Abstract

It is generally accepted that the main strength of subduction boundaries occurs in the shallower region where frictional deformation is dominant. However, estimates of absolute values—commonly expressed as apparent coefficient of friction, μ′—show great variation. Frictional shear heating is closely related to μ′, and estimates of the extra thermal energy supplied by shearing can in principle be used to estimate subduction zone strength. One such approach is based on surface heat flow measurements. However, heat flow in convergent margins shows large local scatter and even in the same area, different studies using this method show large variations in estimates of μ′ indicating large uncertainties. The thermal record of subduction conditions preserved in subduction-type metamorphic rocks is developed over geological time scales that average out local complexities in heat flow and therefore has good potential as an alternative indicator of the amount of shear heating and, hence, shear strength along subduction boundaries. Thermal models that incorporate shear heating were developed for two contrasting and well-known subduction-type metamorphic belts: the relatively warm Sanbagawa belt of SW Japan and the relatively cold Franciscan belt of western USA. High-grade rocks of the Sanbagawa belt show strongly curved P–T paths that display increasing P/T to about 2 GPa. Information on the rate of plate movement and the age of the subducting slab at the time of metamorphism can be combined with modelling results to show that relatively high shear stresses, equivalent to μ′∼0.13 are required to account for the observed curved P–T paths. In contrast, the high-grade rocks of the Franciscan belt show relatively cool P–T conditions that do not allow for strong shear heating with an appropriate upper bound for μ′ of ∼0.03. Modelling suggests subduction rate and lithology are potentially important controls on the development of high- versus low-stress subduction zones. High-stress subduction zones are likely to be associated with high aseismic/seismic slip ratios possibly related to slab roughness.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call