A simple mechanical model for oceanic spreading centers

Abstract

In a simple model for a spreading oceanic ridge, it is assumed that viscous material rises as a result of gravitational instability in a vertical-walled cleft between steadily diverging lithosphere plates of uniform thickness. The requirement that the walls must accrete at the same rate as they spread permits a simple discussion of the energetics of the conduit. A numerical example suggests that the rate of mechanical energy dissipation in the conduit may be comparable on a global scale to the total rate of seismic energy release. This much energy is available to drive or resist plate separation at oceanic ridges, depending on the vertical distribution of pressure in the conduit. If the mean pressure at the conduit wall were greater than the mean lithostatic pressure in the lithosphere, tectonic compression would develop in the spreading direction, dissipation in the conduit would decrease, and the energy saved would go into driving plate separation. If the mean pressure on the conduit wall were less than mean lithostatic, tectonic tension would develop. It would resist plate separation and would increase energy dissipation in the conduit accordingly. If the effective viscosity decreased upward in the conduit from values characteristic of the asthenosphere to values characteristic of crustal intrusions, the mean conduit pressure would be low, and ridges would offer substantial resistance to plate separation (∼300 bars in the numerical example). The model provides a relation between mean viscosity, conduit dimensions, and spreading velocity. For lower viscosities, much narrower conduits are required. An extension of the model to describe viscous drag on the conduit wall accounts for uplift adjacent to spreading centers and the occurrence of axial valleys for slow spreading and axial horsts or volcanic piles for rapid spreading. The model can evidently account for axial relief and its change with velocity in general agreement with observations. Reasonable conduit widths result if the relaxation time of the nearby lithosphere is assumed to be on the order of a million years. The stair-step offsets of oceanic ridges by transform faults suggest that the ridges offer much greater resistance to plate separation than the transforms. This is consistent with the present model, if transforms are loosely coupled with a confined fluid phase at depth and if their resistance is primarily in a superficial seismogenic zone. It is speculated that a similar decoupling mechanism might operate along the base of the lithosphere, with important implications for plate dynamics.