Reheating of old oceanic lithosphere: Deductions from observations

Deep, wide oceanic basins are the only regions of old seafloor where depth is truly representative of thermal isostasy. When the depths of these basins are corrected for the effect of sediment accumulation, and variation in crustal thickness, the principal non-thermal factors have been eliminated. We collect the most precise and reliable values of heat flow for the same basins, from multi-penetration measurements with in situ thermal conductivity, or deep sea drilling thermal gradients backed up by surface surveys. The 9 data points that result from this selection process have been plotted on a depth versus heat flow graph and compared to published thermal models of lithosphere. When considered without regard to age, all the points fall at greater depths than predicted by the 'plate' models with constant temperature lower boundaries, and remarkably close to boundary-layer cooling with parameters determined from the pre-80 Ma depth and heat flow history of the ocean floor. They are differentiated by their heat flows not much by their depths and the order they plot in along the heat flow axis is random with respect to crustal age. Modeling of discrete reheating events shows that near boundary layer conditions are re-established after about 40 Myr, but corresponding to a younger-than-real age. The data therefore favor discrete reheating events rather than a continuously hot basal boundary, as implicitly assumed by the plate model. Lithospheric reheating appears to start only on ocean floor > 100 Ma. The data alone cannot discriminate between a few discrete reheating events due to convective peel-off at the base of the lithosphere or one or more catastrophic events. However, the distribution of points from the Blake-Bahama basin is more consistent with distal reheating associated with the Bermuda hotspot than local convective peel-off.