The topographic and gravitational expression of density anomalies due to melt extraction in the uppermost oceanic mantle

Abstract

Evidence suggests that basaltic extrusion can decrease the density of the depleted source region by up to 2% for 20% partial melting. This hypothesis is tested by calculating the effect of the density anomaly between the depleted and undepleted mantle on the topography, gravity and geoid fields and comparing the anomalies to the observed topography, gravity and geoid anomalies over volcanic features in the oceans. If the source region lies at the base of the lithosphere, we find that the gravity and geoid anomalies are proportionally more sensitive to the density contrast than the topography anomaly. This effect is most pronounced for wavelengths less than 1000 km. Since mid-ocean ridge volcanism, the largest source of volcanism in the oceanic plates, is fairly uniform along the ridge, the depleted material in the oceanic plates, and the corresponding topography and gravity anomalies, should also be somewhat uniform in extent. Any long wavelength anomalies will be obscured in the topography and geoid fields by other anomalies at similar wavelengths such as slab and deep mantle anomalies. However at mid-plate swells and seamounts, if the density contrast is greater than 0.4%, the topography, gravity and geoid anomalies are measurable. To be consistent with the wavelengths represented in the geoid and topography data at Hawaii, the depleted material must be spread into a thin layer by convective flow at mid-plate swells. Even in the thin layer model, the topography and geoid anomalies due to the depleted material would still be as large as 200 m and 2 m, respectively. After the convective flow producing the bulk of the swell is removed, the anomaly due to the depleted material would remain and would be apparent in the geoid and topography fields. At seamounts, if the source region is near the base of the plate, then the presence of depleted material should be distinguishable in the geoid field but be somewhat obscured by flexural effects. Therefore, although a topographic or geoid anomaly due to depletion has not been observed, the density anomaly cannot be discounted because only careful scrutiny and modelling of mid-plate swells and seamounts would distinguish its effect.