Eduction, extension, and exhumation of ultrahigh‐pressure rocks in metamorphic core complexes due to subduction initiation

Abstract

AbstractThe controversy over the exhumation of ultrahigh‐pressure (UHP) rocks centers on whether it involves rising of pieces of crust detached from subducted continental lithosphere or an entire subducted plate that undergoes “eduction,” i.e., reverse subduction. We present a new thermomechanical model of continental subduction showing that these apparently contrasting mechanisms can occur together: crust subducted deep enough is heated and weakened, causing limited diapiric rise, while crust subducted to shallower depths retains strength and is exhumed only by eduction. The model also shows for the first time how eduction followed by seafloor spreading can occur in a zone of regional convergence. This occurs spontaneously when subduction of buoyant crust causes a subduction zone to “lock up” in one place causing a new subduction zone to form in another. The model is consistent with many features of the youngest region of UHP rock exhumation on earth: the D'Entrecasteaux Islands. UHP exhumation and the amount of regional extension, as well as the seismic structure around the islands, can be explained by eduction. Ductile flow fabrics, seen on the islands, would result from exhumation of the most deeply subducted crust heated enough to undergo partial melting. Reversal of motion on the north‐dipping continental subduction zone, required by this model, was likely triggered by initiation of the New Britain Trench, as suggested previously. Our model implies that the crust of Goodenough Basin, south of the islands, was exhumed by eduction in the last 5 Ma and this hypothesis can be tested by drilling.