Gessner, Klaus1, Smithies, R. Hugh1, Lu, Yongjun1
1 Geological Survey of Western Australia, East Perth, WA 6004, Australia
The Yilgarn Craton has been divided into six tectonic domains, of which the South West Terrane, in the southwest of the craton, remains the least studied and understood. Several geological, geophysical and geochemical observations suggest a fundamental geological difference between the South West Terrane and the Archean terranes to the east. Whereas the boundary between the South West Terrane and the Youanmi Terrane is largely hypothetical, a distinct NNW oriented boundary between the Youanmi Terrane and the Eastern Goldfields Superterrane is recognizable in regional geophysical, geochemical, geochronological and isotopic datasets.
We use whole-rock geochemical analyses of Archean granitic rocks in the South West Terrane as a proxy to test whether crustal melts were sourced from spatially distinct lower crustal source domains. We divide the granitic rocks into four groups according to their source composition (K2O/Na2O; sodic if <1, potassic if >1) and depth of melting (Sr/Y; high = >40, low = <40) and also consider variations in trace element characteristics and in whole-rock initial 206Pb/204Pb ratio. As variation in these proxies for melting-pressure shows no spatial relationship with a gravity anomaly identified within the South West Terrane, we propose that this anomaly does not directly relate to Archean felsic magmatism, but more likely relates to the Proterozoic or Phanerozoic evolution of the West Australian Craton.
The spatial distribution of the granitic rocks defines a NE-trend that is independent of the NNW-trend of the eastern boundary of the South West Terrane, and may even extend into the western part of the Eastern Goldfields Superterrane. This spatial geochemical trend is consistent with the interpretation of large wavelength geophysical data including gravity and magnetic anomaly data, is a prominent feature of Nd model age distributions and also appears to be reflected in the spatial distribution of earthquakes. The results of statistical analyses of the geochemical data relative to NE-trending geophysical domains supports our hypothesis that spatially distinct populations exist. We conclude that these NE-trends relate to basement domains that existed before the younger (post-2.73 Ga) terrane boundaries were imposed.
Klaus Gessner leads the Lithospheric Architecture Branch in the Geological Survey of Western Australia. His research interest include Precambrian geology and geodynamics, structural geology and tectonics, mineral deposits and the application of 3D Geomodelling and numerical simulation.