Plate Motion Driving Forces Revisited

The motions of the surface plates are thought to be primarily attributed to the driving forces from the descending slabs in terms of slab pull and slab suction forces. Dense subducting slabs mechanically attached to the surface plates can act as stress guides to transmit the vertical gravity forces into horizontal edge pull forces to drive oceanic plate motions. Dense subducting slabs also induce mantle flow, exerting basal tractions that drag the nearby plates towards the subduction zones. The explicit numerical modelling of stress guides (i.e., strong slabs attached to stiff oceanic lithosphere along with neighboring shallow weak plate boundaries) in finite-element flow models is a major computational challenge. Consequently, previous studies mostly added slab pull forces externally through a slab-pull torques balance approach. However, we found that in the mantle flow models that include both negatively buoyant and strong slabs, the strong upper mantle slabs might inhibit the slab suction forces and their induced basal drag beneath plates. This result is not physically consistent with the previous approach of external linear addition of slab pull forces. There are further compelling reasons to question the efficacy or strength of slab pull forces as significant drivers of plate motions. Firstly, the far-field upper mantle slab suction forces are an important driver of continental plates with only a small fraction of marginal subduction zones (e.g., North American and African plates) (Liu and King, in revision). Secondly, the remarkable lateral fixity of the spreading ridge along Eastern Pacific Rise (EPR) over the past 80 Ma cannot be simply explained by constantly changing upper mantle slab pull forces acting on the Pacific and Farallon/Nazca plates (Rowley et al., 2016). Here we present results of physically consistent, high-resolution mantle flow models that quantify the contributions of slab suction and slab pull on global plate motions extending into the geological past. Our results indicate that lithospheric basal drag generated by slab suction forces and by deep positive buoyancy from hot upwellings (in particular below the EPR) is a more important driver of global plate motions than slab pull edge forces.