11:00 AM - 1:00 PM
[HDS10-P07] Tsunami calculations of M7.8 and M7.6 earthquakes at the Aleutian Trench in July and October 2020
Keywords:Tsunami calculation, DART, Aleutian trench
The Pacific Plate subducts beneath the North American Plate by 5 to 7 cm per year at the Aleutian Trench to cause many earthquakes. In 2020, an M 7.8 interplate earthquake was followed by an M 7.6 intraplate earthquake after three months. These earthquakes generated small tsunamis observed at tide gauges along the Alaska Peninsula and DART buoys in the Pacific Ocean. These two earthquakes may have occurred in conjunction, which may provide important clues for understanding the nature of induced earthquakes. In this study, we calculated the tsunamis of the M 7.8 earthquake and the M 7.6 earthquake using the fault models calculated by USGS and compared them with the observed waveforms by DART.
The crustal movements were calculated using the fault models for the initial water level of the tsunamis. We solved the linear long-wave equation using the tsunami software JAGURS. The topographic data was resampled from GEBCO with a grid interval of 5 minutes. The computational domain is 135°E to 105°W and 0°N to 65°N. The computation time was 10 hours after the earthquake, the time step width was 20 seconds, and the rise time was 60 seconds. We used 13 DART stations for the M7.8 earthquake and 5 DART stations for the M7.6 earthquake.
The maximum amplitude of the calculated tsunami waveform was almost equal to that of the DART observation on the west side of the M7.8 tsunami source. For the M7.6 earthquake, the calculated waveforms could not reproduce the large withdraw in the DART observation waveforms. In response to this result, we improved the numerical model by including the effects of tsunami excitation due to horizontal displacement of the seafloor slope, seawater density, and elastic deformation of the earth due to tsunami loading. However, the accuracy of the tsunami calculations did not improve. Therefore, we plan to perform tsunami inversion analyses to estimate the finite fault models for these earthquakes.
The crustal movements were calculated using the fault models for the initial water level of the tsunamis. We solved the linear long-wave equation using the tsunami software JAGURS. The topographic data was resampled from GEBCO with a grid interval of 5 minutes. The computational domain is 135°E to 105°W and 0°N to 65°N. The computation time was 10 hours after the earthquake, the time step width was 20 seconds, and the rise time was 60 seconds. We used 13 DART stations for the M7.8 earthquake and 5 DART stations for the M7.6 earthquake.
The maximum amplitude of the calculated tsunami waveform was almost equal to that of the DART observation on the west side of the M7.8 tsunami source. For the M7.6 earthquake, the calculated waveforms could not reproduce the large withdraw in the DART observation waveforms. In response to this result, we improved the numerical model by including the effects of tsunami excitation due to horizontal displacement of the seafloor slope, seawater density, and elastic deformation of the earth due to tsunami loading. However, the accuracy of the tsunami calculations did not improve. Therefore, we plan to perform tsunami inversion analyses to estimate the finite fault models for these earthquakes.