Enhancing remote sensing of water temperature using acoustic tomography and 3-D-RPS algorithm

Abstract

Remote sensing of water temperature and flow dynamics is of paramount importance for water environment management. Coastal acoustic tomography (CAT), an innovative method for remote sensing in underwater environments, utilizes the dual-way travel time of multiple sound paths to reconstruct the underwater temperature field. However, the transmission experiments are susceptible to data noise caused by station drift and inaccuracies in travel time extraction. In this study, a three-dimensional Rousseeuw Phase-Space (3-D-RPS) thresholding algorithm is proposed to remove outliers in the travel time data obtained from Huangcai Reservoir using three CAT systems on September 16, 2020. Additionally, this work includes grid partitioning based on sound path propagation, calculation of sound path lengths, and reference transmission time within each grid, construction of a sparse matrix, and inversion of the two-dimensional temperature field and its associated inversion error. The accuracy and applicability of the proposed method are confirmed through a comparison with data obtained from temperature–depth (TD) sensors. The results demonstrate the precision and suitability of this approach. By effectively mitigating the impact of station drift and extraction accuracy issues, this method provides a reliable temperature field estimation.