Maximum Decoupling Depths in Subduction Zones From the Rock Record

The maximum depth of decoupling or MDD is the depth at which subducting oceanic plates—or slabs—become fully viscously coupled with the overlying mantle wedge and has a strong influence on the thermal structure of subduction zones. In many models, this depth is assumed to be around 80 km, based on comparisons between model results and measured surface heat flow data. However, very few convergent margins have a dense enough network of heat flow measurements to provide reliable constraints on this depth. As a result, the range of possible MDD for different regions and different times remains poorly constrained and the suitability of using a fixed value for Dc in thermal models of subduction in unclear.

We propose an alternative method for estimating MDD based on the rock record of subduction-type metamorphic belts. As rocks move along the subduction interface and pass through this depth, they transition from a cold domain—where the thermal structure is dominated by the advection of cool lithosphere—to a much hotter domain, where induced inflow of hot mantle towards the subduction interface leads to significant warming. This transition should result in a sharp increase in temperature over a relatively small increase in depth. If this thermal bend can be recognized in subduction-type metamorphic belts, its depth can serve as a valuable MDD indicator in ancient subduction zones. An important caveat to our proposed approach is that high thermal gradients can also result from shear heating at shallower depths, and these must be distinguished to make a reliable estimate.

We have identified several examples of thermal bends from ancient subduction zone settings. These all suggest that MDD occurs at depths 70–90 km. Our results support the idea that MDD varies little between different subduction zones or over geological time.