[S-GD01] Geodetic Advances in Crustal Dynamics and Environmental Change for Taiwan and Neighboring Regions

[SGD01-12] Evaluation of Post-Processing GNSS Solutions for Depth Modernization

*I Yun Chou¹, Chung-Yen Kuo¹, Chi-Ming Lee¹, Jian-Sing Lyu², Yen-Ti Chen², Tzu-Pang Tseng¹ (1.Department of Geomatics, National Cheng Kung University, Taiwan, 2.Department of Land Administration, Ministry of the Interior, Taiwan)

Keywords:Depth Modernization, IHO Special Order, Precise Point Positioning, Post-Processed Kinematic, GNSS/IMU integration

Depth modernization utilizes a GNSS/IMU system and a multibeam echo sounder to directly measure the ellipsoidal heights of the seafloor, which is subsequently transformed into charting datums, such as the Lowest Astronomical Tide (LAT), using a Separation Model (SEP). Therefore, achieving high accuracy of GNSS vertical positioning is crucial and must comply with the standards set by the International Hydrographic Organization (IHO). In this study, data was collected using an unmanned surface vessel (USV) equipped with a GNSS/IMU system and a multibeam echo sounder at a nearshore port environment, Argo Marina in Kaohsiung. Since attitude corrections for the USV are required, GNSS and IMU integration is essential to minimize uncertainty. This study assesses vertical accuracy against half of the IHO Special Order vertical accuracy standard (12.5 cm) by applying various post-processing techniques, including Precise Point Positioning (PPP), Post-Processed Kinematic (PPK), GNSS/IMU integration (both tightly and loosely coupled), and online PPP modules. Additionally, we also examine the vertical accuracy of tightly coupled PPK in signal obstructed areas and propose semi-automated corrections for unreasonable offsets in GNSS solutions. Preliminary results indicate that vertical positioning differences among the various methods are ±3 cm, with a standard deviation of 5 cm, likely resulting from algorithmic variations. Nonetheless, all results satisfy the IHO Special Order accuracy requirements. The findings further demonstrate the improvements in the 3D point cloud of bathymetric data following the correction of unreasonable offsets.