Petrogenesis and geochronology of the North Australian Kalkarinji low-ti continental flood basalt province

Abstract

A voluminous outpouring of tholeiitic lava, the Kalkarinji low-Ti continental flood basalt (CFB) province, covered an extensive area of northern Australia in Cambrian times. It comprises the Antrim Plateau, Helen Springs, Nutwood Downs and Peaker Piker Volcanics in northern and north-western Australia and the Colless Volcanics in Queensland. To the south, the Kalkarinji province most likely includes the Boondawari dolerite and Table Hill Volcanics (Western Australia). The name Kalkarinji is used here for the first time in this context to describe this vast and poorly known CFB province. New high-precision 40 ArP9 Ar dating for feldspar separates from the Antrim Plateau and Helen Springs Volcanics yield an average radiogenic age of 507 ±4 Ma (2a) which places the Kalkarinji eruption event close to the Early Cambrian - Middle Cambrian boundary, and links it with the global Toyonian faunal mass extinction. The Kalkarinji basalts and dolerites are mineralogically typical of other continental tholeiites in that they are dominated by clinopyroxene (augite and pigeonite) and plagioclase, with lesser ilmenite, titanomagnetite, and primary and secondary quartz and potassium feldspar. They exhibit marked trace element and Sr, Nd and oxygen isotopic enrichments that resemble continental crust. Moreover, the basalts and dolerites show depletion in the HFSE, notably Ti, P and Nb relative to other incompatible elements, along with extreme enrichment in Th and U and high Rb/Ba. PGE abundances are low and PGE patterns are fractionated with high Pt, Pd and low Ir implying early sulphide segregation. Crystal fractionation was the dominant process responsible for the chemical evolution of the basalts from ~9 to 3 wt% MgO. This can be demonstrated using geochemical models based on plagioclase and clinopyroxene crystal-liquid partition coefficients determined for aphanitic Kalkarinji basalts in this study. Isotopic compositions and key crust-sensitive trace element ratios do not correlate with indices of fractionation, e.g.87Sr/86Sr, 6 180 and Ba/Rb show negligible variation over a range of Mg#, indicating that the crust-like signature of the Kalkarinji basalts was already present, prior to near surface crystal fractionation processes at compositions more primitive than 12 wt% MgO. Distinctive geochemical features of the North Australian felsic crust such as elevated Th/U ratios and high LREE are reflected in the geochemistry of the Kalkarinji basalts pointing to North Australian felsic crust as the likely contaminant. Trace element, Sr and Nd isotopic geochemical signatures for the majority of the Kalkarinji basalts can be reproduced by assimilation and fractional crystallisation involving ~ 10% contamination of a pi critic parental liquid by average Proterozoic North Australian felsic crust. Larger, and hence less realistic amounts of crustal contamination are required if other potential crustal compositions, such as global average upper crust, are used for trace element modelling; this highlights the importance of selecting the most appropriate contaminants for assimilation and fractional crystallisation modelling exercises. The distinctive low-Ti signature of the Kalkarinji basalts can be modelled by contamination of an NMORB picrite parental liquid, without the requirement for melting of depleted peridotite. Segregation of the parental picrite magma occurred at ~15-20 kbar, equivalent to depths ~50-70 km in the mantle. However, in Proterozoic times, 200 km thick, dense, eclogite-rich subcontinental lithosphere underlaid the future centre of Kalkarinji volcanism. The presence of such thick lithosphere presents great difficulties for mantle plume models because the plume cannot rise to realistic segregation depths beneath such a barrier. The plume centre would need to be located on the continental shelf, e.g. Bonaparte Gulf to the north, where the lithosphere is considerably thinner, however, this is hundreds of kilometres north of the inferred locus of Kalkarinji volcanicity. Alternatively, in order for parental picritic magmas to segregate from their source at 50-70 km, a substantial portion of the lithosphere would need to be removed by 508 Ma. It is possible that the anomalous ~90° rapid anticlockwise rotation of the Australian continent around the time of basalt eruption was a trigger for catastrophic failure of dense, unstable, eclogite-rich subcontinental lithosphere beneath the Halls Creek orogenic belt, the presumed locus of Kalkarinji volcanism. The sinking lithospheric delaminate would be replaced by an equal volume of upwelling asthenosphere. It is postulated that the upwelling asthenospheric diapir experienced partial melting assisted by degassing of volatiles released from the sinking lithosphere. This model can explain features of some CFB provinces that cannot be explained by mantle plumes including the absence of regional uplift and association between CFB provinces and cratonic margins. Show more