AUTHOR=Ma Chunyong , Hou Qianqian , Liu Chen , Liu Yalong , Duan Yingying , Zhang Chengfeng , Chen Ge 

TITLE=Exploring Siamese network to estimate sea state bias of synthetic aperture radar altimeter

JOURNAL=Frontiers in Marine Science

VOLUME=Volume 11 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/marine-science/articles/10.3389/fmars.2024.1432770

DOI=10.3389/fmars.2024.1432770

ISSN=2296-7745

ABSTRACT=The sea state bias (SSB) is a crucial error of satellite radar altimetry over the ocean surface. However, for operational nonparametric SSB (NPSSB) models, such as two-dimensional (2D) or threedimensional (3D) NPSSB, the solution process becomes increasingly complex and the construction of their regression functions poses challenges as the dimensionality of relevant variables increases. And most current SSB correction models for altimeters still follow those of traditional nadir radar altimeters. Therefore, for Synthetic Aperture Radar altimeter, this paper proposes a deep learningbased SSB estimation model called SNSSB, which takes multi-dimensional variables related to sea state as input to rapidly obtain SSB values, enabling the exploration of multi-dimensional variable combinations influencing SSB. Experiments are conducted using Sentinel-6 self-crossover data from 2021 to 2023, and the model is evaluated using three main metrics, the variance of the sea surface height (SSH) difference, the explained variance, and the SSH difference variance index (SVDI). The results indicate that, on a global scale, compared to the traditional NPSSB, the multi-dimensional SNSSB not only decreases the variance of the SSH difference by over 11%, but also improves the explained variance, particularly by up to 5-10 cm 2 in mid-and low-latitude regions. In the Kuroshio Extension and Gulf Stream Extension, the regional SNSSB also perform well, reducing the variance of the SSH difference by over 10% compared to the NPSSB. Additionally, SNSSB improves the computational efficiency by approximately 100 times. The favorable results highlight the potential of the multi-dimensional SNSSB in constructing SSB models, particularly the five-dimensional(5D) SNSSB, representing a breakthrough in overcoming the limitations of traditional NPSSB for constructing high-dimensional models. This study provides a novel approach to exploring the multiple influencing factors of SSB.