9:45 AM - 10:00 AM
[SGD03-04] Estimation of interplate coupling in the Hyuga-nada using GNSS and GNSS-A data
Keywords:GNSS, Crustal Deformation, Hyuga-nada
1. Introduction
Hyuga-nada is located at the western edge of the source region of the anticipated Nankai Trough megathrust earthquake. Magnitude(M) 7.1 and 6.6 earthquakes occurred in this region on August 8, 2024, and January 13, 2025, respectively. Since slow earthquakes frequently occur in this area (e.g., Yamashita et al., 2015), estimating the interplate coupling distribution is crucial for understanding future seismic potential. Previous studies have estimated the interplate coupling in Hyuga-nada using GNSS data (e.g., Wallace et al., 2009). However, these studies had limitations due to the limited number of GNSS observation points available at the time and the arbitrary modeling of upper plate deformation, leaving room for re-examination of the interplate coupling estimation. Since 2017, approximately 20 new GNSS observation stations have been installed along the Hyuga-nada coast and the Bungo Channel by the Disaster Prevention Research Institute (DPRI) of Kyoto University, Kyushu University, and Kobe University, enabling more detailed analysis of crustal deformation distribution. This study aims to clarify detailed crustal deformation patterns using GNSS and other geodetic data deployed in this region and to estimate the interplate coupling distribution.
2. Data and Methods
In this study, we used two-year-long GNSS data from April 2022 to April 2024, including stations operated by the Geospatial Information Authority of Japan (GSI) and universities. Additionally, we used approximately three years of offshore GNSS-A data from October 2021 to September 2024, provided by the Japan Coast Guard. The number of GNSS stations used in this study is 430, across Kyushu, Shikoku, Chugoku, and parts of the Kinki region, while the number of offshore GNSS-A stations is eight from Hyuga-nada to offshore Shikoku.
We used GNSS daily coordinate data processed with the precise point positioning method implemented in GipsyX (Bertiger et al., 2020) by Disaster Prevention Research Institute, Kyoto University and converted the data to a local coordinate system. We applied a least-squares linear approximation to the local coordinate system data to estimate the average velocity at each observation point relative to the Amur Plate.
Next, we used a block-fault model to estimate the interplate coupling along the boundary between the subducting Philippine Sea Plate (footwall) and the Amur Plate (hanging-wall) from the Nankai Trough to Hyuga-nada. The block-fault model represents crustal deformation as the sum of rigid block motion and elastic deformation due to fault locking. We used the TDEFNODE software (McCaffrey, 2009) for estimation.
For the estimation method, we assumed three blocks: two for the continental plate (southern Kyushu and the rest) and one for the oceanic plate. The fault interface along the Nankai Trough and Hyuga-nada were divided into small fault segments of 6 km × 3 km, represented by 12 × 17 nodes. We estimated the interplate coupling ratio at each node and then converted the estimated coupling ratio into slip deficit rates.
3. Results and Discussion
The estimated interplate coupling distribution (slip deficit rate, Fig. 1) revealed strong coupling near Cape Muroto and offshore Cape Ashizuri. In Hyuga-nada, while strong coupling was observed in some areas in the northern and southern parts, the overall coupling was moderate. The location of the M 7.1 Hyuga-nada earthquake on August 8, 2024, was estimated to be in the transition zone where the coupling changes. However, the extensive strong coupling observed offshore Hyuga-nada may be attributed to systematic errors because the corresponding deformation is not recognized in onshore observed velocity.
Next, we compared the observed and calculated velocity data (Fig. 2). The results showed good agreement in Shikoku and southern Kyushu but poorer agreement in central Kyushu and offshore areas. The discrepancy in central Kyushu is likely due to ignoring intra-plate deformation and postseismic deformation of the 2016 Kumamoto earthquake. The current preliminary model has relatively sparse node spacing to estimate slip deficit rate, and the assumed geometry of plate interface does not reflect its latest research (e.g., JAMSTEC, 2021). We plan to refine the model by updating the plate geometry and increasing the node density.
Acknowledgments
In this study, we used data from the Geospatial Information Authority of Japan and the Japan Coast Guard. For the calculation of strain rates and the estimation of interplate coupling, we utilized programs developed by Dr. Tomohisa Okazaki and Dr. Rob McCaffrey. We would like to express our gratitude here.
Hyuga-nada is located at the western edge of the source region of the anticipated Nankai Trough megathrust earthquake. Magnitude(M) 7.1 and 6.6 earthquakes occurred in this region on August 8, 2024, and January 13, 2025, respectively. Since slow earthquakes frequently occur in this area (e.g., Yamashita et al., 2015), estimating the interplate coupling distribution is crucial for understanding future seismic potential. Previous studies have estimated the interplate coupling in Hyuga-nada using GNSS data (e.g., Wallace et al., 2009). However, these studies had limitations due to the limited number of GNSS observation points available at the time and the arbitrary modeling of upper plate deformation, leaving room for re-examination of the interplate coupling estimation. Since 2017, approximately 20 new GNSS observation stations have been installed along the Hyuga-nada coast and the Bungo Channel by the Disaster Prevention Research Institute (DPRI) of Kyoto University, Kyushu University, and Kobe University, enabling more detailed analysis of crustal deformation distribution. This study aims to clarify detailed crustal deformation patterns using GNSS and other geodetic data deployed in this region and to estimate the interplate coupling distribution.
2. Data and Methods
In this study, we used two-year-long GNSS data from April 2022 to April 2024, including stations operated by the Geospatial Information Authority of Japan (GSI) and universities. Additionally, we used approximately three years of offshore GNSS-A data from October 2021 to September 2024, provided by the Japan Coast Guard. The number of GNSS stations used in this study is 430, across Kyushu, Shikoku, Chugoku, and parts of the Kinki region, while the number of offshore GNSS-A stations is eight from Hyuga-nada to offshore Shikoku.
We used GNSS daily coordinate data processed with the precise point positioning method implemented in GipsyX (Bertiger et al., 2020) by Disaster Prevention Research Institute, Kyoto University and converted the data to a local coordinate system. We applied a least-squares linear approximation to the local coordinate system data to estimate the average velocity at each observation point relative to the Amur Plate.
Next, we used a block-fault model to estimate the interplate coupling along the boundary between the subducting Philippine Sea Plate (footwall) and the Amur Plate (hanging-wall) from the Nankai Trough to Hyuga-nada. The block-fault model represents crustal deformation as the sum of rigid block motion and elastic deformation due to fault locking. We used the TDEFNODE software (McCaffrey, 2009) for estimation.
For the estimation method, we assumed three blocks: two for the continental plate (southern Kyushu and the rest) and one for the oceanic plate. The fault interface along the Nankai Trough and Hyuga-nada were divided into small fault segments of 6 km × 3 km, represented by 12 × 17 nodes. We estimated the interplate coupling ratio at each node and then converted the estimated coupling ratio into slip deficit rates.
3. Results and Discussion
The estimated interplate coupling distribution (slip deficit rate, Fig. 1) revealed strong coupling near Cape Muroto and offshore Cape Ashizuri. In Hyuga-nada, while strong coupling was observed in some areas in the northern and southern parts, the overall coupling was moderate. The location of the M 7.1 Hyuga-nada earthquake on August 8, 2024, was estimated to be in the transition zone where the coupling changes. However, the extensive strong coupling observed offshore Hyuga-nada may be attributed to systematic errors because the corresponding deformation is not recognized in onshore observed velocity.
Next, we compared the observed and calculated velocity data (Fig. 2). The results showed good agreement in Shikoku and southern Kyushu but poorer agreement in central Kyushu and offshore areas. The discrepancy in central Kyushu is likely due to ignoring intra-plate deformation and postseismic deformation of the 2016 Kumamoto earthquake. The current preliminary model has relatively sparse node spacing to estimate slip deficit rate, and the assumed geometry of plate interface does not reflect its latest research (e.g., JAMSTEC, 2021). We plan to refine the model by updating the plate geometry and increasing the node density.
Acknowledgments
In this study, we used data from the Geospatial Information Authority of Japan and the Japan Coast Guard. For the calculation of strain rates and the estimation of interplate coupling, we utilized programs developed by Dr. Tomohisa Okazaki and Dr. Rob McCaffrey. We would like to express our gratitude here.