shallow-dipping (<20°) over most of the belt. Metamorphic grade in the Otago Schist ranges from pumpellyite-actinolite facies on the northeast and southwest flanks to upper greenschist facies (garnet-biotite-albite) within the core of the belt (Bishop 1972; Mortimer 2000). Schists in the northwest part of the belt are in the core of a broad, shallow, west-northwest trending antiformal structure that mainly exposes complexly deformed, upper greenschist facies units of the Aspiring Terrane (Fig. 2). Metamorphic cooling ages from the Otago Schist belt range from c. 148 Ma to c. 110 Ma (e.g., Gray & Foster 2004), indicating that exhumation of this region was mainly finished by middle Cretaceous time.

Initiation of a new plate boundary through the South Island in the early Tertiary resulted in tectonic subsidence and faulted basin formation (Turnbull et al. 1975; Norris et al. 1978), culminating in complete marine inundation in the Oligocene (Landis et al. 2008). Many basin-bounding normal faults were reactivated as reverse faults as regional deformation evolved to transpressional in the Miocene. One such reactivated tectonic feature is the Moonlight Fault, a major tectonic structure which cuts the Otago Schist (Fig. 1, 2) (Turnbull et al. 1975; Cooper et al. 1987). The compressive deformation associated with the Moonlight Fault resulted in widespread development of upright mesoscopic and macroscopic folds of the schist foliation up to 10 km away from the fault, and localised preservation of faulted and folded slivers of Oligocene marine sediments (Turnbull et al. 1975; Craw et al. 2006).

A swarm of lamprophyre dikes (Alpine Dikes of Cooper 1986; Cooper et al. 1987 or Westland Dikes of Barreiro & Cooper 1987) was intruded into the southern portion of the Southern Alps in the Miocene (c. 22–28 Ma; Fig. 1) (Cooper 1986; Cooper et al. 1987). Lamprophyre emplacement was related to the development of folds and extensional fractures associated with the Moonlight Fault during the inception of the Alpine Fault as a through-going plate boundary structure (Cooper et al. 1987). Most dikes, especially in the southern portion of the swarm, fill a conjugate set of extensional fractures that strike east and southeast and have steep dips (Craw 1985; Cooper et al. 1987). Dikes at the northern end of the swarm also fill northeast-striking fractures that are parallel to the structural grain of the host rock, imposed by tight folding of the foliation at the northern extension of the Moonlight Fault zone (Cooper et al. 1987; Craw et al. 2003). Dike emplacement was accompanied by locally extensive ankeritic alteration and fenitisation of host rocks (Cooper 1986).

Alpine Schist

The Alpine Schist is a belt of schistose rocks that were exhumed by Miocene to Recent uplift to the southeast of the Alpine Fault (Fig. 2). This uplift began in the Miocene in the northwest corner of the Otago Schist, where mainly Aspiring Terrane rocks were exhumed (Craw 1995). The locus of uplift evolved northeastwards through the Pliocene, and schists beneath the Torlesse composite terrane were exhumed along the Alpine Fault. Metamorphic grade ranges from pumpellyite-actinolite near the Main Divide of the Southern Alps to amphibolite facies adjacent to the Alpine Fault. Metamorphic zones are steeply dipping and metamorphic boundaries are folded and faulted (Cox et al. 1997; Craw 1998). A prominent fault zone, the Main Divide Fault Zone, generally forms the boundary between low grade Alpine Schist and non-schistose metagreywacke of the Torlesse composite terrane (Cox & Findlay 1995).

Most of the Alpine Schist has been derived from metasedimentary rocks with Rakaia Terrane geochemical affinities. However, zones rich in metabasites and metacherts occur locally in the central part of the Alpine Schist (Craw et al. 1994) and amphibolite facies metabasites are common along the Alpine Fault. Likewise, there is a prominent belt of metabasites and ultramafic pods, including the Pounamu belt (Cooper & Reay 1983) in the northeastern part of the Alpine Schist (Fig. 2). Hence, although individual mafic and ultramafic bodies are generally only traceable on the 0.1–10 km scale, a zone of schist containing these bodies is traceable from the Aspiring Terrane in the Otago Schist northeastwards through the Alpine Schist (Fig. 2) (Cooper 1976).

The structural and metamorphic fabrics in the Alpine Schist are largely inherited from Mesozoic deformation, similar to that which affected the Otago Schist. The Alpine Schist, however, has been variably overprinted by younger deformation and dynamothermal metamorphism that post-dates the youngest metamorphism (c. 110 Ma) that affected the Otago Schist. The nature and timing of these younger metamorphic events are not well understood. Garnet grade metamorphic mineral assemblages in various portions of the Alpine Schist have given U-Pb, Sm-Nd and Lu-Hf ages that range from c. 100 to c. 70 Ma (Mortimer & Cooper 2004; Vry et al. 2004). Volumetrically minor anatectic pegmatite dikes in the Alpine Schist have also given U-Pb ages of c. 70 Ma (Chamberlain et al. 1995; Batt et al. 1999). This prolonged Middle to Late Cretaceous metamorphism and associated deformation is interpreted as related to the collision and subsequent underthrusting of the Hikurangi oceanic plateau (Vry et al. 2004). Post-Miocene deformation becomes progressively more intense from faulting near the Main Divide, through upright folding in greenschist facies, to the development of new foliations in upper greenschist and amphibolite facies. Most K-Ar and ⁴⁰Ar/³⁹Ar ages for the Alpine Schist, especially near the Alpine Fault, are very young (mainly <6 Ma) and are interpreted to reflect post-metamorphic exhumation and cooling during initiation and continued transpressive displacement along the Alpine Fault (e.g., Chamberlain et al. 1995; Batt et al. 2000; Batt 2001). A detailed analysis of K-Ar, ⁴⁰Ar/³⁹Ar and fission track age data from the Alpine Schist (Batt et al. 2000; Batt 2001) demonstrates that metamorphic cooling ages for micas decrease towards the Alpine Fault from c. 15 Ma at the Main Divide. Mica cooling ages range from c. 5 Ma to c. 1 Ma close to the Alpine Fault. There is a strong topographic control on exhumation ages, with older ages coming from higher elevations (Cox & Findlay 1995; Teagle et al. 1998).

The Southern Alps host an extensive active hydrothermal system driven by collisional orogenic processes (Craw et al. 2002; Craw & Campbell 2004). This hydrothermal system includes surface hot springs and numerous and widespread vein swarms, including the auriferous veins that are the topic of this study (Fig. 1). Veins formed from a mixture of meteoric and crustally exchanged waters at 200–350°C and a range of depths from near-surface to>6 km. Deep-sourced crustally exchanged fluids are an important component of gold-bearing fluids in this hydrothermal system (Cox et al. 1997; Craw et al. 2002; Campbell et al. 2004; Craw 2006).

Methods

A representative set of samples from auriferous vein systems in the Southern Alps (Craw 2006) was selected for this study to include material from a wide variety of structural settings from south to north. These sample sites are indicated in Fig. 1. Their nature and structural settings are summarised in Table 1, and their geological context is described in the following sections. Sulphide minerals were separated from clean unweathered sulphide-rich samples. Sample preparation, geochemical separations and Pb isotopic measurements were conducted at the Pacific Centre for Isotopic Geochemical Research at the University of British Columbia, using methods as described in Mortensen et al. (2008). All measured isotopic ratios were corrected for instrumental mass fractionation of 0.12%/amu based on repeated measurements of the NBS 981 standard and the values recommended by Thirwall (2000). Errors were numerically propagated throughout all calculations and are reported at the 2σ level (Table 2).

Table 1  Geological setting of vein material from which sulphide samples were taken for Pb isotope analyses in the Otago and Alpine schists

	Copper Creek	Macetown	Smyth Glacier	Callery River	Whataroa River	Fiddes vein
Metals	Au, Sb, As	Au, Sb, As	Au, Sb, As	Au, W, As	Au, As	Au, As, Cu
Style	Vein, breccia, ankeritic alteration	Vein, breccia, ankeritic alteration	Vein, breccia, ankeritic alteration	Veinlets, disseminated, greenschist alteration	Vein, breccia, ankeritic alteration	Stringer veins, minor breccia, little alteration
Attitude	NW strike, 60° NE dip	NW strike, steep dip	E to NE strike, vertical	NE strike, variable dip	NE strike, steep dip	N strike, 55° W dip
Host rock	Micaceous schist, greenschist facies	Micaceous schist, greenschist facies	Pumpellyite-actinolite facies of schist	Biotite zone micaceous schist	Pumpellyite-actinolite facies of schist	Prehnite-pumpellyite facies metagreywacke
Host terrane	Aspiring	Aspiring	Rakaia	Rakaia	Rakaia	Rakaia
Local structural setting	Associated with Miocene folding of Oligocene marine sediments and schist foliation, and Moonlight Fault	Crosscut Miocene folds of schist foliation, associated with Moonlight Fault	Swarm of small structures containing ankeritic carbonate, cuts across steeply dipping schist foliation	Cut across Alpine Schist folds of schist foliation	Cut schist foliation associated with faults that juxtapose rocks of differing metamorphic grade	Cut metagreywacke slices in active Main Divide Fault Zone
Regional structural setting	Folds and extensional zones related to inception of Alpine Fault in early Miocene	Folds and extensional zones related to inception of Alpine Fault in early Miocene	Associated with Pliocene-Recent faults that affect topographic development of mountains	Host rock exhumed from metamorphic depths and folded in Quaternary	In hanging wall of Main Divide Fault Zone	In vein swarm that cuts across metamorphic zones uplifted along Alpine Fault, near Hope Fault intersection
Lamprophyre relationship	Veins fill same structures as Miocene lamprophyres to N	Veins fill same structures as Miocene lamprophyres to N	None apparent; 20 km from main locus of Miocene intrusion	None	None	None
Crosscutting structures	Late stage fractures	Late stage fractures	Late stage fractures	Quaternary adularia-quartz veins	Late stage fractures	Active Main Divide Fault Zone
References	Cooper et al. (1987); Craw et al. (2006)	Cooper et al. (1987); Craw (2006)	Craw et al. (2003); Craw (2006)	Craw et al. (1987); Teagle et al. (1998)	Craw et al. (1994); Craw & Campbell (2004)	Becker et al. (2000)

Table 2  Lead isotopic compositions of sulphide minerals from gold-bearing vein occurrences in the Otago and Alpine schists (errors given at a 2σ level)

Sample	Mineral	^206Pb/^204Pb	Error (%)	^207Pb/^204Pb	Error (%)	^208Pb/^204Pb	Error (%)	^207Pb/^206Pb	Error (%)	^208Pb/^206Pb	Error (%)
Copper Creek	sb	18.713	0.11	15.633	0.15	38.560	0.19	0.8354	0.051	2.0606	0.092
Copper Creek	sb	18.773	0.09	15.657	0.14	38.738	0.18	0.8340	0.046	2.0635	0.091
Copper Creek	py	18.727	0.06	15.630	0.07	38.613	0.10	0.8347	0.039	2.0620	0.050
Macetown	py	18.740	0.1	15.646	0.09	38.674	0.14	0.8349	0.06	2.0637	0.089
Fiddes	asp	18.724	0.07	15.652	0.09	38.611	0.10	0.8360	0.019	2.0622	0.038
Smyth Glacier	sb	18.842	0.27	15.719	0.27	38.950	0.28	0.8343	0.052	2.0672	0.050
Callery-1	py	18.689	0.16	15.642	0.16	38.526	0.17	0.8370	0.022	2.0615	0.041
WB-1	py	18.739	0.07	15.641	0.08	38.602	0.10	0.8347	0.019	2.0601	0.038
WB-2	py	18.816	0.07	15.705	0.08	38.801	0.10	0.8346	0.020	2.0621	0.038
WB-3	py	18.785	0.04	15.671	0.06	38.714	0.08	0.8342	0.023	2.0609	0.039

Gold-bearing veins in the Southern Alps

Gold mineralisation accompanied deformation of the Otago Schist during initial uplift of the Southern Alps in northwest Otago in the Miocene (Craw 1995; Craw et al. 2006). Gold occurs in quartz veins and breccia zones that fill extensional sites associated with Moonlight Fault folding (e.g., Copper Creek; Fig. 1) and in normal faults that locally crosscut folds related to the Moonlight Fault (e.g., Macetown, Fig. 1) (Craw et al. 2006). Gold mineralisation was accompanied by extensive structurally controlled ankeritic alteration and silicification, abundant pyrite and arsenopyrite and minor stibnite locally. Auriferous veins and associated ankeritic alteration zones are most common in a conjugate set of extensional fractures striking east and southeast. This extensional fracture set is the same as that which hosts lamprophyre dikes at the southern end of the dike swarm (Fig. 1). The best-developed auriferous vein systems occur to the south of the lamprophyre dike swarm, but similar quartz veins (some of which are gold-bearing) occur in close proximity to lamprophyre dikes near the southern end of the dike swarm (Craw 2006).

A vein swarm occurs in the Smyth Glacier area (Fig. 1), crosscutting schist foliation that has been steepened by tight Miocene-Pliocene folding and faulting. Veins include extensive breccias, especially in intersection zones. These veins strike east and northeast and fill extensional sites in a more extensive network of fractures with the same orientations. Veins and fractures have ankeritic alteration along margins, with sulphide (pyrite, arsenopyrite, stibnite) addition. Similar alteration occurs in nearby northward-striking fault zones that have facilitated post-Miocene topographic development in the area (Craw et al. 2003). On this basis, the vein system is presumed to be Pliocene in age (Fig. 1) rather than Miocene as for the veins to the south which they strongly resemble.

Gold-bearing veins in the Callery and Whataroa Rivers of the central Southern Alps (Fig. 1) fill northeast-striking extensional sites. The Callery mineralised zones extend into graphitic biotite zone host rocks, with coarse-grained (centimetre scale) disseminated, gold-bearing arsenopyrite grains scattered through schist up to 20 cm from veinlets. Veinlets contain calcite, biotite and chlorite, especially along vein margins, and vein quartz hosts minor scheelite and pyrrhotite. Pyrite is absent from the mineralised veins but occurs in related fractures in nearby quartzofeldspathic schists that lack graphite. Veins and associated mineralised zones at Callery cut across folds of foliation formed during post-Miocene exhumation and compression (Craw et al. 1987). The mineralised veins are cut by a regionally extensive swarm of euhedral adularia-quartz-chlorite fissure veins, one of which has been dated via ⁴⁰Ar/³⁹Ar at c. 900 ka (Teagle et al. 1998). Whataroa gold-bearing veins are dominated by breccias and have abundant ankeritic alteration up to 2 m from vein walls with coarse-grained auriferous pyrite and subordinate arsenopyrite. These veins occur in extensional sites in a fault zone that has juxtaposed rocks of differing metamorphic grade during Alpine Schist deformation, in the hanging wall of the Main Divide Fault.

A gold-bearing vein swarm extends in a northeasterly direction across the northern Southern Alps, near the complex structural intersection zone between the Alpine Fault and Hope Fault (Fig. 1). Post-Miocene uplift along the Alpine Fault has stacked steeply dipping metamorphic zones in the Alpine Schist between the Main Divide and the fault. The auriferous vein swarm cuts across these stacked metamorphic zones, from greywacke in the south to biotite zone schist in the north (Becker et al. 2000). The veins fill extensional sites in younger fault-related fractures, possibly associated with superimposition of Hope Fault structures on the older Alpine Fault structures (Campbell et al. 2004). The Fiddes vein system (Table 1; Fig. 1) occurs near the Main Divide, with numerous other similar veins, in greywacke host rocks that have been extensively disrupted by the active Main Divide Fault Zone (Becker et al. 2000). Veins have little alteration on their margins other than minor silicification at the centimetre scale and minor brecciation near vein margins. The veins contain abundant albite as well as quartz and minor pyrite, arsenopyrite and chalcopyrite.

Results

A total of 10 sulphide samples were analysed from the six selected sites (Table 1). Analytical data are presented in Table 2 and are shown on conventional thorogenic and uranogenic plots in Fig. 3A and 3B, respectively, and on a ²⁰⁸Pb/²⁰⁶Pb vs. ²⁰⁷Pb/²⁰⁶Pb plot in Fig. 4.

Fig. 3  (A, B) Lead isotopic compositions of sulphides from gold-bearing veins and vein breccias in the northwestern Otago and Southern Alps, together with fields for sulphide Pb compositions from gold-bearing veins and shear zones in the main part of the Otago Schist belt (from Mortensen et al. 2010). The present-day fields for Pb isotopic compositions of Pahau Terrane greywackes on North Island, New Zealand, are shown for reference (data from Graham et al. 1992; McCulloch et al. 1994). Age-corrected fields for Pb isotopic compositions of Pahau Terrane greywackes at 25 Ma are also shown; these were calculated using Pb evolution parameters as discussed in the text.

Fig. 4  Lead isotopic compositions of sulphides from gold occurrences in northwestern Otago and Southern Alps. Fields for Pahau Terrane greywacke Pbs as in previous figure.

Compositional fields for sulphide Pbs from many of the gold-bearing veins and shear zones within the main part of the Otago Schist belt (Mortensen et al. 2010) are also shown for comparison. These data fall into two distinct fields: a ‘less’ radiogenic field that represents deposits and occurrences formed during an Early Cretaceous (135–142 Ma) mineralising event and a ‘more’ radiogenic field defined by analyses of sulphides from a younger (Mid-Cretaceous; 101–106 Ma) mineralising event.

The average upper crustal growth curve of Stacey & Kramer (1975) is also shown for reference on Fig. 2 and 3, along with a field for the present-day whole rock Pb isotopic compositions of Jurassic greywackes from the Pahau (‘Younger Torlesse’) Terrane in North Island, New Zealand (data from Graham et al. 1992, and McCulloch et al. 1994).

Sedimentary strata of the Pahau Terrane are somewhat younger than those that make up the Rakaia Terrane (which includes the Otago Schist and Alpine Schist). It is generally thought, however, that the Pahau Terrane greywackes are largely recycled from the Rakaia Terrane (e.g., Mackinnon 1983; Mortimer 2004; Wandres et al. 2005); the range of isotopic compositions of the Pahau greywackes should therefore provide a reasonable approximation for that of the Rakaia Terrane. Mortensen et al. (2010) argue that the Pb isotopic compositions of the Pahau/Rakaia greywackes are consistent with evolution at slightly higher µ (= ²³⁸U/²⁰⁴Pb) and κ (=²³²Th/²³⁸U) values than those modelled by the Stacey & Kramer (1975) model (µ=c. 9.88 vs. 9.74; κ=c. 3.85 vs. 3.78).

In view of the very young age inferred for gold-bearing vein systems within the Alpine Schist (<5 Ma), Pb extracted from the host schists and concentrated into the veins would be expected to have essentially the same composition as the present-day greywacke compositions (Fig. 2, 3). On geological grounds, mineralised veins and vein breccias in the Shotover-Macetown area are suggested to be no older than Miocene (<23 Ma; Begbie & Craw 2006; Craw et al. 2006; see discussion above). Fields for Pb compositions of the Pahau greywackes (n=42; excluding two anomalously radiogenic analyses), age-corrected to 25 Ma using the revised µ and κ values for Pahau Terrane greywackes from Mortensen et al. (2010), are shown for reference in Fig. 3 and 4. These fields represent the least-radiogenic limits for Pb derived from the Rakaia Terrane schists in the Shotover-Macetown area.

Discussion

Geochemical studies by Pitcairn et al. (2006) demonstrated that upper greenschist and amphibolite facies Rakaia Terrane metagreywackes in the Otago and Alpine schists are depleted relative to their lower grade (lower to middle greenschist facies) counterparts in the same suite of ore-forming elements (Au, Ag, As, Sb, Hg, Mo and W) that characterises orogenic gold-bearing vein systems contained within the schist assemblages. This observation supports the argument that metals (together with S and fluids) were progressively remobilised out of the metagreywacke package as metamorphic grade increased. Base metals such as Pb, however, were not observed to show a parallel depletion in the higher grade rocks. Pitcairn et al. (2006) postulated that this might reflect the fact that base metals are typically transported as chloride complexes, whereas the elements that showed distinct depletion in the higher grade rocks are thought to be transported as complexes with other ligands (e.g., oxide, hydroxide, sulphide and bisulphide complexes). This suggests that although Pb appears to have much utility as a proxy for the source(s) and behaviour of other metals (including gold) in orogenic gold deposits, it must be applied with caution.

Lead isotopic compositions for sulphides from both Pliocene-Pleistocene veins in the Alpine Schist and Miocene veins and vein breccias from the Shotover-Macetown area in the northwestern Otago Schist belt are slightly less radiogenic than would be expected if the Pb was derived solely from Pahau-type greywacke (Fig. 3, 4). Furthermore, samples from the Pliocene-Pleistocene Callery River occurrence yield similar to somewhat less radiogenic isotopic signatures to those of the Miocene Copper Creek and Macetown occurrences, which is inconsistent with the simple extraction of Pb from a homogeneous reservoir at c. 23 Ma and c. 1–2 Ma. It is important to note, however, that most of these mineral occurrences are small, and likely did not result from mobilisation and homogenisation of metals from very large volumes of source rocks. Thus lateral and vertical lithological variability within the source rocks is likely to be reflected by significant variation in the Pb isotopic signature from one gold occurrence to another.

Metabasic rocks (greenschists and amphibolites) are generally more abundant in the Aspiring Terrane that underlies much of northwestern Otago and parts of the Alpine Schist (Fig. 2) than in most other parts of the Rakaia and Pahau Terranes. Although no Pb isotopic measurements are presently available for any of these metabasic rocks in the Otago Schist belt, they typically display N-MORB, E-MORB or within-plate basalt chemistry (e.g., Bierlein & Craw 2009) and it is likely that their Pb isotopic compositions would be considerably less radiogenic than those of the various metagreywacke units (Fig. 5). Thus, Pb extracted from a portion of the Aspiring Terrane that included a substantial proportion of metabasic rocks would be expected to have a less radiogenic composition than Pb derived entirely from metagreywacke, as is observed in the data for the Alpine Schist and northwestern Otago Schist belt occurrences. The Pb isotopic data for the sulphide occurrences indicates, however, that any Pb contribution from the metabasic rocks must have been very minor (Fig. 5).

Fig. 5  Lead isotopic compositions of sulphides from gold occurrences in northwestern Otago and Southern Alps (symbols as in Fig. 2, 3) together with Pb isotopic compositions of lamprophyres in the Alpine (= Westland) Dikes (from Barreiro & Cooper 1987; Hoernle et al. 2006). The Stacey & Kramer (1975) growth curve is shown for reference, along with fields for 25 Ma compositions of Pahau Terrane greywacke Pbs as in previous figures. Black arrow points to 0 Ma on an average mantle growth curve (from Zartman & Doe 1981).

A possible genetic linkage between lamprophyres and orogenic gold systems elsewhere in the world has been suggested by numerous workers (Rock & Groves 1988; Wyman & Kerrich 1988; Rock et al. 1989; Kerrich & Wyman 1990), although the evidence for a direct connection has been subsequently questioned by others (e.g., Ashley et al. 1994; Kerrich & Wyman 1994; Taylor et al. 1994). Widespread alkaline lamprophyre dikes in the northwestern Otago Schist belt and in parts of the Southern Alps (Alpine Dikes of Cooper 1986; Westland Dikes of Barreiro & Cooper 1987) (Fig. 1) give consistent late Oligocene to early Miocene K-Ar, ⁴⁰Ar/³⁹Ar and U-Pb ages (27–20 Ma; Cooper et al. 1987; Adams & Cooper 1996; Hoernle et al. 2006). There is a close spatial association between alkaline lamprophyre dikes and gold-bearing veins and vein breccias in the Shotover-Macetown area, where the dikes and veins are observed to be emplaced into the same set of structures (Craw et al. 2006). Lead isotopic compositions for many of the lamprophyre dikes (Barreiro & Cooper 1987; Hoernle et al. 2006) are highly radiogenic (Fig. 5) with distinctly elevated ²⁰⁶Pb/²⁰⁴Pb compositions, reflecting a very high-µ (‘HIMU’) mantle source for the magmas. The Oligocene-Miocene lamprophyres in northwestern Otago and the Southern Alps are now interpreted to be part of a long-lived (Late Cretaceous to mainly Cenozoic) phase of diffuse alkaline mafic magmatism that affected much of central and southern Zealandia (e.g., Hoernle et al. 2006; Panter et al. 2006). On the basis of age, the lamprophyres could only be genetically related to the Miocene gold systems in the Shotover-Macetown area. The lamprophyres are volumetrically very minor in this area, however. Although widespread ankeritic alteration of wall rock schists is commonly observed adjacent to the lamprophyre bodies, it seems unlikely that this phase of small-volume magmatism would have been capable of driving hydrothermal circulation that might have significantly influenced the formation of orogenic gold systems in the area. Furthermore, any metals derived from the lamprophyres would be expected to share the distinctive high ²⁰⁶Pb/²⁰⁴Pb signature of the lamprophyres, and there is no evidence for this in the measured Pb isotopic compositions of the veins (Fig. 5). Craw et al. (2006) review light stable isotope evidence for carbonates and gold-bearing veins in the Shotover-Macetown area that also argues against a genetic relationship between the lamprophyres and veins. Our Pb isotopic data are therefore in good agreement with the contention of Craw et al. (2006) that, although the lamprophyres and gold-bearing veins and vein breccias in the Shotover-Macetown area are of approximately the same age and were emplaced into the same generation of fault and fold structures, there is no direct genetic connection between them.

Conclusions

This study tested the relative role of three distinctly different sources for Pb in hydrothermal gold-bearing veins that formed in the Otago and Alpine Schists between Miocene and Recent time: (1) host and underlying metasedimentary rocks of the Rakaia Terrane; (2) metabasic rocks interlayered with the metasedimentary rocks; or (3) lamprophyre dikes that were emplaced coevally with Miocene gold-bearing veins in similar extensional structural sites. The Pb isotopic signatures of sulphide minerals in gold-bearing veins are consistent with the derivation of most of the contained Pb from underlying host metasedimentary rocks with similar Pb isotopic signature to the Pahau Terrane. There are some differences in Pb isotopic signatures between gold deposits, and these differences are probably a result of lateral and vertical lithological variability within the source rock mass. The observed Pb isotopic signatures in the gold-bearing veins precludes significant incorporation of Pb from metabasic rocks contained within the host metasedimentary package, which have N-MORB, E-MORB or within-plate basalt chemistry and would be reasonably expected to have a much less radiogenic isotopic signature than the metasedimentary rocks. In contrast, lamprophyre dikes in the region have Pb isotopic signatures that are much more radiogenic than the host metasedimentary rocks or the sulphide minerals, which appears to rule out any genetic relationship between dikes and gold mineralisation.

Acknowledgements

This research was partly funded by the New Zealand Foundation for Research, Science and Technology and University of Otago (DC) and the Natural Sciences and Engineering Research Council (JKM). Discussions with Alan Cooper, Nick Mortimer, Andy Tulloch and Michael Palin were helpful in developing ideas herein. The paper also benefited from a critical review by Richard Goldfarb on behalf of the journal.