Bathymetry Prediction with SWOT Gravity Anomaly using Machine Learning Methods: Paper 2 - Model Evaluation and Uncertainty Analysis

This is the predicted bathymetry based on an ensemble of machine learning models utilizing the gravity field derived from the Surface Water and Ocean Topography (SWOT) satellite. The methodology and evaluation is determined from a pair of companion papers:

Sandwell D, Phrampus B, Salajegheh F, Nilsson B, Liu B, Yu Y, Harper H, Andersen O, Smith W, Elmore P, Kirby J, Beale J, Roberts J, Perez L. Bathymetry Prediction with SWOT Gravity Anomaly using Machine Learning Methods: Paper 1 - Model Development

Nilsson B, Sandwell D, Phrampus B, Salajegheh F, Liu B, Yu Y, Andersen O, Smith W, Elmore P, Kirby J, Perez L Bathymetry Prediction with SWOT Gravity Anomaly using Machine Learning Methods: Paper 2 - Model Comparison and Uncertainty Analysis

Along with the predictions, a dataset containing the confidence levels of bathymetry is provided. This is computed as a function of features of interest from the gravity field as well as the distance between ship soundings.