Fabric development as the key for forming ductile shear zones and enabling plate tectonics

Lithospheric deformation on Earth is localized under both brittle and ductile deformation conditions. As high-temperature ductile rheologies are fundamentally strain-rate hardening, the formation of localized ductile shear zones must involve a structural or rheological change or a change in deformation conditions such as an increase in temperature. In this contribution, I develop a localization potential that quantifies the weakening associated with these changes. The localization potential corresponds to the increase in strain rate resulting from that change under constant stress conditions. I provide analytical expressions for the localization potential associated with a temperature increase, grain size reduction, an increase in water fugacity, melt content, or the abundance of a weak mineral phase. I show that these processes cannot localize deformation from a mantle convection scale (10³ km) to a ductile shear zone scale (1 km). To achieve this, is it necessary to invoke a structural transition whereby the weak phase in a rock forms interconnected layers. This process is efficient only if one phase is much weaker than the others or if the weakest phase has a highly non-linear rheology. Micas, melt, and fine-grained aggregates – unless dry rheologies are used – have the necessary characteristics. As none of these phases is expected to be present in the dry lithosphere of Venus, this concept can explain why Venus, unlike the Earth, does not display a global network of plate boundaries. The diffuse plate boundary in the Central Indian Ocean may be as yet non-localized because serpentinization has not reached the ductile levels of the lithosphere.