Evolution of the lithosphere beneath Tasmania and Western Norway
thesisposted on 28.03.2022, 11:54 authored by Eloise Ellen Beyer
Two suites of mantle-derived peridotites have been studied to evaluate the linkages between mantle geochemistry and crustal age in Proterozoic and Phanerozoic terrains.Spinel-bearing peridotite xenoliths entrained in Tertiary alkali basalts have been erupted through Proterozoic and Phanerozoic crust in northern Tasmania. The xenoliths can be divided into three major compositional groups. The first, and most widespread, group consists of spinel Iherzolites that are moderately fertile in terms of whole-rock Al₂O₃ and CaO contents and olivine Mg# and lie in, or adjacent to, the “oceanic peridotite trend”of Boyd (1989). The second group is less common and consists of harzburgites, which have depleted compositions and plot in the Proterozoic field on the olivine Mg# vs modalolivine diagram. The third group comprises highly fertile Iherzolites and is restricted to the northwestern localities. Modelling of compatible trace element abundances in clinopyroxene indicates that the fertile Tasmanian peridotites have experienced less than 5% (batch or fractional) partial melting in the spinel stability field while the highly fertile peridotites have experienced less than 2% melting. The depleted xenoliths require higher degrees of partial melting (8-15% fractional melting). Most Tasmanian xenoliths show evidence for cryptic metasomatism. Clinopyroxene trace element patterns show enrichment in the LREE, Sr, U and Th which can be modelled by simple mixing between moderately depleted (~5% melting) peridotite and 3-15% basalt. The depth to the crust-mantle boundary across northern Tasmania has been determined by extrapolating xenolith temperatures to the SE Australian geotherm. The CMB is well-defined at 31-32 km beneath much of northern Tasmania and coincides with Moho depths derived from wide-angle seismic surveys. The CMB is poorly constrained beneath northwestern Tasmania but may be as shallow as 26 km. Estimates of crustal thickness are complicated in the northwest by high heat Bow which would raise the local geotherm and thus raise the estimated depth to the crust-mantle boundary. Rare depleted xenoliths found at some localities in Tasmania may represent relics of ancient lithosphere preserved in the shallow lithosphere. This is supported by an Archean Re-Os sulfide age for a depleted xenolith from northeastern Tasmania. A Re-Osisochron age of 1.45±0.7 Ga for sulfides from a fertile xenolith from the same locality suggests that the region might be underlain in part by refertilised Proterozoic lithospheric mantle. The extensive occurrence of hot, fertile lithosphere beneath northern Tasmania suggests that there has been large-scale removal or refertilisation of the original depleted lithosphere. This is attributed to asthenosphere upwelling and lithosphere thinning during rifting between Australia and Antarctica. Mantle peridotites preserved in orogenic massifs in the Western Gneiss Region (WGR), Norway are thought to represent slices of lithospheric mantle tectonically emplaced into continental crust during subduction. The WGR peridotites have whole-rock compositions that are depleted relative to estimates for primitive mantle. The relatively fertile Almklovdalen garnet peridotites have the least depleted bulk compositions and have been modelled as residues after ~20-35% partial melting at low pressure (2 GPa). The more depleted garnet peridotites from Gurskpy and Otrpy have been modelled as residues after 20-40% melt extraction at high pressure. The extremely depleted dunites from Almklovdalen appear to be residues after high degrees (60%?) of partial melting at high-pressure (7 GPa). The discrepancy in the pressure estimates for the garnet peridotites and dunites at Almklovdalen suggests that these rock types are not related by a simple melt-depletion process. Rare earth patterns for the WGR garnets have high HREE and depleted LREE similar to garnets from high-T sheared peridotite xenoliths in kimberlites. This suggests that the WGR rocks represent refertilisation of a depleted precursor. The close structural relationship between the fertile garnet peridotites and the highly depleted dunites at Almklovdalen suggests that the garnet peridotite bodies represent zones of refertilised dunite. Whole-rock compositions for the Almklovdalen garnet peridotites indicates that the refertilising agent was rich in Fe, Ca, Al and Na but not Ti, similar to pyroxenites within the garnet peridotites. Re-Os Trd model ages for Almklovdalen sulfides define a series of peaks, some of which can be matched with known crustal events. However, peaks in the Archean do not correspond with any known event in the WGR crust and suggest that the peridotites experienced an Archean partial melting event. The preservation of Archean ages in the WGR garnet peridotites supports compositional evidence that some Proterozoic mantle sections elsewhere represent strongly modified Archean lithospheric mantle. In conclusion, this study has shown that crustal age is not a definitive indicator of the composition and age of the underlying mantle due to processes such as lithosphere removal and modification which can spatially or temporally decouple crust and mantle. These processes also may leave residual ancient mantle beneath relatively young crust.