Analysis of the oxidation states of iron in silicate rocks by fusion decomposition

Abstract

The generation of granitoid magmas by partial melting of crustal rocks during continental thickening events is well documented in many geological provinces throughout the world, including the late Mesozoic Sevier belt of western North America. We present a thermal and petrologic model of anatexis and metamorphism in regions of crustal thickening where the only mantle contribution is the normal conductive supply of heat through the base of the lithosphere (i.e. advection of mass and energy are excluded). We distinguish between formation of migmatites and generation of mobile granitoid magmas and examine the temporal and spatial relationships between these two distinct anatectic processes, between anatexis and regional deformation and between anatexis and metamorphism. A fundamental conclusion is that, if protoliths rich in hydrous minerals are present, regional anatexis is the end-product of classical Barrovian metamorphism in thickened continental crust, even in the absence of a free water-rich fluid phase. Barrovian metamorphic facies series are predicted with thickening ratios (maximum crustal thickness attained/initial crustal thickness) as low as 1.3, but mobile granitoid magmas are not formed if this ratio is less than approximately 1.5. Above these lower bounds, Barrovian metamorphism and anatectic granitoid magmatism occur independently of the magnitude of thickening and of the way in which thickening is accomplished. Both processes are sensitive to a diminished heat supply; lowering either Moho heat flow or crustal radioactive heat production results in blueschist-eclogite metamorphism and inhibits the formation of mobile granitoid magmas. We model anatexis under fluid-absent conditions and show that, with such a constraint, migmatization is always a syn-kinematic process (relative to the crustal thickening event), whereas generation of mobile granitoid magmas is in most cases post-kinematic (relative to crustal thickening) but can be syn-kinematic if thickening takes more than approximately 50 Myr. The typical time intervals for melting are consistent with geological observations; mobile granitoid magmas are predicted by most of our models within approximately 10 Myr of the end of the crustal thickening event. This “incubation period” results primarily from the temperature increase required for the dehydration-melting reactions capable of producing large melt fractions to occur. The energetic requirements of anatexis are relatively minor compared to conductive crustal thermal budgets, as shown by the fact that once the necessary P-T conditions are attained, melting reactions are completed within time intervals on the order of 1 Myr, i.e. 1–2 orders of magnitude smaller than the characteristic time scales of the tectonic processes involved in crustal thickening.