Comparative Study on Topography Prediction Using Gravity Anomalies and Gravity Gradient Anomalies

Direct bathymetry is a primary data source for creating high-precision seafloor topography but cannot easily achieve high-resolution global coverage. Therefore, using gravity data for seafloor topography prediction is an alternative method. Current algorithms, reliant on the measurement of ship soundings and geophysical parameters, still have room for improvement in accuracy and coverage. In this study, we establish observation equations based on vertical gravity anomalies (VG, also called gravity anomalies) and vertical gravity gradient (VGG) anomalies generated by a rectangular prism and develop a new analytical algorithm for predicting topography. We validate its effectiveness through numerical simulations and actual applications. We determine the source of the errors, the mid–high frequency error caused by the boundary area is reduced by regularization, whereas the low-frequency error caused by the far area is reduced by the error equation algorithm. In the shallow areas with a maximum depth of 2 kilometers and deep-sea areas with a maximum depth of 5 kilometers, the root mean square errors (RMS) for VGG anomalies prediction are 93.8 m and 233.8 m, respectively, while for VG anomalies prediction they are 101.2 m and 239.4 m. We also find that VGG anomalies are more sensitive to topography fluctuations, but VG anomalies have a stronger linear correlation with topography. Additionally, we propose a cubic spline interpolation algorithm to effectively fuse the prediction results and ship soundings, improving topographic prediction accuracy in shallow and deep-sea areas by up to 53.2% and 39.67%, respectively.