Deep-focus earthquakes: From high-temperature experiments to cold slabs

Abstract Deep-focus earthquakes (DFEs) present an interesting scientific challenge in that they occur at depths where brittle failure should be impossible. The fact that their occurrence is confined to locations where subducting lithospheric slabs are crossing through the transition zone suggests that olivine phase transformations may be involved in the production of these earthquakes. Experimental studies have shown that olivine can persist metastably in subducting slabs and that olivine phase transformations can lead to faulting at high pressures. However, it has been argued that large DFEs are too large to be contained within a metastable olivine wedge preserved in the interior of subducting slabs. We demonstrate, using experiments on olivine-analog materials, that transformational faulting can continue to propagate via shear-enhanced melting into the stable high-pressure phase. We also show that transformational faulting is controlled by the ratio between strain rates and the olivine-ringwoodite transformation rates, and extrapolate this relationship to the natural conditions of DFEs. Counterintuitively, these results imply that cold and fast-subducting slabs produce transformational faulting at higher temperatures, which results in more numerous DFEs.