GEMOC ARC National Key Centre

Figure 1.  Eloise Beyer knee-deep in her research area.

The WGR can be broadly divided into two principal lithotectonic units: (1) the HP/UHP metamorphosed Fjordane complex, a mixed assembly of gneiss, supracrustal rocks, anorthosites, augen gneisses and eclogites, exposed to the north and west; and (2) the Jøstedal complex, a more monotonous migmatitic orthogneiss unit to the east.  Sample localities were selected from both units, though the Fjordane complex was of particular interest as it is this unit which contains the mantle fragments.

Detrital zircons were collected in stream sediment from several localities throughout the WGR.  Preliminary results from analysis of four samples indicate a predominance  of Proterozoic ages in the range 1600-1700 Ma.  These ages match the period of extensive crustal formation in this area, known as the Gothian Orogeny and it is reasonable to assume that zircons of this age were sourced from rocks formed during this mid-Proterozoic event.  There were also a high number of zircons with U-Pb ages in the range 1400-1500 Ma which can be correlated with a phase of intermittent mafic intrusions and possible sporadic granitic magmatism.  A smaller number of zircons fall in the range 900-1300 Ma which corresponds with the extensive intrusion of dolerites and granites during the Sveconorwegian (Grenvillian) Orogeny.

Figure 2.  A screen concentrate with red and black rings defined by the garnet and spinel from the peridotites – the zircons are concentrated with these heavy minerals.

The U-Pb ages give no direct evidence for an Archean-aged event in the WGR basement rocks.  However, the Hf isotopic compositions of the zircon can be used to determine the age of the sources of the magmatic parent rocks of the zircons and hence constrain the age of the deep crust.  Hf isotopic data for two samples analysed to date is predominantly very juvenile, indicating that the zircon source has not undergone significant crustal reworking.  Crustal model ages for the oldest group of zircons (1600-1700 Ma U-Pb ages) range from 1700-2300 Ma and show no evidence for any Archean input.  The zircons with ages less than 1600 Ma have Hf compositions which suggest they were derived from a mixture of younger more juvenile material and older mid-Proterozoic crustal rock.

The lack of evidence for Archean crust in the WGR basement rocks suggests that the crust and the mantle peridotite bodies enclosed within it are not genetically related.  With no evidence for an Archean lower crust, the data favour a Laurentian provenance for the peridotites (see Fig. 3).  The WGR, already famous for its insights into HP/UHP metamorphism, may also contain hard physical data on the geometry and mechanisms of continental collision and the partial subduction of continental plates.

Contacts: Eloise Beyer, Bill Griffin, Sue O’Reilly
Funded by: ARC, Industry

Figure 3.  "Slam-dunk” model for the WGR peridotites, stuffed into the top of the descending Baltica plate.

Keeping the lid on: How long do island arc magmas degas before erupting?

AT ISLAND ARCS, fluids coming off the subducting plate facilitate partial melting in the wedge of mantle above the plate.  Subsequent decompression and differentiation leads to volatile build up, degassing, the formation of hydrothermal systems and associated economic deposits, and often to violent eruptions (eg Krakatau, Tambora, Vesuvius, Mount St Helens).  However, the relationships between degassing, the composition of erupted magmas and the time scales of deeper magma degassing remain unclear.  For example, degassing raises the solidus, increases plagioclase stability and forces crystallisation, such that magma differentiation might be driven by degassing during magma ascent.  However, volatile loss leads to increased magma density which will retard magma-crystal separation, so the link between degassing and differentiation is not straightforward.  Finally, volatile build-up will be an inevitable consequence of differentiation unless it is counterbalanced by degassing -- which is why the more differentiated and volatile-rich island arc magmas typically erupt most explosively.

Figure 1.  Once an exsolved volatile phase exists, gases can move independently and escape from the magma through the surrounding rock or via the magma itself.  A bottle neck or physical obstruction in the magmatic plumbing system can cause accumulation of gas bubbles at certain depths, which leads to potentially explosive situations.

If we can measure the time scales of these magmatic processes, we can gain important insights into the physical mechanisms at work beneath volcanoes.  There has been much interest in links between the timing, composition and volume of gas emission and eruptive behaviour.  Measurement of the short-lived isotope ²¹⁰Pb (half life = 22.6 years) can constrain the time scales of magma degassing because the decay chain from its progenitor ²²⁶Ra includes the noble gas ²²²Rn, which has a half life of only 3.8 days.  ²²²Rn is likely to be efficiently and continuously lost from a magma so long as exsolution and degassing of a major volatile species (eg CO₂, H₂O, SO₂…) is occurring.  However, its short half life means that its concentrations will grow back into secular equilibrium with ²²⁶Ra within a couple of weeks after degassing ends.  Thus, ²¹⁰Pb concentrations will not “see” the effects of ²²²Rn degassing, and ²¹⁰Pb/²²⁶Ra will remain =1, unless the duration of degassing begins to approach the half life of ²¹⁰Pb.  However, if the duration of radon degassing is on the order of several years or more significant ²¹⁰Pb deficits, i.e. ²¹⁰Pb/²²⁶Ra < 1, can be produced.

Figure 2.  Variation in (²¹⁰Pb/²²⁶Ra) in several subsequent eruptions are similar to the variation in Li concentration in plagioclase phenocrysts of those same eruptions. Both (²¹⁰Pb/²²⁶Ra) and Li are indicators of degassing behavior of the magma.  Prior to the May 18^th gas bubbles with Li and ²¹⁰Pb accumulated, whereas on May 18^th magma with low ²¹⁰Pb and low Li was erupted indicating a deeper degassed reservoir was tapped during this eruption.  Figure after Berlo et al. (Science vol. 306).

Measurements of ²¹⁰Pb/²²⁶Ra in a wide range of island arc lavas from around the world show that more than 60% have significant ²¹⁰Pb deficits, which require degassing periods of 2-47 years.  Thus it appears that most island arc magmas sit and degas for decades prior to eruption.  A detailed study of Sangeang Api volcano in Indonesia indicates that this degassing post-dates magma differentiation, which occurred by crystallisation due to cooling over thousands of years deeper within the arc crust.  At Mount St. Helens it appears that the ²¹⁰Pb/²²⁶Ra ratio may vary with eruptive style (plinian versus effusive).  Some lavas have ²¹⁰Pb excesses, which probably reflect the addition of ²¹⁰Pb via gas transport from deeper, degassing magma.  ²¹⁰Pb excesses therefore might be used to identify volcanoes underlain by freshly emplaced magma bodies.  Such information can ultimately inform models for hazard mitigation.

The resurgent lava dome in the crater of Mount St. Helens, looking down the “blast corridor” of the 1980 eruption.

PGE fractionation in mantle sulfides: fingerprints of metasomatic processes

ANALYSES OF MANTLE SULFIDES have demonstrated the high solubility of the Platinum Group Elements (PGEs) in base metal sulphides (BMS) and show that the PGE budget of mantle rocks is fully controlled by the BMS.  If they are not enclosed inside refractory silicate or oxide phases, sulfides are likely to be mobile in the lithospheric mantle and their absolute and relative abundances of PGEs can be altered by metasomatic processes.  Therefore, studies of mantle sulfides may provide critical information on metasomatic processes in the mantle.  Abundant primary sulfides occur both as inclusions in silicates and as interstitial grains in spinel lherzolite and pyroxenite xenoliths hosted by Miocene intraplate basalts from the Penghu Islands, Taiwan.  Modally metasomatised xenoliths with amphibole can have sulfide abundances up to 1 vol.%, whereas cryptically metasomatised samples, with re-enriched trace-element patterns in clinopyroxenes, have 0.1~0.3 vol.% sulfide, well above the mantle average.  These sulfides are mixtures of Fe-rich and Ni-rich monosulfide solid solutions (MSS), pentlandite, millerite and chalcopyrite, exsolved from high- temperature (>900°C) MSS bulk compositions.

Figure 1.  Bulk Ni (a) and Co (b) versus Cu in Penghu mantle sulfides.

Most Penghu sulfides are interstitial to the silicate phases, suggesting a metasomatic origin.  Two distinct groups of sulfides, KP and TCY, are distinguished by their reconstructed major-element compositions (Fig. 1).  The KP sulfides are Fe-Cu-rich, S-deficient MSS, whereas the TCY sulfides are Ni-rich MSS with unusually high Co contents, ranging up to 11.5 wt.% (Fig. 1).  Their major-element characteristics suggest that each group of sulfides resulted from the crystallization of sulfide melts at high temperature: KP from a Cu-rich melt and TCY from a Ni-Co-rich one.  Sulfides from both groups have lower Os and Ir abundances than typical residual MSS and have PGE patterns within the range of typical sulfide melts (Fig. 2); the KP sulfides have more irregular PGE patterns than the TCY sulfides, with high Ru-Rh and low Pt-Pd.  The Os-isotope systematics of both groups of sulfides give Proterozoic model ages, and it seems probable that these differences in the PGE patterns reflect reaction between the sulfides and metasomatic fluids between Proterozoic and Miocene time.

Figure 2.  Chondrite-normalised PGE patterns for Penghu mantle sulfides.
Data for shaded fields representing residual MSS and sulfide liquids are from Griffin et al. (2002).

The so-called P-PGEs (Pd+Pt) in the two groups of sulfides are fractionated relative to the I-PGEs (Ir + Os) but in different ways (Fig. 3a), and they have different Re-Pd relationships (Fig. 3b).  Because the P-PGEs and Re are relatively incompatible elements during melting processes, these patterns suggest that both groups of sulfides have been metasomatised by silicate melts, but of different composition.  Os (and Ru and Rh; not shown) behave coherently with Ir in the TCY sulfides, but in the KP sulfides, Ir and Os are only weakly correlated (Fig. 3c).  This I-PGE fractionation is not likely to be caused by silicate metasomatism, considering the similar particient coefficients of the I-PGEs, so another metasomatic agent is suggested.  Aqueous fluids/low density supercritical fluids may fractionate Os from Ir (Lee, 2002; Lorand et al., 2004), and it is likely that such fluids have affected some of the KP sulfides to fractionate their I-PGEs.

Our previously published Os isotope data (see Research Highlights, GEMOC Annual report 2003) show these sulfides have recently undergone one unique type of Os disturbance: the addition of radiogenic Os, with little or no addition of Re.  The introduction of highly radiogenic Os, which is only observed in some of the KP sulfides, may be caused by an oxidizing hydrous fluid carrying Os, but not Re.  In these KP sulfides, metal/sulfur ratios are negatively correlated with ¹⁸⁷Os/¹⁸⁸Os (Fig. 3d), indicating that sulfides equilibrated under oxidising conditions have higher ¹⁸⁷Os/¹⁸⁸Os.  The unique metasomatic agent causing fractionation of I-PGEs coupled with enrichment in (radiogenic) Os may reflect specific processes that remain to be identified.  Sulfide studies can provide more constraints on these processes, and sulfur saturation in metasomatic fluids may play a major role in controlling Re and Os (and other PGE) distribution in magmas and residual peridotites.

Figure 3.  (a), (b) and (c) Bivariate bulk abundance plots for Penghu mantle sulfides. (d) ¹⁸⁷Os/¹⁸⁸Os versus bulk metal/Sulfur ratios.  Solid circles: KP sulfides showing negative correlation between ¹⁸⁷Re/¹⁸⁸Os and ¹⁸⁷Os/¹⁸⁸Os. Open circles: other KP sulfides.  Open squares: TCY sulfides.

Fault-scarp-derived volcaniclastic deposits on Macquarie Island; sedimentation at a fossil ridge-transform intersection?

STUDIES OF OCEANIC CRUST are hindered by limited subaerial exposure.  Our knowledge of in situ oceanic crust therefore comes from bathymetry and other geophysical data, dredge samples and in a few areas survey and sampling by submersibles.  Moreover, even where rarely exposed on land within ophiolites, oceanic crust has typically been deformed to varying degrees during obduction.

Figure 1.  (a) Location of Macquarie Island indicated by a star.  (b) Geologic map of Macquarie Island.

Macquarie Island, located approximately 1200 km southwest of New Zealand in the Southern Ocean  (Fig. 1), forms the apex of the Macquarie Ridge Complex (MRC), a system of ridges and troughs along the currently active Australian-Pacific oceanic transform plate boundary between the Alpine Fault of New Zealand and the Australian-Pacific-Antarctic triple junction.  The island exposes the eastern side of the ~5 km high, ~50 km wide submarine ridge and lies ~4.5 km east of the major active plate boundary fault zone.  Macquarie Island offers a globally unique opportunity to examine oceanic crust in situ – it is the only exposed piece of non-plume-related oceanic crust that still lies within the basin in which it formed, and in a relatively well-constrained present-day plate tectonic setting.

Figure 2. (a) Volcaniclastic breccia, Unity Point. (b) Gabbro and diabase clasts in volcaniclastic breccia.

Recent research results:  Upper Miocene to lower Pliocene sedimentary rocks on Macquarie Island are dominantly volcaniclastic breccia (Fig. 2a), sandstone  (Fig. 3), and siltstone produced by the physical disintegration and tectonic abrasion of oceanic crust in fault zones, and mass wasting of these tectonic features.  They represent small debris fans and small-scale turbidites deposited at the base of active fault scarps, related to Late Miocene to Early Pliocene seafloor spreading.  Most of the sediment is derived from basalts, but diabase and gabbro clasts in some sedimentary rocks (Fig. 2b) indicate that middle and lower oceanic crust was exposed to erosion on the sea floor.  A lack of exotic clasts and a low degree of clast roundness imply a local source for the sediment without input from continental rocks.  Spatial relationships between sedimentary rocks and major faults associated with seafloor spreading on the island and correlations between sedimentary clast and adjacent up-thrown block compositions reflect paleo-tectonic relief for Macquarie Island crust during deposition.   Our data support a model involving the deposition of these rocks at the inside corner of a ridge-transform intersection.  Furthermore, a tectonic reconstruction of the Australian-Pacific plate boundary for the approximate time that Macquarie Island crust formed (10.9 Ma) also shows that Macquarie Island crust most likely formed near to a ridge-transform intersection.  Sediments associated with active faulting at a ridge-transform intersection have been uplifted in situ above sea level along with the surrounding oceanic crust, which indicates that high-angle faults have more pronounced influence on sedimentation than do low-angle faults, in this tectonic environment (see GEMOC publication #388).

Figure 3.  Flame structures and graded bedding.

Contacts: Nathan Daczko (partly done at Department of Geological Sciences and Institute for Geophysics, Jackson School of Geosciences, University of Texas, Austin, Texas and Macquarie University)
Funded by: Australian Antarctic Division, University of Texas

Figure 4.  Sampling hazards – female elephant seals moulting amongst rotting smelly kelp.

A continental near-neighbour during Abitibi crustal growth

THE ABITIBI SUBPROVINCE (part of the Superior craton of the Canadian Shield) is widely regarded as an example of juvenile Archaean crust generated during volcanic and plutonic activity in an ocean setting, far removed from the influence of older continental crust.  However, our recent work along the southwestern margin of this subprovince casts doubt on this hypothesis.

Figure 1.  Geological map of the southwestern Abitibi subprovince, Ontario, Canada.  The locations of two studied plutonic complexes are indicated by rectangles. Isotopic evidence for the involvement of older continental crust ( 2.92 Ga) during Abitibi magmatic activity is known from the locations shown by the dots.  The dashed line denotes the probable eastward extent of this older crust at depth.

The Kenogamissi plutonic complex is located southwest of Timmins in the Ontario segment of the Abitibi subprovince (Fig. 1).  Samples of tonalite, granodiorite, granite, and diorite from this complex have zircon U-Pb primary ages ranging between 2747 ± 3 Ma and 2676 ± 4 Ma, representing nearly the entire period of Abitibi magmatic activity.  Surprisingly, two of these samples also contain ca. 2850 Ma zircon grains.  We believe that these zircons were derived from partially melted older continental crust during magma formation.  Hf isotopic measurements were made in situ on dated igneous zircons from this complex.  ¹⁷⁶Hf/¹⁷⁷Hf ratios vary widely for individual samples (Fig. 2), consistent with final magma generation by mixing of at least two magma types: juvenile mafic melts derived from the depleted mantle, and melts of slightly older continental crust.  Our in situ isotopic approach, combined with the radiometrically sensitive nature of the Lu-Hf system, partly explains why Hf isotopic data from individual zircons can be used to detect this older crust.  In contrast, whole-rock Sm-Nd isotopic data from this region typically fails to indicate its presence.  However, some of the Nd data are consistent with this interpretation, and rare 2.93-2.85 Ga inherited zircons have now been documented by other workers in both plutonic and volcanic samples from north of the Kenogamissi plutonic complex (Fig. 1).

Figure 2.  Hf isotopic data for igneous zircons.  Most samples are characterised by a wide range of eHf, suggesting contributions of isotopically distinct magma batches to the final magma.  Low eHf values (near or below CHUR) are consistent with the presence of older crust in the source region.  Ca. 2850 Ma inherited zircons found in two Kenogamissi samples support this hypothesis.

Based on this growing evidence for the presence of older crust beneath the southwestern Abitibi subprovince, we propose the following crustal model (Fig. 3).  The adjacent Wawa subprovince contains rocks as old as 2.92 Ga and is currently separated from the Abitibi subprovince by the Kapuskasing structural zone (KSZ), an east- to southeast-verging intracratonic thrust system.  Movement on the KSZ began in the late Archaean and culminated at roughly 1.9 Ga.  Prior to its development, there must have been a boundary between 2.75-2.70 Ga granite-greenstone belts underlain by older crust (Wawa side) and 2.75-2.70 Ga granite-greenstone belts lacking this older basement (Abitibi side).  We suggest that part of the present-day Abitibi subprovince formed above an eastward-tapering wedge of older Wawa crust.  This lower crustal wedge may have formed during earlier episode of rifting and continental margin development.  Subsequent tectonic activity on the KSZ left a region of autochthonous crust ‘stranded’ on the Abitibi side of the system, isolated from tectonically similar counterparts in the Wawa domain.  Development of the KSZ was probably controlled in part by the location the older, rift-weakened(?) crustal boundary.  The Abitibi subprovince thus appears to have formed in a pericratonic setting, close to and partly on top of an older continental fragment.

Contacts: John Ketchum, John Ayer (Ontario Geological Survey), Otto van Breemen (Geological Survey of Canada)
Funded by: ARC (to O'Reilly/Griffin) and Geological Survey of Canada

Seismic properties of the Amery Ice Shelf – How thick is it, and how does it move?

THE AMERY ICE SHELF (AIS) is located on the east coast of Antarctica, at approximately 70ºS, 70ºE (Fig. 1).  It is the third largest embayed shelf in Antarctica, and the largest entirely in East Antarctica, and as such is one of the largest glacier drainage basins in the world.  Due to this, and its thermal isolation, the AIS plays an important role in the global climate system.  Knowing the structure of the AIS is important to studies of the impact of global warming on present-day ice shelves and the subsequent effect on global ocean circulation and climate.  Understanding the structure of the AIS can also provide tighter control in mass balance calculations of the shelf and in ocean circulation models beneath the ice cover.  The seismic geophysical technique is a powerful tool for surveying the thickness and structure of the subsurface, and this makes it perfect to investigate the AIS.

Figure 1.  Map of the Amery Ice Shelf, East Antarctica. Inset: Map of Antarctica showing the major ice shelves.

A detailed seismic reflection survey was undertaken in the summer of 2002/2003 at the centre of the AIS to map the thickness of both the ice and water column, to investigate the sediment structure and most importantly to see if a delineation could be made between glacial and marine ice.  The detailed seismic reflection survey was performed utilising a 24-channel spread with a 10 m geophone spacing, giving a maximum fold coverage of six.  The survey successfully delineated the subhorizontal glacial and marine ice layers, with average thicknesses of 754 and 20 m, respectively.  The water column is 590 m thick, placing the seafloor at a depth of 1364 m below the ice surface. A 60 m thick, subhorizontally-layered sedimentary unit can be delineated (Fig. 2).

Out for a days work, photo by K. McMahon.

During the 2004/2005 Antarctic summer season, seismic data were used primarily to investigate anisotropy within the ice.  Anisotropy in ice refers to a difference in seismic velocity within the ice due to ice crystal orientation.  As ice streams merge to become the AIS, the margins undergo stress, and ice crystals realign and elongate, meaning the ice structure is no longer homogenous, i.e. it is no longer isotropic.  This could produce a measurable difference in seismic velocity with orientation relative to relict ice stream margins.  These margins were picked from satellite images of the shelf, which reveal flowlines – the preserved boundaries of ice streams.

Figure 2.  (a) CDP stacked section of processed reflection data showing all major reflections.  Legend: IP – primary ice reflection; IM1 – first order ice multiple; IM2 – second order ice multiple; SP – primary sediment reflection; SM1 – first order sediment multiple; SM2 – second order sediment multiple.  Circled objects are the deep hyperbolic reflections. (b) Detailed image of ice boundaries, with interpretation. (c) Detailed image of sedimentary layering, with interpretation.

This season’s work area was adjacent to Gillock Island  (Fig. 1) where twelve sites were investigated for anisotropy using refraction surveying techniques, four regional sites were surveyed (extending the area of the shelf surveyed by Australian seismologists) and a detailed seismic reflection line was carried out, spanning ~3km of the shelf E-W over a relict stream margin.  This was done using a 24-channel spread (48 for the detailed line), with a 10 m geophone spacing.  For the refraction surveys, the data reveal information up to approximately 200 m depth within the ice.

Ray-inverse modelling is being used primarily to process the anisotropy refraction data, using a model for the ice shelf incorporating a thin snow layer overlaying firn that changes gradually to solid ice within the shallow surface material.  This is more realistic than using a simple layered model, and is supported by data from ice cores elsewhere on the AIS.

The Gillock Island area of ice is older than the ice of the southern grounding zones, which will be the site for further investigation in the 2005/2006 season.  This will lead to a comparison of the strength of anisotropy with age of the ice, and proximity to the zone of initial stress.

Contacts: Kathleen McMahon, Mark Lackie
Funded by: ASAC Grants 2542 and 2581

Base camp on the AIS: “Camp Tropical”, Photo by K. McMahon.

Seismic work at AM01 on the AIS, Photo by K. McMahon.

Tracing gold with zircon

AS PART OF AN ONGOING STUDY of the relationships between granites and intrusion-related gold deposits in the Georgetown Inlier, north Queensland, the Black Cap Diorite and Lochaber Granite within the Lochaber-Bagstowe Complex (Fig. 1), and the Median Dyke within the gold-rich porphyry deposit at Kidston (Fig. 1 and 2) were studied to evaluate existing models for temporal and genetic relationship between these intrusive rocks. Geologic, fluid inclusion and stable isotope studies suggest that the mineralisation and brecciation at the Kidston breccia pipe is spatially, temporally and genetically related to the north-west trending Permo-Carboniferous rhyolite dyke magmatism (Paddys Knob Dyke Swarm). 

The U-Pb ages and Hf isotope composition of zircons are used to determine the timing and degree of crust/mantle interaction involved in the genesis of porphyry intrusions.  Zircons from selected rocks were analysed by in situ LAM-ICPMS (Laser Ablation Microprobe Inductively Coupled Plasma Mass Spectrometry) for U-Pb ages and Hf isotope composition.

Figure 1.  Regional geology of the Kidston area and the relationship of the Kidston to Permo-Carboniferous volcanism, the NW trending rhyolite dyke swarm and coincident gravity low (after Baker and Andrew, 1991.)

The gold-mineralised Median Dyke (Fig. 2 and Fig. 3C) within the gold-rich subvolcanic breccia-hosted porphyry deposit at Kidston has a U-Pb zircon crystallisation age of 335.7±4.2 Ma. The same rock, dated by SHRIMP I ion probe, yielded a ²⁰⁶Pb/²³⁸U age of 334.7 ± 3.3 Ma.  The Black Cap Diorite (Fig. 3A) and the Lochaber Granite  (Fig. 3B) within the Lochaber-Bagstowe Complex (LBC), have U-Pb zircon ages of 350.6±1.6 Ma and 338.9±2.7 Ma respectively.

The Median Dyke has e_Nd=-13.45 (like other Kidston porphyries); the Lochaber Granite, e_Nd=-10.68; the Black Cap Diorite, e_Nd =-7.81, and the Paddys Knob Dyke Swarm have an e_Nd range from -9.6 to -18.2.  The range of e_Nd values was interpreted to reflect either magma mixing or assimilation of highly isotopically evolved Georgetown crust (Nd model age ca 1.70 Ga) into the evolving magmas.

The range of relatively non-radiogenic Hf isotope signatures in each rock (Median Dyke, e_Hf=-6.3 to -18.3; Lochaber Granite, e_Hf=0.26 to -8.7; the Black Cap Diorite, e_Hf =3.8 to -9.3) is interpreted to reflect magma derivation by remelting of heterogeneous deep crust.  The Hf model age of 1.22-2.26 Ga suggests the source could be the regionally developed Proterozoic crust, but Siluro-Devonian granitoids were derived earlier from the Proterozoic crust (see Research Highlights 2003), and it therefore is difficult to distinguish between a Proterozoic or a Siluro-Devonian source rock using only Hf data.  The younger Lochaber Granite could have been derived by fractional crystallisation from the older Black Cap Diorite, which has a similar Hf-isotope signature.

Figure 2.  Surface geology of the Kidston breccia pipe (after Baker and Andrew, 1991).

The Median Dyke contains zircons with a wide range of Hf isotope compositions, which barely overlaps the range seen in the Black Cap Diorite and Lochaber Granite.  The chemical characteristics of the Median Dyke fall along the trends defined by the Paddys Knob Dykes and the LBC, suggesting that they are all part of the same suite and spatially, temporally and genetically related.  However, the Hf-isotope data rule out a simple fractional crystallisation relationship.  The less radiogenic Hf in the zircons from the Median Dyke suggests instead that magmas like the Black Cap Diorite were mixed with melts derived from Proterozoic crust with Hf model ages between 1.78 Ga and 2.26 Ga, to produce the more felsic compositions of the Median Dyke magmas.

This study illustrates once again the power of the combined U-Pb and Hf-isotope analysis of zircons as a petrogenetic tool.

Contact: Valeria Murgulov, Phil Blevin, Sue O’Reilly, Bill Griffin
Funded by: AMIRA, ARC, Macquarie University (APA)

Figure 3. Plot of eHf vs U-Pb zircon age showing the range in eHf values.  Error bars shown for grains that define the age of the rock (open diamonds).

Zircon – a multi-faceted tool for petrogenetic modelling