Metallogenesis

5.4.5. PALEOTECTONIC SETTINGS

Layered complexes do not seem to have formed in one particular tectonic setting. Some layered intrusions, such as the Duluth Complex (Minnesota, USA), are clearly related to a continental rift system (Basalt Volcanism Study Project, 1981), but the settings of others, such as the Bushveld and Stillwater, are controversial. For example, Irvine and Sharp (1982) suggested that the U„ and A„ melts involved in the formation of the Bushveld Complex came, respectively, from deep in the upper mantle (perhaps from a depth of 180-200 km) and from the bottom of thick (40-50 km) segments of continental crust. Hatton (1989), on the other hand, favored a supra-subduction setting for the Complex. In his model, the U_a magma was derived from a depleted-mande source that had been metasomatized by fluids from subducted sediments and the A„ magma from subducted oceanic lithosphere in a region of retro-arc spreading when the Kaapvaal and Zimbabwe cratons collided around 2700 Ma. Stowe (1989) proposed that the Complex was intruded in a region of crustal extension, and von Gruenewaldt and Harmer (1992) discussed evidence favoring the emplacement of the Complex into an intracratonic environment in which riftiing had occurred during deposition of the sediments of the Transvaal Sequence. An extensional setting has also been proposed

for the Great Dyke (Bichan 1969, Prendergast 1987) and the Skaergaard (Brooks & Nielsen 1982). The intrusion of the Stillwater Complex is believed to be related to strike-slip faulting in a continental environment (Page & Zientek 1985). A continental environment for the Stillwater magmas is supported by Rh-Os and Sm-Nd systematics. Lambert et al. (1989) determined that U-type magmas for the Stillwater Complex had initial e_Nd of -0.8 to -3.2 and a chondritic initial ¹⁸⁷0s/¹⁸⁶0s ratio of «0.88, whereas A- type magmas had e_Nd of -0.7 to +1.7 and initial ¹⁸⁷Os/^l86Os ratio of »1.13. These data suggest that U-type magmas were derived from a lithospheric mantle source containing recycled crustal materials, whereas A-type magmas originated either by crustal contamination of basaltic magmas or by partial melting of basalt in the lower crust. Despite some lingering uncertainties, it is reasonable to conclude that layered complexes were emplaced into already stabilized continental crust. This conclusion is consistent with the pervasive cumulate layering, including stratiform chromitites, in the complexes.

Because of the remarkable similarity with oceanic crust formed at modem oceanic spreading centers, ophiolites are believed to have formed originally in a similar setting (Coleman 1977). However, there is a growing body of geochemical evidence in favor of a subduction-related model for many ophiolites, probably a back-arc basin rift environment that can account for both the sheeted diabase dike complex, requiring sea- floor spreading, and the geochemistry of the pillow basalts, requiring magma generation above a subduction zone (Pearce 1980, McCulloch & Cameron 1983, Coleman 1984). As an extension of the melt-rock interaction model for the origin of podiform chromite deposits (Zhou et al. 1994), Zhou and Robinson (1997) suggested that the occurrence of podiform chromitites and their chemical compositions could be correlated with formation in two environments: in island arcs, and in nascent spreading centers, such as back-arc basins. The island-arc environment, characterized by high degrees of partial melting because of high input of volatiles from the subducting slab into the overlying mantle wedge, is conducive to the formation of high-Cr chromite (e.g., Troodos ophiolite). Back-arc basins, on the other hand, should favor the crystallization of high- A1 chromite because of smaller degrees of partial melting due to a lower influx of volatiles in this environment. Mature spreading centers, such as mid-ocean ridges, do not provide a favorable environment for the formation of podiform chromite deposits. The composition of mid-ocean ridge basaltic magmas remain remarkably uniform through time, providing little drive for melt-rock interaction once the magmas have equilibrated with the wallrocks through which they pass.

Stowe (1994) proposed that the chemical composition of chromite ores might be used as an indicator of their tectonic setting. It is well established that the principal control on chromite chemistry is the composition of the melt from which it crystallizes (Dick & Bullen 1984, Roeder & Reynolds 1991) and that the melt (magma) composition varies from one tectonic setting to another because of differences in the parent material and conditions of partial melting. There is a good correlation between liquidus chromite composition — e.g., Cr:(Cr+Al) ratio versus Mg:(Mg+Fe²⁺) ratio — and magma type, but the extrapolated correlation between magma type and tectonic setting can be equivocal (Rollinson 1995). Moreover, the composition of chromites subjected to subsolidus re-equilibration during serpentinization and regional metamorphism may be significantly different from the original liquidus compositions. An evidence for subsolidus re-equilibration is the development of Fe- and Ti-enriched “ferrit-chromite” rims on serpentine-hosted chromite grains.

Date: 2015-02-16; view: 973