17:15 〜 19:15
[SCG55-P27] オープンソースコードGARPOSを用いた南西諸島海溝における海底地殻変動解析
The Nansei-Shoto Trench is a plate boundary where the Philippine Sea Plate subducts beneath the Eurasian Plate. Historical records show that the 1771 Yaeyama Tsunami (Nakamura, 2009) was triggered by a subduction earthquake, indicating the significant tsunami risk in this region. Understanding the dynamics of this trench is crucial for assessing seismic hazards and mitigating potential disasters. In this study, we conducted seafloor crustal deformation measurements at three observation points (RKB, RKC, and RKD) in the Nansei-Shoto Trench using the GNSS-Acoustic (GNSS-A) positioning method.
The GNSS-A positioning method determines the position of a seafloor station by combining GNSS measurements for ship positioning with acoustic ranging measurements between the ship and the seafloor instruments.
For seafloor station position analysis, we utilized the GARPOS analysis code (Watanabe et al., 2020). This code estimates a temporally and spatially varying sound speed field, addressing one of the primary error sources in the GNSS-A method.
We then estimated the displacement velocity of each observation point relative to Okinawa Island and Miyako Island (defined as the Okinawa Block) using the least squares method, based on the seafloor station positions analyzed in the ITRF2014 reference frame and the F5 solutions of nearby GEONET observation points provided by the Geospatial Information Authority of Japan.
As a results, relative to the Okinawa Block, the estimated displacement velocity was 0.9 ± 4.2 mm/yr eastward and 1.3 ± 2.0 mm/yr northward at RKB, 8.1 ± 20.2 mm/yr westward and 5.2 ± 14.0 mm/yr southward at RKC, 2.0 ± 10.1 mm/yr eastward and 12.3 ± 29.8 mm/yr southward at RKD.
Although the uncertainties for RKC and RKD are relatively large, no significant movement was observed at any of the observation points. These results suggest the absence of a coupled zone immediately beneath the observation sites. However, further improvements in accuracy are necessary for more detailed discussions.
The sound speed gradients estimated through GARPOS indicate no systematic directional bias. Additionally, the sound speed gradient in the shallow region was larger and more scattered compared to the deeper region, which aligns with the general understanding that sound speed is more stable at greater depths and that there are no significant ocean currents in the southeastern Okinawa region. This supports the reliability of the sound speed structure estimated by GARPOS.
Furthermore, some survey cruises exhibited linear variations in sound speed gradient at the beginning or end of observations. This trend may result from segments were excluded, the linear variation disappeared, suggesting that the sound speed gradient derived from GARPOS can serve as a useful tool for data quality assessment.
Reference
Nakamura, M., (2009) Fault model of the 1771 Yaeyama earthquake along the Ryukyu Trench estimated from the devastating tunami. Geophysical Research Letters, 36, L19307
Watanabe, S., Ishikawa, T., Yokota, Y., & Nakamura, Y. (2020). GARPOS: Analysis software for the GNSS-A seafloor positioning with simultaneous estimation of sound speed structure. Frontiers in Earth Science, 8, 597532.
The GNSS-A positioning method determines the position of a seafloor station by combining GNSS measurements for ship positioning with acoustic ranging measurements between the ship and the seafloor instruments.
For seafloor station position analysis, we utilized the GARPOS analysis code (Watanabe et al., 2020). This code estimates a temporally and spatially varying sound speed field, addressing one of the primary error sources in the GNSS-A method.
We then estimated the displacement velocity of each observation point relative to Okinawa Island and Miyako Island (defined as the Okinawa Block) using the least squares method, based on the seafloor station positions analyzed in the ITRF2014 reference frame and the F5 solutions of nearby GEONET observation points provided by the Geospatial Information Authority of Japan.
As a results, relative to the Okinawa Block, the estimated displacement velocity was 0.9 ± 4.2 mm/yr eastward and 1.3 ± 2.0 mm/yr northward at RKB, 8.1 ± 20.2 mm/yr westward and 5.2 ± 14.0 mm/yr southward at RKC, 2.0 ± 10.1 mm/yr eastward and 12.3 ± 29.8 mm/yr southward at RKD.
Although the uncertainties for RKC and RKD are relatively large, no significant movement was observed at any of the observation points. These results suggest the absence of a coupled zone immediately beneath the observation sites. However, further improvements in accuracy are necessary for more detailed discussions.
The sound speed gradients estimated through GARPOS indicate no systematic directional bias. Additionally, the sound speed gradient in the shallow region was larger and more scattered compared to the deeper region, which aligns with the general understanding that sound speed is more stable at greater depths and that there are no significant ocean currents in the southeastern Okinawa region. This supports the reliability of the sound speed structure estimated by GARPOS.
Furthermore, some survey cruises exhibited linear variations in sound speed gradient at the beginning or end of observations. This trend may result from segments were excluded, the linear variation disappeared, suggesting that the sound speed gradient derived from GARPOS can serve as a useful tool for data quality assessment.
Reference
Nakamura, M., (2009) Fault model of the 1771 Yaeyama earthquake along the Ryukyu Trench estimated from the devastating tunami. Geophysical Research Letters, 36, L19307
Watanabe, S., Ishikawa, T., Yokota, Y., & Nakamura, Y. (2020). GARPOS: Analysis software for the GNSS-A seafloor positioning with simultaneous estimation of sound speed structure. Frontiers in Earth Science, 8, 597532.