16:00 〜 16:15
[HDS10-09] 海洋短波レーダーを用いた2024年日向灘地震の津波震源インバージョン
キーワード：海洋短波レーダー、震源インバージョン、2024日向灘地震、N-net
On 8 August 2024, at 16:42:55 JST (07:42 UTC), a Mw 7.1 earthquake occurred in the Hyuganada Sea off the coast of Miyazaki Prefecture, Kyushu. The earthquake triggered a tsunami, and the Japan Meteorological Agency (JMA) issued a tsunami advisory for nearby Kochi and Miyazaki Prefectures within 3 minutes. The tsunami peaked about 40 minutes after the earthquake, with a maximum height of about 40 cm at the Aburatsu tide gauge. Moreover, after this earthquake, JMA issued an unprecedented public advisory on higher-than-usual risks of a megaquake in the Nankai Trough.
Source inversion based on tsunami data is an effective method to understand the source properties such as earthquake slip distribution. Previous studies on source inversion have used tsunami data from tide gauges, offshore bottom pressure gauges (OBPG), and satellites, which record tsunami height. High-frequency (HF) radar emits shortwave radio waves from land stations towards the sea surface and receives backscattered radio waves. It measures the velocity of sea surface currents and detects incoming tsunami signals. The information on tsunami velocity can also be used in source inversion based on linear long-wave models.
We conducted a joint inversion of the source of the 2024 Hyuganada earthquake using OBPGs, tide gauges, and HF radar. The OBPGs belong to the Nankai Trough Seafloor Observation Network for Earthquakes and Tsunami (N-net), extending from offshore Kochi Prefecture to Hyuganada Sea. The HF radar is installed along the coast of Miyazaki Prefecture (frequency: 13.5 MHz). Within each hour, continuous 18-minute current velocity observations are conducted. We set up 25 subfaults (5 along the strike × 5 along the dip), each with a size of 12 km × 12 km. Green’s functions were computed between the subfaults and tsunami observational points (height or velocity). We estimated the slip distribution using non-negative least squares regression.
The magnitude of the finite fault model obtained from the joint inversion is generally consistent with that obtained from seismic wave inversion, but the model provides better tsunami simulation results. Additionally, HF radar data has a significant impact on the slip distribution in the western part of the fault. After including the HF radar data, the fit with the tide gauge on the Kyushu coast (west of the fault) improves. The main innovation of this study is the introduction of HF radar for tsunami source inversion. It demonstrates that current velocity data can also be used in inversion, allowing for a more diverse set of data in source inversion.
Source inversion based on tsunami data is an effective method to understand the source properties such as earthquake slip distribution. Previous studies on source inversion have used tsunami data from tide gauges, offshore bottom pressure gauges (OBPG), and satellites, which record tsunami height. High-frequency (HF) radar emits shortwave radio waves from land stations towards the sea surface and receives backscattered radio waves. It measures the velocity of sea surface currents and detects incoming tsunami signals. The information on tsunami velocity can also be used in source inversion based on linear long-wave models.
We conducted a joint inversion of the source of the 2024 Hyuganada earthquake using OBPGs, tide gauges, and HF radar. The OBPGs belong to the Nankai Trough Seafloor Observation Network for Earthquakes and Tsunami (N-net), extending from offshore Kochi Prefecture to Hyuganada Sea. The HF radar is installed along the coast of Miyazaki Prefecture (frequency: 13.5 MHz). Within each hour, continuous 18-minute current velocity observations are conducted. We set up 25 subfaults (5 along the strike × 5 along the dip), each with a size of 12 km × 12 km. Green’s functions were computed between the subfaults and tsunami observational points (height or velocity). We estimated the slip distribution using non-negative least squares regression.
The magnitude of the finite fault model obtained from the joint inversion is generally consistent with that obtained from seismic wave inversion, but the model provides better tsunami simulation results. Additionally, HF radar data has a significant impact on the slip distribution in the western part of the fault. After including the HF radar data, the fit with the tide gauge on the Kyushu coast (west of the fault) improves. The main innovation of this study is the introduction of HF radar for tsunami source inversion. It demonstrates that current velocity data can also be used in inversion, allowing for a more diverse set of data in source inversion.