Crust-mantle interaction during syn-collisional magmatism – evidence from the Oamikaub diorite and Neikhoes metagabbro (Damara orogen, Namibia)

Abstract The early-syntectonic 563.7±6.1 Ma old Oamikaub diorite (Damara orogen, Namibia) consists of metaluminous, magnesian, calc-alkalic to calcic diorites, granodiorites and granites. Associated gabbro-diorites and gabbros belong to the Neikhoes metagabbro. Linear major and trace element variations imply that the rock suite evolved through fractional crystallization processes involving amphibole, biotite, Fe-Ti oxides, zircon and apatite. Initial Sr (87Sr/86Sr: 0.7058-0.7123) and Nd (e Nd: -2.1 to -18.8) isotopic compositions are highly variable and negatively correlated indicating that assimilation of crustal components occurred. Unradiogenic initial 206Pb/204Pb (16.23-17.23) and 207Pb/204Pb ratios (15.50-15.57) suggest derivation from or interaction with ancient crust with low U/Pb. Two gabbro-diorites have MgO, Ni and Cr abundances that are compatible with derivation of these rocks from upper mantle lithologies. Their initial e Nd values (-2.1 and – 7.4) and 87Sr/86Sr ratios (0.7058 and 0.7076) imply derivation from an aged metasomatized lithospheric mantle. Other mafic samples have MgO abundances and compatible element concentrations that exceed the values commonly accepted for primary mafic melts implying some accumulation of clinopyroxene and amphibole. The granodiorites form a homogenous group in which the isotope data (initial e Nd: -12.4 to -14.1; initial 87Sr/86Sr: 0.7083-0.7096) imply a lower crustal source. The granites are also magnesian and calc-alkaline but two of them are strongly peraluminous. Their isotope data (initial e Nd: -13.2 to -18.8; initial 87Sr/86Sr: 0.7099 to 0.7123) imply derivation from more ancient sources, alternatively these samples gained their isotope systematics through extensive AFC processes from parental granodiorites. A common subduction zone environment as suggested from negative Nb-Ta anomalies in multi-element diagrams seems unlikely for all samples because of a lack of isotopically depleted signatures. The data from the Oamikaub diorite and other mafic complexes are better explained by a “flat” subduction model involving mainly continental mantle lithosphere and crust with limited, if any, melting of asthenospheric mantle.