SDUST2026SM: a global seamount catalog based on vertical gradient of gravity anomaly and bathymetric data

Abstract. This study applies the integrated seamount detection and height estimation method based on vertical gradient of gravity anomaly (VGGA) and bathymetry (ISDHEB) to detect and catalog seamounts across the global ocean between 80° N and 80° S. This analysis uses the SWOT_04 VGGA field released by the Scripps Institution of Oceanography (SIO) and the GEBCO_2025 bathymetry grid. The global ocean was divided into 10° × 10° processing tiles, each surrounded by a 1° overlapping buffer. Candidate features were screened using multiple constraints, including local maxima, closed contours, area and amplitude, circular or elliptical fitting, Jaccard similarity, and seamount exclusion zones. Marker-controlled watershed segmentation was used to separate multi-peak anomalies. Foot-of-slope points were then identified from the slope and second derivative of bathymetric profiles to estimate base depth and seamount height, followed by visual quality control. A total of 68,934 candidate seamounts at least 100 m high were identified. Small, medium, and large seamounts accounted for 67.6 %, 30.9 %, and 1.5 % of the candidates, respectively. After visual inspection, 67,374 targets were retained in the high-confidence seamount set, corresponding to an overall candidate acceptance rate of 97.74 %. The high-confidence catalog contains 55.0 % more seamounts than the SIO 2023 catalog, with most additional targets having heights below 1 km. The cumulative height-frequency relationship follows a clear power law above 0.8 km but progressively deviates at lower heights. The expanded catalog reflects the combined benefits of SWOT-enhanced gravity information, updated bathymetry, and the integrated detection and height-estimation procedure. The resulting catalog provides a basis for studies of seafloor tectonics, shipborne bathymetric survey planning, and marine ecological and resource investigations. The SDUST2026SM dataset is publicly available at https://doi.org/10.5281/zenodo.23277209 (Gao et al., 2026).