Heat flow and the thermal origin of hot spot swells: The Hawaiian Swell revisited

Abstract

We present 150 new heat flow measurements obtained at eight sites along a 1230‐km‐long profile across the Hawaiian Swell about 700 km ESE of Midway Island. Most of the measurements include in situ thermal conductivity determinations, which helped to reduce the statistical uncertainties (95% confidence) at all sites to <±2.l mW m−2. Surprisingly, there is no systematic variation in heat flow across the axis of the swell. With one exception, the mean heat flow (corrected for sedimentation) at each site is within ±10% of the mean value of 57.7±4.3 (S.D.) mW m−2 for all sites. At the two sites at either end of the profile, clearly located off the swell, the observed heat flow is >59 mW m−2, about 20% higher than the predicted heat flow for 100 Ma seafloor based on simple lithospheric cooling models. Thus even though there is a small increase in heat flow with age along the swell, when compared with the off‐swell heat flux, the anomalous heat flow associated with the Hawaiian Swell is probably of the order of 5–10 mW m−2 and arguably may not exist at all. A previous investigation [Von Herzen et al., 1982] may have overestimated the magnitude of the heat flow anomaly associated with the Hawaiian Swell by comparing heat flux measurements on the swell with values expected for simple lithospheric cooling models. The heat flow anomalies associated with the Bermuda Rise and the Cape Verde Rise may also be smaller than previously estimated, given the uncertainties in the heat flux on the normal seafloor surrounding these swells. The lack of a significant heat flow anomaly associated with the bathymetric expression of the Hawaiian Swell is inconsistent with simple models of lithospheric reheating. Dynamic support, accompanied by modest temperature increases (<100°–200°C) confined to the lower lithosphere, and underlying asthenosphere appear to be required to explain both the height of the swell and its small heat flow anomaly. However, detailed modeling of the origin of midplate swells, and their associated heat flow anomalies, is hampered by our lack of understanding of the thermal evolution of old (>80 Ma) oceanic lithosphere.