Origin of continental arc andesites: The composition of source rocks is the key

Abstract

The generation of continental arc andesites is generally attributed to subduction of oceanic slabs beneath continental margins, but the origin of crustal components in andesites and the petrogenetic processes of andesites remain widely debated. Common hypotheses include differentiation and crustal contamination of the mantle-derived basaltic magmas, subducted oceanic igneous rock melting and subsequent melt-peridotite interaction, relamination and melting of the subducted sediment, and partial melting of the hydrous mantle wedge peridotite. Because the relatively enriched signature of radiogenic isotope composition is generally comparable between andesites and their emplaced continental crust, the crustal contamination at the crust-mantle transition zone is often envisaged in the hypotheses for andesite petrogenesis. However, these hypotheses are not compatible with quantitative constraints from the mass balance of major elements, trace elements and their pertinent radiogenic isotopes in andesites. On the other hand, source mixing is another possible mechanism for incorporation of crustal components into the mantle source of andesitic magmas. This is realized by reaction of the mantle wedge peridotite with hydrous felsic melts that derived from partial melting of the subducted oceanic crust. Based on compilation and analysis of the geochemical data for basaltic and andesitic volcanics from the Middle and Lower Yangtze Valley, a SARSH (subduction, anataxis, reaction, storage and heating) model is advocated for andesite petrogenesis above oceanic subduction zones. According to this model, the sediment weathered from the overlying continental margin was subducted with the oceanic slab, and it then underwent dehydration and melting at subarc depths. This gives rise to the hydrous felsic melts for crustal metasomatism of the mantle wedge, generating the mantle source that is mafic in lithochemistry and enriched in trace elements and their pertinent radiogenic isotopes. Such a source would be located in the lower part of the mantle wedge and undergo partial melting upon heating for andesitic magmatism in still active continental margins or in post-subduction settings. In this case, the property and amount of slab-derived material (hydrous melts vs. aqueous solutions) dictate whether andesitic or basaltic melts would be produced above oceanic subduction zones. Therefore, this model provides an alternative to the classical views on andesite petrogenesis and thus a resolution to long-standing questions about when and how the crustal components were incorporated into the andesites.