Seamount gravity anomaly modelling with variably thick sediment cover

Ramillien, G. and Wright, I.C. (2002) Seamount gravity anomaly modelling with variably thick sediment cover. Marine Geophysical Researches, 23 (1), 13-23. (doi:10.1023/A:1021231221660).

Abstract

Inversion modelling of marine gravity anomalies to derive predicted seafloor topography has provided significant advance in delineating deep-ocean bathymetry where the seafloor both conforms to the half-space cooling model of seafloor spreading, and largely sediment-free. Similar modelling for elevated ridges and seamounts, that are formed by processes other than seafloor spreading and/or have proximal sediment sources (e.g., continental margins and volcanic arcs), have significantly higher errors when validated against modern shipborne echo-sounding data. A three-dimensional, five-layer gravity model is emulated for the cases of both synthetic and real seamounts, with varying degrees of sediment burial, to establish the sensitivity of variable sediment cover as a source of error. A simple 'Gaussian' seamount with base radius of 30 km, 2000 m of relief, has a maximum 140–160 mGal anomaly, that decreases to 50 mGal with the addition of 1 km of sediment cover with simple 'flood' geometry. Complete burial, with a typical sediment density of 2300 kg m^–3, results in a 120 mGal difference from a sediment-free seamount model. Increasing sediment density results in an exponential decay of the seamount anomaly. More complex synthetic geometries of varying basement relief and sediment thickness show that the anomaly amplitude remains significant, especially where the latter is >700–800 m thick. For the real case, seamounts of the Three Kings Ridge (northern New Zealand) imaged with seismic reflection data, with varying degrees of sediment cover of up to 1 km, when modelled both with and with-out the inclusion of a sediment layer, typically have rms differences of 30 mGal between observed and modelled gravity anomalies. Significantly, the rms errors are reduced by 50% with the inclusion of a sediment layer that corresponds to a reduction of predicted seafloor topography rms errors of 192–684 m to 78–360 m.

This record has no associated files available for download.

Contributors

Download statistics