An Integrated Calibration Method for Sounding Velocity Errors in ROV-Borne Multibeam Bathymetric Data

Accurate sound speed information is critical for bathymetry collected by multibeam echosounder systems (MBESs). However, achieving this remains challenging for high resolution underwater topographic surveys performed by the remotely operated vehicle (ROV), leading to significant sounding velocity errors. To address this, we propose an integrated calibration method that sequentially corrects positioning inaccuracies, removes outlier beams, and performs sound speed inversion. First, ROV trajectory is smoothed to mitigate positioning errors under Global Navigation Satellite System (GNSS)-denied conditions. Second, edge-beam noise is filtered based on beam travel time characteristics prior to georeferencing. Finally, equivalent sound speed profiles are inverted by minimizing depth discrepancies in overlapping areas between adjacent swath. The method was validated using deep-sea data collected by the ROV “Discovery”. In the validation dataset, for the overlap region between adjacent survey lines with the largest pre-correction depth errors, the mean and standard deviation (SD) of depth deviations decreased from 2.020 m and 1.805 m to−0.066 m and 0.378 m, respectively, after correction. Overall, the mean depth deviation was reduced to the centimeter level, while the SD decreased by approximately 50%. These results demonstrate that the proposed approach significantly enhances the consistency and accuracy of ROV-based MBES data, making it compliant with the Special Order standards of the International Hydrographic Organization (IHO).