Petrogenesis of the mafic-ultramafic intrusions of the Mesoproterozoic Giles Event, Musgrave Province, central Australia
thesisposted on 03.03.2017 by Seubert, Roland Edgar Benedikt
In order to distinguish essays and pre-prints from academic theses, we have a separate category. These are often much longer text based documents than a paper.
The 1090–1040 Ma Giles Event in the Musgrave Province, central Australia, led to the formation of the Ngaanyatjarra Rift. Mafic-ultramafic layered intrusions formed contemporaneously with massive gabbroic intrusions within this tectonic setting. The Bell Rock Range, Latitude Hill and Wingellina Hills intrusions represent three of the main different types of the ‘G1’ layered intrusions in this region, and were sampled as representative examples covering much of the compositional spectrum of layered intrusive bodies in this area. The Bell Rock Range olivine gabbronoritic-troctolitic intrusion forms a large segment of the Mantamaru intrusion that was rapidly emplaced into the crust at a maximum depth of c. 10 km. In comparison, the Latitude Hill and Wingellina Hills gabbronoritic-ultramafic intrusions are smaller intrusive bodies containing large proportions of websterite cumulates and gabbronorites with intercumulus plagioclase. Several of the G1 layered intrusions are associated with massive ‘G2’ gabbroic intrusions that were locally contaminated by felsic magmas. Evidence for the previously proposed chronological order of emplacement in which G1 was followed by G2 is ambiguous: the boundary between G1 and G2 was revisited during this study. Hierarchical cluster analysis enabled the extraction of chemical elements that vary most between the G1 and G2 intrusions, with principle component analysis outlining natural clusters within the data. The clustering within the G1 and G2 compositions can be readily explained by simple cumulate effects, where the G1 layered intrusions lost late-stage melts and as a result became incompatible element depleted, whereas the G2 gabbros have compositions that closely represent original liquid compositions. This indicates that the G2 intrusions may have formed from the same parental magmas as the gabbronoritic-ultramafic G1 intrusions. In addition, this study indicates that linear discriminant analysis can be used to determine exploration vectors for Nebo-Babel style Ni-Cu-PGE targets in the area. Although initial rifting in this region was most likely passive (i.e. driven by plate dynamics) it is still unclear whether the parental magmas for these intrusions were derived from the asthenospheric mantle (potentially plume-related) or from the subcontinental lithospheric mantle. Derivations from a likely enriched mantle reservoir as well as a maximum melting depth of c. 75 km permit both interpretations. Partial mantle melting generated magmas that then underwent moderate crustal contamination before being rapidly emplaced at mid-crustal levels to form the intrusions in the study area. Recent advances in the classification of rift settings indicate that active and passive processes can occur together in the same setting. Thus, the involvement of mantle dynamics, at least for later stages of the Giles Event (potentially generating the Warakurna Large Igneous Province), cannot be ruled out. Most magmatic sulphides within these intrusions are Cu-rich, are located in interstitial spaces and formed by extensive fractionation rather than by assimilation of crustal S, indicating that G1 intrusions may be prospective for smaller but PGE-rich deposits. This is supported by the presence of Ni-rich olivine and pyroxene, indicating an early major sulphide segregation event at depth is unlikely; hence, the G1 intrusions are likely to be unprospective for large orthomagmatic Ni-Cu ore deposits.