This paper presents a study of the evolution of the oceanic lithosphere from a thermomechanical approach. We have investigated the finite-amplitude development of secondary convection cells beneath the oceanic plates by means of the single-mode mean-field equations and the fully 2-dimensional convection equations, using finite-element techniques. Both Newtonian and non-Newtonian rheologies with highly temperature- and pressure dependences, and an activation energy of 100 kcal mol-1 have been employed. The temperature at the base of the convecting medium governs the final thickness of the lithosphere. The mean interior temperature varies only slightly during the temporal evolution. Non-Newtonian rheology has the tendency to induce oscillatory time-dependent behavior of the flow. Heat flow, topography and gravity are influenced by secondary convection in two ways. Small scale perturbations with wavelengths of around 600 km arise from the lateral thermal differences between the uprising and descending convective limbs; large-scale features are also produced as a consequence of lithospheric growth. The calculated quantities of the heat-flow topography and gravity associated with small-scale convection are typically in the range of O(HFU), O(102m), and O(10 mgal). The horizontal mean-temperature profiles from the convection model are used to calculate long wavelength geophysical observables as a function of age. Convective processes are found to reduce the rate of lithospheric thickening. Predictions from our model can fit well the observed data of heat flow, ocean floor topography and geoid offsets along fracture zones, the last data base exhibiting the most sensitivity to thermal perturbations below the lithosphere. Our calculations show that the oceanic lithosphere is able to grow continuously up to O(109 y), long past the flattening of the seafloor. We report here that the thickness of a thermally equilibrated lithosphere could reach ∼ 250 km, which lies within the appropriate range of values for the continental lithosphere, as inferred from studies in seismology, flexure observations, and secular polar motions.
Bibliographical noteFunding Information:
We thank Dr. C.A. Anderson at the Los Alamos National Laboratory for supplying us with his program SANGRE and Dr. G. Sewell at the Urn-versity of Texas, El Paso, for numerical advice concerning elliptic—parabolic systems. We are indebted to stimulating discussions with our friends Uli Christensen and Henri-Claude Nataf. This research has been supported by “Institut National d’Astronomie et de Geophysique” (INAG) in the framework of the A.T.P. (Sismogenese). (Grant no. 1150), N.S.F. grants EAR-8117439 and EAR-8214094, Petroleum Research grant 13550-G2, administered by the American Chemical Society, and by a N.A.T.O. research award (grant no. 27681).