Terrestrial Origins and Evolution


Geochronology and Isotope Geochemistry





The isotopic compositions of the elements are not constant. The transmutation of nuclides during radioactive decay changes the abundance of both parent and daughter nuclides. Given the rate of nuclear decay of the parent, the age of a given sample can be calculated from parent daughter ratios and this forms the basis of geochronology. Fractionation of isotopes in chemical reactions occurs because of the mass difference between isotopes. An understanding of the physical processes involved can therefore be ascertained by measuring isotopic abundances. Spallation induced isotopic shifts form the basis of cosmogenic dating and are also evident in extraterrestrial samples exposed to cosmic rays. Finally, the elements themselves may retain a memory of their nucleosynthesis in stars. Our group carries out measurements of a variety of elements that show the effects of one or more of these processes.

The noble gases, helium, neon, argon, krypton and xenon, are very useful geochemical tracers. Studies of their abundances and isotopic compositions in geological samples provide important constraints on hypotheses concerned with the origin and evolution of the Earth’s atmosphere, crust, mantle and core. The identification of the primordial noble gas composition of the Earth is critically important for understanding how and when the Earth acquired its volatiles and how its atmosphere evolved. We have found a remarkable correlation between helium and neon isotope systematics in mantle-derived samples. These results have provided strong evidence for primordial helium and neon of solar composition within the Earth. It is very important to ascertain whether the heavier noble gases, argon, krypton and xenon in the Earth are also solar in composition. To better constrain the evolution of noble gas compositions in the mantle we have started undertaking noble gas studies on old mantle-derived materials, including diamonds and early Archaean materials from Greenland. We have also started research projects on cosmogenically-produced noble gases in rocks at the Earth’s surface in relation to exposure ages.

The GIG Group is involved in a wide range of study areas from the formation of the Earth from its primordial constituents through to the history of the Earth as a dynamically evolving planetary body. Our research techniques are primarily focused on variations in isotopic abundances that can be used for radiometric age dating as well as tracers of processes affecting the Earth. Additional information is derived from geochemical characteristics of rocks as well as their mineralogy and petrology, and when appropriate, field geology.

The thematic areas are exemplified by some of the specialties of the group members. Dr Ireland is interested in the nature of material that makes up the Early Solar System and the processing of that material into planetary bodies. Dr Bennett is working on the origin and chemical evolution of the earth as well as applications of isotope geochemistry to problems in the Earth Sciences. Dr Nutman specializes in early Archaean geology and Precambrian tectonics. Dr Williams is focused on SHRIMP zircon geochronology and its application to the development of the Australian continent and once contiguous areas of Gondwana. Dr Honda is involved in the application of terrestrial noble-gas studies to the origin and evolution of the Earth's atmosphere, crust, mantle, and core.

Analytical facilities within the group include the SHRIMP ion microprobes and the noble gas mass spectrometers. The facilities are available to outside researchers through PRISE. We also collaborate with Geoscience Australia in optimizing the SHRIMP facilities for geochronology.


SHRIMP Development

S. W. Clement, J. J. Foster, T. R. Ireland, B. Jenkins, P. Lanc, N. Shram, R. Waterford, A. Welsh, I. S. Williams


The design of the reverse geometry SHRIMP RG should produce approximately four times the mass resolution of the forward geometry SHRIMP II design at the same sensitivity level. This year saw us take a major advance in the performance of the SHRIMP RG with a redesign and refitting of the mass analyzer. The main problem in the original design was an assumed fringe field distribution of the electrostatic quadrupole lenses that was based on a magnetic quadrupole. The electrostatic lenses were modeled in 3D with SIMION resulting in significant changes to both the lens structures and the chambers, effectively allowing the electrostatic field to expand without truncation by ground potential. The recommissioning of the SHRIMP RG immediately revealed a dramatic improvement in image quality with mass resolution at close to design specification (Figure). In terms of sensitivity, SHRIMP RG can run at or above SHRIMP II levels at 6,000 R. There remains a vestige of aberration from wide divergence into the mass analyzer but at this stage it is unclear whether this is due to residual problems with the electrostatic lenses, or a known discrepancy between the theoretical fringe field of the magnet and that measured. We are continuing a study on the modeling of the magnet design through simulations produced by the OPERA program. The results of these studies will also have implications for our future designs of the SHRIMP II instruments.

SHRIMP RG has been successfully used for U-Pb analysis in the standard SHRIMP configuration, measurement of trace elements at ca. 12,000 R, as well as higher resolution measurements for Sr and Ti isotopes. The main limitation for high mass resolution measurements at ca. 20,000 R and above is the limited field step size available. A redesigned field controller with effectively unlimited steps is being produced and should be fitted to the SHRIMP RG in early 2002.


A catastrophic blow out of an oil seal resulted in the contamination of the source chamber. Following an heroic refurbishment of the source chamber by Dr John Foster, SHRIMP I is performing better than ever. SHRIMP I is currently only being used for U-Pb geochronology.


This year has seen the testing of the prototype multiple collector and Cs gun/negative ion configuration. Testing of the multiple collector during the year indicated general agreement with expectation, but there were problems in the control of individual counters that had not been expected, and some stray ions causing cross talk in the individual counting systems. Based on the experience gained from testing of the prototype, a series of modifications have been proposed and fabrication is underway of the modified counting system.

SHRIMP II was also tested in negative ion mode during early 2001. The Kimball Physics Cs ion gun was attached to the primary column and a stable negative ion beam from Si metal was produced at the collector. The coming year will be busy with the final refinements for the multiple collector and an extended push for a stable isotope capability for SHRIMP II.

Some traumatic changes have been made to SHRIMP II that have clearly affected our ability to keep it in a stable configuration for routine analysis this year. The multiple collector requires a new transfer lens in order to make use of the single ETP multiplier on the central beam trajectory. Specific conditions for tuning the transfer lens to obtain optimal peak-shape have been more painstaking than for the single collector and tuning appears to be variable from day to day. Operation of SHRIMP II was also affected by arc discharges in the primary column and required several extensive cleaning episodes to return to stability.

Computing and Control:

SHRIMPs are controlled by the National Instruments LabVIEW platform. This year we have moved the SHRIMPs from the Macintosh platform to Windows with a clear increase in performance in terms of stability and response time. Both SHRIMP RG and SHRIMP II are now operating under Windows LabVIEW. It is expected that we will transfer SHRIMP I from its remarkably robust Macintosh II computer control of ca. 1990 to LabVIEW in the coming year. In addition, we are in the process of moving SHRIMP RG off GPIB/optical fiber communication to direct optical fiber communication. This will increase communication speed as well as reliability. Developments in the control programs are ongoing. This year has seen the development of an automatic tuning facility for SHRIMP analysis further reducing the impact of human variability in analysis. We are working on the development of an automatic mode of operation where stored points can be revisited for unattended analysis.


Noble Gas Development

M. Honda, I. Iatsevitch

During the year we have successfully modified the noble gas extraction system for the automation of gas handling procedures, and programming for automating much of the operation of the gas extraction system and the VG5400 mass spectrometer has been completed principally through the efforts of Dr I. Iatsevitch. For the next phase of the automation, the tasks we plan to undertake include: (1) automatic tune-up of the mass spectrometer source conditions to maximize ion beam intensities, (2) automatic adjustment of the mass spectrometer magnet pole piece position, required to optimize measurement of each of the five noble gases, and (3) automatic cooling of the activated charcoal trap with liquid nitrogen during neon analysis.


Early solar system processes and chronology

T. R. Ireland

The solar system formed approximately 4,560 million years ago. The solar nebula evolved from a cold cloud of dust and gas with high-temperature processing of the dust into a variety of macroscopic solid objects now found in meteorites. Chondrules, composed predominantly of Mg silicates, are by far the most abundant of these, and their origins remain enigmatic. No less enigmatic are calcium aluminium-rich inclusions CAI, that preserve an inconsistent picture of a number of high-temperature processes. CAI are the oldest materials known to have formed from the earliest solar system at 4568 ± 1 Ma. Chondrules likely formed at a similar time. Planet building processes are also recorded as differentiated meteorites (e.g. eucrites) and well-dated examples of these have an age of 4558 ± 1 Myr. These two time markers represent the best absolute age constraints of the early solar system.

Relative time scales can be addressed through short-lived radionuclides with half lives of order 10 Myr or less. One of these short-lived nuclides, 182Hf that decays to 182W with a half life of 9 Ma, has attracted attention of the past few years because of the possibility of dating metal-silicate fractionation (specifically core formation) of large planetary bodies. The presence of a 182W deficit in iron meteorites (relative to bulk chondrite meteorites) suggests that Fe meteorites formed, that is Hf was fractionated away from W, prior to the complete decay of 182Hf. The equivalence of chondritic and terrestrial W isotopic compositions therefore indicates that terrestrial core formation, which causes Hf-W fractionation, can only take place after all 182Hf has decayed to suitably low levels, approximately 60 Ma.

Of fundamental importance in the Hf-W system is the initial 182Hf/180Hf. The low level of Hf and W in meteorites requires the analysis of rather large samples and in fact only whole-rock isochrons have been produced through ICP-MS analysis. In this regard there is a question over the absolute age of a meteorite or suite of meteorites that affects the calculated initial of the 182Hf/180Hf. Specifically, is the W isotopic composition affected by the formation of the Fe meteorite or is it a signature of earlier processing.

To address these issues, the W isotopic composition of meteoritic zircons has been addressed through analysis with the SHRIMP RG. Zircon has high concentrations of Hf (typically 1-2 wt%) and has low levels of W making it ideal for the determination of initial 182Hf/180Hf. The zircons clearly show elevated 182W at high Hf/W. The relative difference between the inferred 182Hf/180Hf in the chondritic and eucritic zircons is consistent with the 10 Myr age difference between these different formation events. However, a specific calibration of Hf/W has proved difficult because of the absence of W in terrestrial (and meteoritic) zircon. This calibration is fundamental to the interpretation of the mineral isochrons in the meteorites and the interpretation of the whole-rock isochrons.


The significance of impacts on the early (3.8-4.0 Ga) Earth: Constraints from studies of coeval lunar (Serenetatis Basin) and terrestrial (West Greenland) samples

Vickie Bennett, Marc Norman and Allen Nutman

Much speculation exists as to the role of impacts in the development of early life. They have been variously accused of sterilizing the planet, promoting suitable environments for life development, or of providing the transport mechanism from other planetary bodies. The large impact basins on the Moon (Figure 1) are now generally considered to have been formed by a terminal cataclysmic bombardment at ca. 3.8-4.0 Ga. Coincidentally or not, the large nearside lunar basins are almost identical in age with the oldest terrestrial rocks, and are therefore relevant for consideration of the possible role of impacts in shaping the terrestrial continents and early life environments. Consideration of the higher cross section and greater gravitational focussing of the Earth requires that during this time period, the Earth as well as the Moon underwent significant meteorite bombardment. Owing to limited preservation and the pervasive tectonic overprinting and metamorphism which has affected all >3.6 Ga terrestrial samples, however, the record for the early Earth is difficult to decipher. In contrast lunar samples may represent a much more straightforward record of impact events and have much to tell us about events affecting the early Earth. What is not known, in both cases, are the numbers and compositions of the various impactors.

This past year, as an alternative approach to understanding the role of impacts in the early Earth as while as providing new data on lunar processes, we have initiated studies on lunar impact samples. This is a logical extension of our on-going program of integrated geochemical and geologic investigations of well-preserved, extensive 3.6 Ga->3.9 Ga terranes of south West Greenland.

Some of the clearest indicators of the presence and type of meteorite impactors are the amounts and relative proportions of the highly siderophile ("metal-loving") elements, most notably Re, Pt, Pd Os, Rh, Ru and Ir. Here at RSES, we have developed low blank, high precision methods for measurement of these analytically difficult elements even in low concentration (parts per trillion level) samples. The chemistry is based on isotope dilution combined with carius tube digestion and ion exchange separation of the HSE conducted in a purpose built, HEPA filtered clean laboratory. The resulting solutions are then analysed using an HP 7500 ICP-MS. At present we are one of the few labs in the world, and the only Australian laboratory to have this capability. Through the use of these techniques, we are able to obtain precise HSE signatures from both >3.8 Ga lunar and terrestrial samples. The goal of the lunar work is to use the recently recognized, distinctive signatures of different meteorite classes (representing potential impact materials) to determine the number and types of impactors represented in the lunar cratering record as represented by impact breccias collected during the Apollo 16 and 17 missions.

Our first lunar results are from Apollo 17 impact melt samples that all likely represent ejecta from the Serenitatis basin. Ar-Ar ages are consistent with the formation of these breccias in a single impact event at 3893 ± 9 Ma. HSE from 11 representative samples have W-shaped patterns on CI chondrite -normalized diagrams, with enrichments in Re, Ru and Pd relative to Ir and Pt, and absolute abundances ranging from ~0.5 to 4% of CI chondrite reference values. Stronger depletions of Ir and Pt relative to Re, Ru, and Pd are correlated with decreasing HSE concentrations. The samples with the highest HSE concentrations have patterns that are identical to those of EH chondrites, but the patterns become increasingly less diagnostic of meteorite group with decreasing concentrations. An additional sample (77035) has a distinctly different HSE pattern and is more consistent with an impactor of ordinary or EL-type chondrite composition. It likely represents a discrete impact event that was sampled at the Apollo 17 site These data demonstrate that lunar impact melts contain signatures that link them to specific types of meteorite impactors that hit the Moon and provide new constraints on the types of materials that must have reached the Earth as well.


Figure 1: Lunar impact crater. The highly siderophile element characteristics of melts from similar craters are being studied to determine the characteristics of the meteorite flux on the ancient Earth and Moon.

The second aspect of this project is the analysis of samples from >3.8 Ga well preserved terranes of southern West Greenland. This area includes, but is not limited to the Isua Supracrustal Belt and the Akilia Island "early life" locality. In contrast to a previous studies our early terrestrial work is focussing on the HSE compositions of the oldest preserved (>3.8 Ga) mantle materials rather than on metasediments. Whilst early Archean metasediments can potentially record meteoritic fluxes, it would be extremely fortuitous for the key layers to be preserved and sampled in these rare and highly altered sequences. The HSE component within the Earth's upper mantle must have been added subsequent to core formation and thereby provides a time integrated average of late meteoritic additions to the Earth. The HSE compositions of the oldest mantle samples therefore could provide a monitor of meteoritic contributions.


Rhenium Systematics in Lau Backarc Basin Basalts from in situ Laser Ablation ICP-MS Analysis of Glasses

W. Sun, V.C. Bennett, S. Eggins, T. Falloon

Recent results from the long-lived 187Re-187Os isotopic system are providing new insights into the chemical evolution of the Earth. However, the fuller application of this system is strongly limited by the lack of knowledge of the global budgets of the parent element Re. For example studies on eclogites and blueschists show that large amounts of Re may be lost from the oceanic slab during subduction. The most straightforward destination of the "lost Re" should be the mantle wedge and subsequently, arc and/or back arc volcanics. This is not supported by the published Re contents of arc volcanic rocks which are much lower than expected. In order to further understand the behavior of Re in the arc environment we have undertaken a study of Lau Backarc Basin basalts.

The approach used here differs from prior work in two important ways. Firstly we analysed only submarine basaltic glasses obtained by shipboard dredging, rather than samples that had been erupted sub-aerially. Previous work on Hawaiian samples (Bennett et al. 2000) suggests that Re may be lost during degassing on eruption leading to an underestimate of Re concentrations. This is likely to an even more significant problem for volatile rich arc basalts. Secondly, Re concentrations were determined in the glasses in situ using an excimer laser coupled to a HP 7500 quadrapole ICP-MS. This ensured that only fresh glassy material without micro-phenocrysts was analysed. Submarine MORB glasses were also analyzed for comparison. The Re contents of samples from the Lau Basin spreading centers (1.2 to 1.6 ppb) are higher than published concentrations for both MORB and arc basalts. This suggests that literature averages do significantly underestimate the Re contents in arcs. The high Re contents however cannot be explained by simple addition of slab components. For example Re abundances do not correlate with other fluid mobile elements such as Pb. Additionally the systematic correlations between Re and other moderately compatible elements such as Yb (Figure 2), suggest that Re was not controlled directly by subduction released materials. Rather, it is likely controlled by melting processes.

The systematic behavior of Re with Yb as well as other heavy REE during melting has been well documented for many sample suites (Figure 2). However the Lau Basin basalts show a much steeper correlation than MORB and with a slope similar to that of the komatiite trend. The difference between MORB and komatiite has been previously explained by the presence and absence of residual sulphide, respectively. This results from Re being a chalcophile element whereas Yb is lithophile such that when there is no sulphide in the residue, Re behaves more incompatibly. It is thus likely that conditions in the arc mantle wedge such as higher oxygen fugacity coupled with the involvement of hydrous fluids led to smaller amounts of residual sulphide during melting.

The Lau Basin samples also exhibit regional differences with the Re contents of Eastern Lau Spreading center basalts systematically higher than those from the other areas. As the Eastern Lau Spreading center is closer to the Tonga arc, the higher Re contents could reflect a complex enrichment of the local mantle source by subduction released materials, or that larger amounts of S are dissolved in the melts as the result of the greater involvement of hydrous fluids derived from the nearby slab.



Figure 2: Re versus Yb diagram for the Lau Backarc Basin and MORB submarine basaltic glass samples. The Lau Basin samples define a much steeper slope than do mid-ocean ridge basalts. This steep slope is more komatiites arrays suggesting derivation from a source with low residual sulphide.


 Searching for an Archaean Alps in Greenland: An ancient continent-continent collision

Nutman A.P. and Friend C.R.L.

Plate tectonics is geology’s unifying theory, providing the framework for understanding the modern Earth. Notably, many geologically recent (past ~1 Ga, billion years) mountain belts are recognised to be collision zones between once distant parts of Earth’s crust. Archaean (pre-2.5 Ga) belts of deformed and metamorphosed rocks might be the sites of continental-continental and/or volcanic island arc collisions developed under an ancient regime resembling modern plate tectonics, or might have formed by different processes. Whether or not there is evidence of Archaean plate tectonic processes is important for understanding both crustal evolution and the ancient mineralisations that are of great economic significance for Australia.

In plate tectonics, collision and accretion at convergent plate boundaries produces regions of crustal blocks (terranes) of different age and history bounded by faults/mylonites. At Archaean higher crustal levels, prime evidence for collisional/accretionary tectonics has come from collages of tectonically-bounded packages of unrelated rocks in low to medium metamorphic grade granite-greenstone terranes, particularly the Superior Province of Canada, and parts of the Yilgarn Craton of Western Australia. Archaean high grade gneiss complexes, with upper amphibolite to granulite facies metamorphism at medium to high pressures (6 to 10 kbar) contain a record of deeper Archaean crust. The superbly exposed high grade gneiss complex in Godthåbsfjord, West Greenland with 1.5 km of topography, shows juxtaposition of unrelated amphibolite-granulite facies gneiss complexes (terranes) along high grade mylonites. Such discoveries make it the world’s best candidate for a well-exposed, deep section through an Archaean (at ~2.7 Ga) continent-continent collision zone. However the case for plate tectonic style collisions to explain the Archaean evolution of areas such a Godthåbsfjord is weakened by the lack of convincing evidence of suture zones (junction between unrelated parts of the crust once far apart). From Alpine analogies, the most diagnostic would be finding that unrelated terranes of crustal rocks are separated by a tectonic panel with lenses of upper mantle ± exhumed crustal rocks with relicts of high pressure metamorphism. Panels of such rocks represent vestiges of oceanic lithosphere and sediments, trapped between colliding blocks of continental crust.

In 2000 in Godthåbsfjord, we discovered of <200 m thick tectonic panel, containing late Archaean massive ultramafic rocks, lenses of gabbro, metabasalt and metasediment. This panel is bounded on its top and bottom by folded, metamorphosed mylonites. The mylonites separate the panel from extensive blocks of gneiss structurally above and below of different ages, and which did not have a common metamorphic history until the late Archaean. A deformed, but non-mylonitic pegmatite intruded along a mylonite has yielded a SHRIMP U/Pb zircon date of 2.7 Ga. In 2001, we found similar ultramafic rocks in the panel’s likely extension up to 20 km away. So far, the panel seems to form an extensive sheet no more than a couple of hundred metres thick, between the two different gneiss blocks. Preliminary geochemical data (Mg/Si versus Al/Si) suggest the ultramafic rocks are candidates for abyssal peridotites. This panel is the best candidate yet for the suture zone in the Godthåbsfjord region, making it more promising that this region contains a late Archaean collisional orogen, formed like the European Alps.

This panel will be the focus of further planned research, such as geochemical studies to confirm or deny the abyssal peridotite affinity of the ultramafic rocks. Another key indicator of a suture panel would be to find relicts of high pressure metamorphism, because, as in the Alpine model, the rocks in it would have been exhumed from greater depth. Even in modern collisional settings, high pressure assembles are rare, because of large-scale re-equilibration at lower pressures during their exhumation. In Godthåbsfjord, the difficulty of finding relict high pressure assemblages will be compounded by recrystallisation during further ductile deformation and amphibolite facies metamorphism in the period 2.7-2.5 Ga. Therefore we hold-out very little hope of actually finding relict eclogites. Instead we will employ a novel zircon geochemical method developed by Dr D. Rubatto which will allow us to determine if any metamorphic zircons in the mafic rocks in this panel formed as part of plagioclase-free (eclogitic) assemblages. Both the ultramafic geochemistry and metamorphic investigations are to address criticism over lack of undeniable evidence of sutures in Archaean orogens and to understand the nature of Archaean tectonic processes.


Zircon formation during fluid circulation in eclogites (Monviso, Western Alps): implications for Zr and Hf budget in subduction zones

D. Rubatto, J. Hermann

Recent studies demonstrated that metamorphic zircon can form by recrystallisation, metamorphic reactions, dissolution/precipitation or even aqueous fluid circulation at variable pressure and temperature conditions. During high-pressure metamorphism, i.e. subduction, metamorphic zircon can form in a variety of rock types: it is commonly found in metasediments and in eclogites derived from ophiolitic gabbros, especially highly differentiated Fe-Ti gabbros and leucogabbros. In addition to dating, the study of zircon in these rocks bears important information for recycling of Zr, Hf and U in subduction zone processes. Subduction zone magmas are characterised by a depletion of HFSE (Zr, Nb, Hf, Ta) with respect to REE of similar incompatibility. These features can be explained either by processes dominated by the mantle wedge or by fractionation during liquid extraction from the subducted oceanic crust. Based on experimental partitioning between rutile and aqueous fluids/melts, it has been proposed that residual rutile in eclogites might be responsible for limited release of HFSE from the subduction zone. Up to now, the possible influence of residual zircon on the Zr and Hf budget in subducted crust are not well known. We address this issue using isotope and trace element data from zircon and associated minerals from eclogite-facies rocks.

The zircons from an eclogite and an enclosed eclogite-facies vein from the Monviso ophiolite (Western Alps) display contrasting chemical and morphological features and document different stages of the evolution of the ophiolite. The zircons from the eclogite, which was metamorphosed at 600°C and 20 kbar, are inherited from the protolith and do not display significant metamorphic recrystallisation. They have a typical magmatic zoning and trace element composition with strong enrichment of heavy rare earth elements (HREE) over middle rare earth elements (MREE), and an accentuated negative Eu-anomaly indicating formation in the presence of plagioclase. The age of these magmatic zircons documents the formation of oceanic crust at 163±2 Ma. Zircons from the eclogite-facies vein contain garnet, omphacite and rutile inclusions demonstrating that they precipitated under eclogite-facies conditions. These zircons are rather homogeneous in composition: they have Th/U ratios < 0.09, are generally depleted in trace elements, display no Eu-anomalies and only weak enrichment of HREE with respect to MREE, consistent with a garnet-bearing, plagioclase-free, i.e. eclogite-facies, paragenesis. Metamorphic vein zircons yield an age of 45.0±1.0 Ma providing evidence for Eocene subduction of the Monviso ophiolite.

In the vein, the contemporaneous formation of metamorphic zircon with eclogite-facies omphacite and garnet permits determination of a set of trace element distribution coefficients between these minerals at high pressure. This set of partitioning can be used to identify chemical equilibrium among these phases in high-pressure rocks where the textural relationships are less clear than in the studied vein. This is especially important to relate zircon formation, and hence age, to sensors of metamorphic conditions such as garnet and omphacite.

The presence of zircon and rutile in the vein demonstrates that high field strength elements (HFSE) are mobile at least over a short distance in aqueous fluids at eclogite-facies conditions. However, the concentrations of Zr and Hf in the aqueous fluid are estimated to be at least a factor 10 lower than primitive mantle values.

Mass balance calculations demonstrate that zircon hosts more than 90% of the bulk Zr and Hf and about 70% of U in the vein. Our new data combined with available literature data indicates that zircon is a residual phase in subducted basalts and sediments up to at least 800-900°C. Therefore, residual zircon in subducted crust together with rutile completely buffer the HFSE in liberated subduction zone fluids/melts and might be partly responsible for negative Zr and Hf anomalies in subduction zone magmas. However, the extremely low Zr and Hf contents of the eclogite-facies aqueous fluid, from which the vein minerals precipitated, indicate that aqueous fluids are not capable of enriching a mantle source in these elements. Thus, subduction zone magmas showing a slab HFSE component require metasomatism of the mantle wedge by hydrous melts, which are able to transport significant amounts of such trace elements.


Further evidence that eastern Australian S-type granites are derived from early Paleozoic sediments

I.S. Williams


Geochemical studies over many years have demonstrated that the S-type granites of southeastern Australia are derived predominantly from source rocks that have been chemically fractionated by weathering, namely sediments. Initially it was suggested that those sediments were deeply-buried Proterozoic continental crust, but as more work was done on zircon from the granites and their host rocks, it became evident that both the granites and the extensive Paleozoic flysch into which they were intruded share a common provenance. The question is now whether the granites might be the product of partial melting of the flysch itself.

Inherited zircon, namely zircon that is significantly older than its igneous host rock, is common in the granites of southeastern Australia. It usually occurs, not as discrete grains, but as cores surrounded by an overgrowth of younger zircon precipitated from the melt phase of the magma, and varies widely in its relative abundance. Inherited zircon is extremely rare in the more mafic metaluminous rocks, such as gabbros and diorites, and generally scarce but ubiquitous in the more felsic metaluminous rocks such as the majority of the I-type tonalites, granodiorites and monzogranites. In contrast, it constitutes a major fraction of the total zircon content of the S-type granites, particularly the mafic S-type granites, in which virtually every zircon grain contains a large inherited core.

These differences in abundance are governed by several factors, for example the abundance of zircon in the magma protolith, the capacity of the magma to incorporate zircon-bearing country rock during its ascent, and the solubility of such assimilated zircon in the magma. High-temperature metaluminous magmas have the greatest power to assimilate, but it is just such magmas in which assimilated zircon is most likely to dissolve. In contrast, zircon is relatively insoluble in low-temperature peraluminous melts with little assimilation power, so zircon from the magma protolith has a greater chance of being preserved uncontaminated as inheritance in the resulting granite.

Detrital zircon in sedimentary rocks commonly survives high grade metamorphism, simply being overgrown by new zircon precipitated during the metamorphic event. If sufficient partial melt develops for a magma to mobilise, then the former detrital zircon is preserved in that magma as zircon inheritance. Case studies of progressive metamorphism indicate that incorporation of detrital zircon into an S-type magma has a negligible effect on the zircon U-Pb isotopic systems. Once it is recognised that the inheritance in S-type granites is predominantly source-derived, then the inheritance pattern in those granites becomes a direct guide to their possible source rocks.

A feature of the S-type granites in southeastern Australia is the close similarity between their inheritance age patterns and the detrital zircon age patterns in the enclosing early Paleozoic flysch. If part of the flysch sequence itself was the source of the magmas, then the age of the youngest inherited component is of particular interest because it places an upper limit on the deposition age of the source metasediment. Young inherited zircon is relatively rare, however, and there is always the question of whether the apparent ages of such cores have been reduced by partial loss of radiogenic Pb. By combining cathodoluminescence imaging of the zircon internal structures with the observation that inherited zircon in S-type granites commonly preserves the external morphology and growth zoning of the original detrital grains, it has recently been possible to identify, and use SHRIMP to date, several very young inherited zircon cores in one of the most inheritance-rich S-type granites in the region, the Jillamatong Granodiorite. The ages, most confirmed by replicate analysis, extend to just below the age of the Cambro-Ordovician boundary, indicating not only that the granite was derived from Paleozoic, not Proterozoic sediment, but also that that sediment was deposited after the Cambrian sediments that in places can be seen to form the base of the flysch pile. The inherited zircon is readily explained without invoking the presence of Proterozoic continental crust beneath southeastern Australia.


The Svecofennian Province of southern Finland— accretionary arc complex or marginal basin?

R.W.R. Rutland and I.S. Williams

The extensive mid-Proterozoic Svecofennian Province covers over 300,000 km2 of southwestern Finland and eastern Sweden. In recent decades this vast region, dominated by migmatised greywackes and volcanic rocks intruded by voluminous Proterozoic granites, has been interpreted to represent a sequence of juvenile island arcs and subduction zones that were successively accreted to the margin of the Archean Baltic Shield to the north. This concept is now being called into question, however. Firstly, the greywackes contain zircon much older than the associated volcanic rocks, and secondly, the igneous rocks themselves are much more evolved chemically and isotopically than would be expected had they been produced in a primitive arc environment. The alternative suggestion is that the greywackes of the Svecofennian Province are in fact remnants of the fill of a large basin developed at a continental margin, akin to the Paleozoic Lachlan Fold Belt in eastern Australia. The volcanic rocks would thereby be the preserved remains of extensional volcanism which followed closure of that basin.

A premise of this alternative view is that the greywacke sequence predates, rather than is contemporaneous with, the volcanic succession. This is not a novel idea—Sederholm, as early as 1897, proposed that the two sequences were separated by a great unconformity. The difficulty in testing the hypothesis, however, is that the sequences have been extensively metamorphosed, commonly to granulite facies, so structural and isotopic evidence of their earlier history is extremely difficult to recover.

An experiment is currently in progress to try and establish the deposition age and thermal history of the Svecofennian greywackes using SHRIMP ion microprobe analysis of individual zircon and monazite grains. The work is being done on a small number of samples carefully selected on structural criteria to be as free as possible of the effects of the late metamorphic overprint. Early results are encouraging. Consistent with earlier work, preserved detrital zircon shows that the greywackes are derived from sources dominated by rocks up to 100 million years older than the overlying volcanic sequence. More importantly, there is evidence in the tiny overgrowths on individual zircon grains, that not only were the source rocks in places metamorphosed to high grade before erosion, but some of the greywackes were metamorphosed to the point of partial melting about 60 million years before being metamorphosed again by the thermal overprint associated with the late igneous activity. So pervasive is the late overprint that in most cases evidence of the early metamorphism is preserved only by zircon, but in one case, evidence for this main metamorphism is also preserved by monazite (Figure 3). These results strongly support the proposition that the migmatitic basement was deformed and metamorphosed before the overlying volcanic sequence was deposited, and suggest that the Svecofennian Province formed in a tectonic setting very different from that commonly assumed.



Figure 3: Relative probability histograms of the Proterozoic zircon and monazite ages from selected Finnish Svecofennian metasediments showing the presence of minerals recording three thermal episodes. The oldest (c.1980 Ma) is represented by detrital grains, the younger two (c.1920 and c. 1880 Ma) are represented by post-deposition metamorphic mineral growth.


Carboniferous and Triassic eclogites in the western Dabie Mountains, east-central China: evidence for protracted convergence of the North and South China Blocks

Weidong Sun, Ian S. Williams, Shuguang Li

SHRIMP U-Pb dating and laser ablation ICP-MS trace element analyses of zircon from four eclogite samples from the northwestern Dabie Mountains, central China, provide evidence for two eclogite facies metamorphic events. Three samples from the Huwan shear zone yield indistinguishable late Carboniferous metamorphic ages of 312 ± 5, 307 ± 4 and 311 ± 17 Ma, with a mean age of 309 ± 3 Ma (e.g. Figure 4). One sample from the Hong’an Group 1 km south of the shear zone yields a late Triassic age of 232 ± 10 Ma, similar to the age of ultra-high pressure (UHP) metamorphism in the east Qinling-Dabie orogenic belt. REE and other trace element compositions of the zircon from two of the Huwan samples indicate metamorphic zircon growth in the presence of garnet but not plagioclase, namely in the eclogite facies, an interpretation supported by the presence of garnet, omphacite and phengite inclusions. Zircon also grew during later retrogression. Zircon cores from the Huwan shear zone have Ordovician to Devonian (350-440 Ma) ages, flat to steep heavy-REE patterns, negative Eu anomalies, and in some cases plagioclase inclusions, indicative of derivation from North China Block igneous and low pressure metamorphic source rocks (Figure 4). Cores from the Hong’an Group zircons are Neoproterozoic (610-780 Ma), consistent with derivation from the South China Block. In the western Dabie Mountains, the first stage of the collision between the North and South China Blocks took place in the Carboniferous along a suture north of the Huwan shear zone. The major Triassic continent-continent collision occurred along a suture at the southern boundary of the shear zone. The first collision produced local eclogite facies metamorphism in the Huwan shear zone. The second produced widespread eclogite facies metamorphism throughout the Dabie Mountains – Sulu terrane and a lower grade overprint in the shear zone.


 Figure 4: Zircon U-Pb isotopic analyses, chondrite-normalised REE patterns, representative CL images of zircon structures and inclusions of high pressure minerals (garnet, omphacite and phengite) restricted to the overgrowths, indicative of late zircon growth under eclogite facies conditions from Xiongdian eclogite 99XD-1 (garnet quartzite) from the Huwan shear zone. Analytical uncertainties 1s .
  • Geochronological correlation within the Curnamona Province of southern Australia, and links to Mount Isa and other north Australian Pb-Zn bearing successions

    R.W. Page

    The advent of a more robust temporal framework (reported in RSES Annual Report 2000) for the Palaeoproterozoic history of the southern Curnamona Province provides the basis for scrutinising lithostratigraphic and sequence stratigraphic correlations on a regional and inter-basin scale. The Willyama Supergroup sequence in the Curnamona Province is host to Broken Hill’s Ag-Pb-Zn orebody. Improved correlation across the Province, from the Olary Domain (eastern South Australia) to the Broken Hill Domain (western NSW) allows comparisons in this sequence that generally affirm earlier lithostratigraphic correlation. This includes the equivalence of 1710-1720 Ma Curnamona and Thackaringa Groups, and the recognition of a coeval 1700-1710 Ma felsic magmatic event in the Olary and Broken Hill Domains.

    New SHRIMP U-Pb ages from several tuffaceous sediments immediately above the calcareous and gossanous, base-metal enriched Bimba Formation in the Olary Domain are remarkably consistent on a regional scale. They define the depositional age of this horizon at 1693±3 Ma. This helps resolve a longstanding question as to which part of Olary sequence is correlated with the ~1690 Ma-Broken Hill Group. The base, middle, and upper part of the latter are defined by ages in the Ettlewood Calc-Silicate Member (1693±4 Ma), Parnell Formation (1693±5 Ma), and Hores Gneiss (1685±3 Ma). The new ages reported here and elsewhere thus substantiate earlier lithostratigraphic correlations between the lower Broken Hill Group and the lower Strathearn Group.

    These correlations, together with the potential importance of syn-depositional base-metal ore formation, call for closer comparison with other Palaeoproterozoic Pb-Zn mineralised sequences — especially the ‘Carpentaria Zinc Belt’ of northern Australia. The gross lithostratigraphy, depositional ages, and possible hiatuses within the Curnamona Province have parallels and provide a quantitative basis for correlation with basin development in Mount Isa and McArthur Basin. The paucity of rocks equivalent in age to Broken Hill Group is possibly linked to the widespread 'Gun' erosional event in northern Australia between 1690 Ma and 1665 Ma ago. However, in the upper Willyama Supergroup, a clear age correlation with northern Australian sequences emerges. This derives from (a) close comparison of middle Paragon Group ages with tuff ages associated with the Mount Isa and Hilton orebodies at ca. 1655 Ma, and (b) similarly comparable ages of ca. 1640 Ma for the upper Paragon Group and Isan Superbasin hosts of the HYC deposit of the McArthur Basin, the Walford deposit, and other mineralised sequences of Lawn Hill Platform. Revisions and improvements to the age and stratigraphic framework of the Willyama Supergroup, and links that emerge with northern Australia sequences might provide a fresh basis and refocus of exploration efforts in the Curnamona Province.


    Age and stratigraphic placement of new hominid genus from Kenya

    I. McDougall, F.H. Brown, P.N. Gathogo5, M.G. Leakey, and F. Spoor

    In March, 2001, Leakey et al. reported in Nature the finding of fossils assigned to a new genus of hominid, Kenyanthropus platyops, with important implications for our understanding of early human ancestry. The fossils were found in near flat-lying sediments of the Nachukui Formation in the Turkana Basin, northern Kenya, just to the west of Lake Turkana in the Lomekwi drainage. The stratigraphic placement of the fossils was undertaken by F.H. Brown and P.N. Gathogo and most of the numerical age measurements on associated tuffaceous horizons were made at ANU. The type specimen of the new genus, a nearly complete cranium, is from a horizon 3.5 Ma old, and additional jaws and teeth, assigned to the same genus, are from sediments deposited between about 3.5 and 3.3 Ma ago. The new fossils are distinctly different from those of Australopithecus afarensis, which are of similar age and from the Hadar region in Ethiopia. The new finds indicate that as far back as 3.5 Ma ago there were at least two lineages that could have ultimately given rise to modern humans, so that the early stages of human evolution are more complex than previously thought.

    Stratigraphic sections shown in Figure 5 also indicate the level from which the hominid fossils were recovered. Sample 4000 refers to the type specimen of Kenyanthropus platyops within the sequence comprised mainly of sands, salts and clays deposited in fluvial and lake-margin environments. Interbedded with the normal detrital sediments are several tuffaceous horizons. These tuffs are the key to the stratigraphy as their identification through chemical fingerprinting and their widespread distribution throughout the Turkana Basin has enabled a robust relative time framework to be established. In addition, isotopic age measurements on alkali feldspar crystals in small pumice clasts in some of the tuffs, has provided numerical ages for the explosive eruptions that produced the tuffs. Deposition of the tuffs is considered to have occurred very soon after their explosive eruption, so that the age of eruption in most cases approximates closely that of deposition also. The 40Ar/39Ar dating of single crystals of feldspar from pumice clasts in the Topernawi and Moiti tuffs yielded precise and concordant ages for each of the two horizons, actually indistinguishable in age at 3.96 ± 0.03 and 3.94 ± 0.03 Ma, respectively (Figure 5). The Lokochot Tuff has not been directly dated, but an age of 3.57 Ma is derived by stratigraphic interpolation, and the estimated age of the Tulu Bor Tuff of 3.40 ± 0.03 Ma is derived from earlier measurements by other workers on a correlative of the tuff in Ethiopia. Together these results provide good control, with an interpolated age of 3.5 Ma for the type specimen of the new hominid. From the stratigraphy and the age of tuffs higher in the sequence in the Turkana Basin, the fossils found above the Tulu Bor Tuff in the area are regarded as no younger than about 3.3 Ma.

    This study is a further excellent example of the value of an integrated interdisciplinary approach involving detailed stratigraphic mapping and geochronology done in conjunction with the palaeontological exploration. This has lead to remarkably tight relative and numerical age control, facilitating discussion of the evolution of hominids toward modern humans relatively free of some of the dating problems associated with hominid finds of decades past.




     Figure 5: Stratigraphic sections and placement of hominid specimens in the Lomekwi drainage, west of Lake Turkana, northern Kenya. Reproduced from Leakey et al. (Nature, 410, 433-440, 2001)


    Exposure dating in young lava flows using cosmogenic neon-21

    D. Gillen, M. Honda and A. Chivas3

    A study was designed to further develop the relatively new method of cosmogenic exposure dating with noble-gas isotopes, particularly neon-21. The motivation for the study was the need for a reliable way of dating exposed surfaces in relation to a variety of geomorphological problems, and possibly measurement of minimum eruption ages for young volcanic rocks that have never been covered.

    Cosmogenic neon-21 was used to determine exposure ages of young basaltic lava flows from the Newer Volcanic Province, southwestern Victoria. Ages determined from neon-21 analyses in olivine showed correlation with exposure ages previously determined by radiocarbon dating and cosmogenic chlorine-36 exposure dating. An age of about 20 ka for the Tyrendarra flow agreed with C-14 results; the Harman Valley flow from Mount Napier returned ages of 20 and 27 ka which show agreement within uncertainties with both Cl-36 and conventional C-14 ages; ages of 22 and 26 ka determined for Mount Porndon were much younger than the corresponding Cl-36 age of 59 ka, indicating some uncertainty over this age. Helium-3 analyses were also attempted but low concentrations relative to detection limits of the VG5400 noble gas mass spectrometer caused large uncertainties in the measurements, so the resulting exposure ages were considered to be less reliable.

    The achievements of this study include the obtaining of meaningful exposure ages for such young basalts, and the reasonable agreement of these ages with C-14 and Cl-36 ages. These outcomes help to validate this dating technique, and continuing study in this area is important to further develop this approach.


    Noble Gases in black bort diamonds from Jwaneng, Botswana

    M. Honda, D. Phillips and J. W. Harris2

    Research goals in noble gas geochemistry include understanding the structure of the Earth’s mantle and the creation of a coherent model of its evolution. In this regard, noble gas compositions in mid-ocean-ridge basalts (MORBs) and in ocean island basalts (OIBs) such as from Loihi Seamount, Hawaii, and Iceland, have provided very useful information on the mantle. However, virtually all these data are from samples that are effectively of zero-age, and therefore, they only give information about the present composition of mantle noble gases. It is critically important to determine if there is any systematic variation of mantle noble gas composition with time. In particular, in relation to the current debate about mantle structure and evolution (e.g. two layered convection vs. whole mantle convection), information on noble gas compositions in ancient mantle can provide very strong constraints. If noble gas measurements are made on mantle-derived samples of different ages, these can be used to evaluate as to what degree, if any, the upper mantle had interacted with the lower mantle, and allow further refinement of models concerning mass transport, including volatiles, in the mantle. However, attempts to determine the noble gas composition of ancient mantle by analysis of older geological samples have, with few exceptions, been unsuccessful, in part owing to the lack of suitable samples.

    Diamonds have unique characteristics that make them potentially very useful as sources of noble gases from the mantle. This is because: (1) diamonds have been demonstrated to retain significant amounts of mantle-derived noble gases, (2) most diamonds appear to be derived from 150 km to 200 km depth in the Earth, (3) diamonds cover a wide range of crystallization ages of between 0.6 and 3.5 billion years, (4) diamonds have been shown to have low diffusivities for noble gases so that they are highly retentive of noble gases, and (5) diamonds typically have suffered little interaction with crust or atmosphere, owing to their great crystallization depths and extremely rapid emplacement to shallow crustal levels in kimberlite and lamproite pipes. Thus, diamonds form under a wide range of conditions and compositions and provide a direct window into the ancient mantle. As such they are unique as sources of mantle-derived noble gases.

    As a pilot study, this year we have undertaken noble gas analysis of four black bort diamonds from the Jwaneng kimberlite pipe, Botswana, by stepwise heating and vacuum crushing. From these diamond samples we found xenon isotope anomalies, relative to atmospheric, and the xenon data lie on the MORB correlation line when they are plotted on the 129Xe/130 vs. 136Xe/130Xe diagram (Figure 6). In contrast, 40Ar/36Ar ratios in the samples are much lower than the MORB value of > 60,000. Neon isotopic ratios are close to atmospheric; there is no evidence for existence of MORB-like neon in the samples. These observations indicate that the MORB-like noble gas composition in the mantle source where the diamonds crystallized was overwhelmed by the lighter-gas enriched atmospheric noble gases. If this is correct, the question as to why elementally fractionated atmospheric noble gases have been introduced into the mantle source remains to be answered.

     Figure 6: Plot of 129Xe/130 versus 136Xe/130Xe observed in bort diamonds from the Jwaneng kimberlite, Botswana. The xenon data from the Jwaneng diamonds lie on the MORB correlation line determined by Kunz et al. (1998). Some of the Jwaneng diamonds show the highest xenon isotope anomalies, relative to atmospheric, so far observed from mantle-derived samples.


    Thermal evidence for Early Cretaceous metamorphism in the Shyok suture zone and age of the Khardung volcanic rocks, Ladakh, India

    W.J. Dunlap and R. Wysoczanski

    The Dras island arc (NW India) is intruded by the Ladakh Batholith and rimmed along its southern margin by the Indus suture zone (Figure 7), which developed ca. 50 Ma at the start of the India-Asia collision. Along its northern margin the Ladakh Batholith intrudes the Shyok Formation, a series of folded and faulted metasedimentary and metavolcanic rocks that are thought to mark an older suture of Cretaceous age. Restoration of Miocene and younger strike-slip movement of ~150 km on the Karakoram fault suggests that the Shiquanhe suture in China was once continuous with the Shyok suture in Kohistan, but no geochronologic evidence for this connection has been demonstrated in the intervening region in Ladakh.



    Figure 7: Tectonic cartoons, in cross-sectional form, of the region where the Karakoram fault zone has formed, in the vicinity of the Shyok-Shiquanhe and Indus suture zones. The figure is centred on the Ladakh sector of the Dras island arc. Schematic interpretive cross-sections of the region shown at ca. 140 Ma and 120 Ma (partially after Matte et al., 1996): acretionary prisms shown in black, KV is future site of deposition of Khardung Volcanic rocks from about 67.4 Ma to 60.5 Ma.


    The Khardung calc-alkaline volcanic rocks were deposited unconformably on the Shyok Formation and are thought to be of Late Cretaceous age on the basis of fossils and regional correlations, yet no reliable radiometric ages have been published. New SHRIMP U/Pb ages on single zircon grains from Khardung volcanic rocks have confirmed that a ~7 km thick section was deposited between 67.4 Ma and 60.5 Ma. The underlying Shyok Formation has been difficult to date due to strong thermal overprinting related to both intrusion by the ca. 102-50 Ma Ladakh granites and movement on the younger Karakoram fault. Near Digar a series of metasedimentary and metavolcanic rocks in structural and metamorphic continuity with the Shyok Formation has experienced less thermal overprinting and a muscovite from a marble unit yields a 40Ar/39Ar maximum age ca. 124 Ma, which indicates that greenschist facies metamorphism took place prior to this time. The geochronological evidence is consistent with an Early Cretaceous age for the Shyok Formation, but it further suggests an Early Cretaceous metamorphic and deformational event related to convergence in an oceanic arc setting between the Dras island arc and the Shiquanhe island arc. This metamorphism was followed in the Late Cretaceous by suturing of the Dras island arc to the continental rocks of the Qiangtang block in westernmost Tibet along the Bangong suture.




    Research staff from Geoscience Australia (GA, formerly Australian Geological Survey Organisation) currently work within RSES mostly on U-Pb SHRIMP geochronology. This capability was recently enhanced and diversified by expanding into Ar-Ar geochronology. The geochronologists principally work in the Minerals Division within a variety of regional projects spread over several provinces across Australia (Yilgarn, Gawler, Tanami-Arunta, Broken Hill, Mt Isa, and Tasmania). Andrew Cross has recently been appointed to GA as a SHRIMP geochronologist.

    GA research is based on the longstanding relationship with the Research School, in particular within the Geochronology and Isotope Geochemistry Group. The scientific outcomes address GA's role in Minerals Promotions under the National Geoscience Agreement (NGA), Petroleum Promotions, and the Predictive Minerals Discovery Cooperative Research Centre (pmd*CRC). A selected range of research activities from these projects is described below.


    40Ar/39Ar geochronology

    Geoff Fraser

    In mid 2000, Geoscience Australia and the Northern Territory Geological Survey commenced a joint project investigating gold mineralisation in the Tanami Region in the context of the geological development of the North Australian Craton. The Tanami Region, located 600 km northwest of Alice Springs, is one of the most important new gold provinces in Australia. It straddles the Northern Territory-Western Australia borders along the southern margin of the Palaeoproterozoic North Australian Craton. The Tanami Region contains over 50 gold occurrences, including three established goldfields (Dead Bullock Soak, The Granites and Tanami), as well as several significant gold prospects (Groundrush, Titania, Crusade, Coyote, Kookaburra). The region has produced 4.1 Moz Au and the remaining resource is ~8.4 Moz (260 t) Au. This figure is steadily growing as a result of extensive exploration.

    Considerable uncertainty surrounds the age of gold mineralisation in the Tanami Region. Prior to the current study, understanding of its geological history held that a phase of major tectonism, referred to as the Tanami Event, occurred between ~1848 and 1825 Ma, immediately followed by rifting, volcanism and granite intrusion in the period 1825 to 1815 Ma. Several of the Tanami gold deposits exhibit a spatial relationship between mineralisation and granitoids, which has led to the proposal of a genetic link between granitoid intrusion and mineralisation; this proposal is currently being tested with both U-Pb and 40Ar/39Ar geochronology. In contrast, the Callie deposit does not appear to be spatially related to intrusives, raising questions about the role of granitoids in gold mineralising events.

    Two samples of biotite from gold-bearing quartz veins in the Callie deposit yield 40Ar/39Ar age spectra in which the majority of the gas has apparent ages in the range ~1670 to 1720 Ma. Textural evidence clearly suggests that these biotites crystallised synchronously with the quartz veins in which they are hosted, which in turn are interpreted as synchronous with ore mineralisation. Evidence from fluid inclusions suggests vein formation temperatures were likely in the range 310 to 330°C, similar to typical biotite closure temperatures for Ar diffusion. It is, therefore, likely that the 40Ar/39Ar ages preserved in vein-biotite closely approximate the timing of vein crystallisation and associated ore formation.

    The limited amount of 40Ar/39Ar data currently available is consistent with at least some of the gold mineralisation in the Tanami Region significantly post-dating the Tanami Event. In the case of the Callie Deposit the data are suggestive of a link with the Strangways Event, (1720 – 1730 Ma, Collins and Shaw, 1995) responsible for widespread deformation and metamorphism in the Arunta Province several hundred kilometres south-west of the Tanami Region. Work is currently in progress to further evaluate the possible influence of Strangways age tectonic activity in the Tanami region, and the extent to which this younger event may have been responsible for gold mineralisation.


    The major building blocks of Tasmania, as identified by SHRIMP U-Pb zircon dating

    L.P. Black11, 12

    This study is part of Geoscience Australia’s TASmanian Mapping Accord Project (TASMAP). Based on geological mapping and aeromagnetic images, pre-Late Carboniferous Tasmania has been divided into seven different strato-tectonic elements – King Island, Rocky Cape, Dundas, Sheffield, Tyennan, Adamsfield-Jubilee and Northeast Tasmania. Each of these is believed to have a geological history and internal structure that is at least partly different from those of the other elements. SHRIMP U-Pb dating is being used to better define the geological histories of these elements, in an attempt to understand the significance of their inter-relationships. It is hoped that this information will allow us to determine whether the elements have always been in their present juxtaposition, whether they are reassembled components of what was a single terrane, or whether they represent the aggregation of originally unrelated terranes.

    Recent and current geochronological effort is largely concentrating on the dating of zircon from Devonian-Carboniferous granites, which occur in six of the seven elements. One facet of this exercise is to establish from comagmatic zircon the emplacement ages of those granites, and their inter-relationships. In addition, the zircon that these granites have brought up from depth provides evidence of the age of components within the underlying crust. Each of these datasets therefore helps better constrain different temporal and spatial aspects of the geological evolution of the individual elements.

    The new emplacement ages readily demonstrate the relative resistance of the U-Pb system in zircon to isotopic resetting. Thus, previously obtained Rb-Sr and K-Ar mineral ages tend to be somewhat younger, even though emplacement was at a high crustal level where subsequent geological effects had been interpreted as being insignificant. There was a general westward progression of igneous activity over about 50 million years (from the Early Devonian to the earliest Carboniferous). The youngest granites in the Northeast Tasmania element are essentially coeval with the oldest granites in the western elements. Both I-and S-type granites (including Heemskirk red and Heemskirk white varieties) in the latter elements share a common age of about 360 Ma (though I-type granites also formed both earlier and later than this). Coeval I- and S-type activity also appears to have occurred in northeast Tasmania about 25 million years previously. A small pluton on King Island yields a relatively complex array of individual zircon ages, consistent with it representing a second attempt at granite formation at 350 Ma. Whereas elevated temperatures about 10 million years earlier had generated substantial granite production in what is now west coast Tasmania, that event is interpreted to have been insufficiently intense to have produced more than localised partial melting in the King Island element.

    Magmatic activity within the Blue Tier Batholith (BTB) of Northeast Tasmania lasted for about 23 million years. Unlike some of the previously reported Rb-Sr ages, all of the U-Pb zircon ages are consistent with field observations, which reveal that the hornblende-biotite granodiorites are older than the biotite granites, which predate the alkali-feldspar granites. The new data support a previous Rb-Sr-based conclusion that the mineralised Lottah Granite is distinctly younger than the Poimena Granite, and that the two are genetically unrelated to each other. Much of the zircon in a mafic enclave from The Gardens Granodiorite is of the same age as the host intrusive, an important but not necessarily diagnostic observation for deciding between the competing models currently being used for magma genesis in the Lachlan Fold Belt. Igneous activity within the adjacent Scottsdale batholith commenced about 10 million years after granite emplacement was initiated in the BTB, and lasted for less than 10 million years.

    A quantitative correlation between U-Pb zircon age and the presence or absence of regional foliation within Northeast Tasmania granites, allows that deformation to be dated at about 390 Ma.

    To properly understand the significance of inherited zircon components within the Devonian – Carboniferous granites, it is important to know just what zircon components are present in the enclosing metasedimentary sequences. Consequently, representative samples (often of quartzitic composition) of those sequences were collected from each of the seven different strato-tectonic elements. All six of the westerly elements have strikingly similar zircon age arrays. Only the Tyennan sample, which yields no zircon younger than about 1680 Ma, has a signature that appears to be slightly different from the others; zircon at least as young as about 1400 Ma is present in the other westerly elements. This could signify that the Tyennan element is older than the others, although it might merely reflect a sampling bias. There is no other chronological evidence to potentially distinguish between the six more westerly units. A dominance of 1700-1800 Ma zircon indicates that all of these rocks were primarily derived from Palaeoproterozoic source terranes. For all but the Tyennan sample, deposition clearly occurred after about 1400 Ma, and the presence of ~1200 Ma zircon at some locations suggests an even younger limit.

    The early Palaeozoic Mathinna Group of Northeast Tasmania yields a very different age pattern, one that is typical of the Lachlan Fold Belt on the Australian mainland. 500-600 Ma zircon is dominant, 800-1400 Ma zircon is less common, and minor quantities of older (including Archaean) zircon are also present.

    Several useful observations can be made on the inherited zircon within the Devonian-Carboniferous granites. First, there is no evidence of different age components in the I- and S-type granites, though the former generally contain less inheritance. Neither is there any obvious difference between the inheritance in the granites from any of the 6 western elements, once again suggesting these elements are at least broadly comparable. In contrast, there is a most pronounced difference between that array and the inheritance within the Northeast Tasmania granites.

    Importantly, the inheritance patterns for all the granites were found to closely mirror those of the rocks they intrude, to the extent that it is possible to tell whether or not a granite sample derives from Northeast Tasmania purely on its inherited zircon signature. Previously dated Neoproterozoic granites in western Tasmania have similar inheritance to their Devonian-Carboniferous counterparts, suggesting that the sources of the granites in that region did not significantly change over a 400 million year period. The data also indicate that those deep crustal source rocks in western Tasmania are considerably older than those in the northeast.


    Terrane recognition in the southern Arunta Province, central Australia

    J.C. Claoué-Long11,12 and A. Cross11,

    Geochronology effort in northern Australia has been strengthened this year by the Northern Territory Geological Survey (NTGS) funding the appointment of Andrew Cross to join Geoscience Australia’s geochronology group. NTGS is remapping the fundamental geology of northern Australia and is using SHRIMP U-Pb geochronology intensively to identify and correlate major rock packages over very wide areas.

    This year a major focus has been an area west of Alice Springs where NTGS geoscientists have been mapping Proterozoic rocks which are well exposed in a series of ranges in the southern Arunta Province. Extrapolation from existing mapping to the east had inferred that the Proterozoic units in the area are divided about a major east-west discontinuity, the Redbank Deformed Zone, thought to juxtapose the northern (older) and southern (younger) Arunta terranes. However, the extensive new dating coverage shows that the region is dominated by gneisses and granites formed in the period ca. 1680 – 1640 Ma, long after the ca. 1870-1770 Ma crust-forming events that created the Arunta terranes and the North Australian Craton to the north. Clearly this package of rocks developed independently from the main Arunta terrane, and the location of the structural junction with the North Australian Craton is yet to be determined.

    Rich information about the evolution of the south Arunta terrane has come from the age spectra of detrital zircon grains in metasediments, which preserve the event histories of the source regions from which the sediments were derived. A quartzite in the north of the region contains detritus with a wide range of formation ages from the late Archaean to ca. 1765 Ma, clearly sourced from the North Australian Craton to the north. In contrast, another sediment near the southern limit of the exposed region appears to be the product of a sediment dispersal system off the craton, because its detrital zircons have a very different age spectrum lacking the Archaean and earlier Proterozoic components. The main correspondence with the North Australian Craton lies in the development of thin zircon overgrowths in rocks throughout the region at ca. 1590 Ma, the age of the major Chewings thermal event documented in the Reynolds Range to the north; this cryptic evidence provides a clue to the time when the south and north Arunta terranes became juxtaposed.

    Three features distinguish this program of geochronology. First is the intensive use of modern isotopic dating in a relatively small region of Proterozoic rocks, enabling detailed correlations and event chronologies to be developed for the whole range of magmatic, metamorphic and sedimentary units. Second, the essential nature of microbeam sampling with SHRIMP in zircons that preserve 3 or even 4 stages of growth reflecting the prolonged episodic crystallisation of a complex magmatic and metamorphic terrane. Thirdly, close integration with the field and geophysical mapping process has converted structural observations into a dynamic interpretation of terrane evolution, and enabled mappers to enter their second field season armed with factual age correlations as a guide.