Best-Fitting Ellipsoid and Surface Disturbance Adjustment of GNSS-Derived Sea Surface Heights

This paper introduces a novel methodology for modeling mean sea surface topography based on GNSS-on-boat observations—continuous satellite positioning from a floating platform. The method is applied for the first time using real-world data collected in a small lagoon in Western Greece. A best-fitting ellipsoid is initially computed to represent the overall geometry of the observed sea surface. A local surface disturbance is then fitted atop the ellipsoid using a polynomial model, capturing small-scale deviations from the ideal form. The resulting hybrid surface significantly improves accuracy: initial mean discrepancies between the ellipsoid and GNSS measurements (3.1 cm) are reduced to below 0.1 cm after incorporating the local disturbance. This innovative approach offers a powerful tool for detailed sea surface modeling and holds promise for local geoid refinement in coastal or semi-enclosed marine areas.