REE geochemistry and isotopic data of Archean silicic volcanics and granitoids from the Pilbara Block,Western Australia: implications for the early crustal evolution

Eighteen silicic volcanic rocks of the Warrawoona Group and ten associated plutonic rocks from the Pilbara Block, Western Australia, have been chosen for geochemical and isotopic studies. Silicic volcanics of the UNSB (Upper member of North Star Basalt) are dated at 3.56—3.57

, by both the Rb-Sr and the Sm-Nd methods. The respective 1 (initial isotopic composition) values are 0.7005 ± 5 (Sr) and 0.50810 ± 39 (Nd). This age is consistent with the stratigraphic interpretation that the TalgaTalga Subgroup, in which the North Star Basalt occupies the lowermost position, is overlain by the Duffer Formation, whose age was earlier established at 3.45

by the zircon U-Pb method. The new Rb-Sr data on six silicic lava samples from the Duffer Formation yield an isochron of 3.23 ± 0.28 (2v). Though imprecise, this age agrees with the zircon age within error limits. Rb-Sr ages of 2.3–2.4.

obtained for the ‘Panorama’ rocks and the Wyman Formation do not correspond to their initial eruption ages. Chemical arguments suggest that these ages represent the time of metasomatism associated with the widespread thermal event in this region about 2.3–2.4

ago.Geochemically, most of these analyzed rocks (volcanic and plutonic) are of tonalite-trondhjemitegranodiorite (TTG) composition, a typical feature found in many other Archean terrains. They generally show fractionated REE patterns, except the Panorama Formation rocks. Furthermore, the Wyman Formation rhyolites and the post-tectonic adamellites show significant negative Eu anomalies, suggesting a similar mode of magma generation and a probable genetic link. Theoretical considerations suggest that most of these TTG rocks could have been generated by partial melting of amphibolitic or basaltic sources, followed by fractional crystallization.Although the Archean granitic gneisses often possess mantle-like I_sr values, the trace element data indicate that they could not have been derived by direct melting of upper mantle materials. The immediate tectonic implication is that in any Archean terrain, the formation of Na-rich continental crust of TTG suite must be preceded by the presence of basaltic crust. The occurrence of this basaltic crust is a matter of controversy. Such crust might have been totally destroyed by repeated melting processes, or its remnants are now represented by some of the mafic-ultramafic enclaves within the tonalite-trondhjemite batholiths.