Abstract
It is important to appropriately model underwater sound speed structures to detect seafloor displacements accurately using GNSS-acoustic seafloor geodetic observations. In recent years, various sea surface platforms (e.g., wave gliders) have been developed for GNSS-acoustic observations. Sub-mesoscale oceanic phenomena can be detected by simultaneously employing multiple sea surface platforms. However, the use of a single sea surface platform with slow navigation speeds may degrade the modeling accuracy of underwater sound speed structures, even when compared to conventional ship-based observations. Therefore, the development of a GNSS-acoustic positioning technique that expresses a complex underwater sound speed structure and simultaneously provides constraints on sound speed parameters, if necessary. This study arranges the observation equation by considering multiple-layered sound speed gradients and develops a GNSS-acoustic positioning scheme using a Bayesian framework. The performance of the proposed GNSS-acoustic positioning method was investigated using synthetic datasets. The proposed method successfully modeled a complex underwater sound speed structure (e.g., temporal variations in sound speed gradients) using a dataset collected by dual sea surface platforms, which is highly sensitive to the underwater sound speed structure. It also provides robust solutions, even for a dataset with low sensitivity, by appropriately introducing constraints on the sound speed parameters. Moreover, the proposed method was applicable to an actual observational dataset, and it was confirmed that the GNSS-acoustic positioning method under special conditions (assumption of a temporally constant single-layered sound speed gradient) in a previous study can be reproduced by the constraints in the proposed method. Thus, the proposed method enabled us to flexibly model the underwater sound speed structure and accurately detect seafloor displacements for various types of observation datasets. The proposed method was implemented in the open-source GNSS-acoustic positioning software "SeaGap.”