The Global Navigation Satellite System-Acoustic ranging combination technique (GNSS-A) is a recently developed technology to precisely detect seafloor crustal deformation. This method can also estimate km-scale underwater sound speed structure (SSS) as a by-product of monitoring seafloor crustal deformation. This paper evaluates the validity of the spatial gradient and its temporal variation of the SSS estimated by GNSS-A observations off the Kii channel before and after Kuroshio meandering. According to the comparison of the JCOPE2M reanalysis data and the in situ observation data, the GNSS-A estimated SSS has local structures that are not reproduced in this reanalysis but were detected by in situ data. In addition, we investigate the effect of observation time on the stability of SSS estimation. The results suggest that the sufficient time required for stable estimation depends on the spatial coverage of observation data, which depends on the depth of the site. As a result, the time resolution was derived to be about one hour at a site whose depth is 1500 m.

全球导航卫星系统-声音测距组合技术（GNSS-A）是一项最新开发的技术，用于精确检测海底地壳变形。这种方法还可以估算公里尺度的水下声速结构（SSS），作为监测海底地壳变形的副产品。本文评估了由GNSS-A观测到的黑潮蜿蜒前后在Kii通道外观测到的SSS的空间梯度及其时变的有效性。根据JCOPE2M重新分析数据与原位 观测数据，GNSS-A估计的SSS具有局部结构，该局部结构在此重新分析中未复制，但通过 原位数据。此外，我们研究了观测时间对SSS估计稳定性的影响。结果表明，稳定估计所需的足够时间取决于观测数据的空间覆盖范围，这取决于站点的深度。结果，在深度为1500 m的地点，时间分辨率约为1小时。