ACCRETION, UNDERPLATING, DETACHMENT AND EXHUMATION ALONG THE SUBDUCTION PLATE INTERFACE: IMPLICATIONS FOR SHORT-TERM AND LONG-TERM RHEOLOGY OF THE PLATE INTERFACE, FROM NASCENT SUBDUCTION TO CLOSURE

The presence of km-scale accreted terranes/units in both ancient and present-day subduction zones points to changes in strain localisation along the plate interface. The rock record indicates that, for a given subduction zone, exhumation is episodic: no more than a few My compared to the ~100 My lifetime of typical subduction zones.

Not much is known, however, regarding this process: what exactly is episodic (i.e., detachment from the slab and/or exhumation?), how long and where? Arguably too, changes in mechanical coupling may impact, or be impacted, by the initial structure of the incoming plates (structural/lithological heterogeneities, thermo-fluid regime, geodynamic boundary conditions,...).

We herein present new and literature data from depths between 15-20 and 120 km along the subduction interface, which span a range from long-term to short-lived events of underplating and/or exhumation, and confront them with the recent wealth of geophysical/chemical data.

Data indicate that the slicing of km-scale units in fact mostly occurs at specific depths where major coupling changes occur along the plate interface: at 30-40 km (downdip of the seismogenic zone) and 70-80 km (where mechanical coupling between the two plates resumes and where eclogites get critically dense). This suggests that the detachment of material from the slab is a (steady or incidental?) depth-dependent process controlled by interplate coupling, whereas later exhumation is rather controlled by large-scale, lithospheric-scale boundary conditions.

We then discuss the possible link between tectonic slicing and seismicity, and the recent finding of eclogite breccias. In the long-term, we propose that mechanical coupling may vary through time, from weak to strong, as a function of the contrast of effective viscosity on either side of the interface: a young and wet subduction interface will promote the formation of knockers and sole accretion, whereas a fluid-present yet drier and colder one will lead to mainly metasedimentary underplated material and large-scale slivers of oceanic lithosphere.

This interpretation is supported by bi-phase numerical models allowing for fluid migration and by the transient accretion of metamorphic soles during nascent intra-oceanic subduction, when similar effective rheology exists on both sides.