|
Journal: |
Journal of Geophysical Research 2017 No.122(12)
clicks:465 |
|
Title:
|
|
|
Author:
|
Takuto Minami, Hiroaki Toh, Hiroshi Ichihara, Issei Kawashima |
|
Adress: |
Earthquake Research Institute, The University of Tokyo, Tokyo, Japan |
|
Abstract:
|
We present a new finite element simulation approach in time domain for electromagnetic (EM) fields associated with motional induction by tsunamis. Our simulation method allows us to conduct three-dimensional simulation with realistic smooth bathymetry and to readily obtain broad structures of tsunami-generated EM fields and their time evolution, benefitting from time domain implementation with efficient unstructured mesh. Highly resolved mesh near observation sites enables us to compare simulation results with observed data and to investigate tsunami properties in terms of EM variations. Furthermore, it makes source separations available for EM data during tsunami events. We applied our simulation approach to the 2011 Tohoku tsunami event with seawater velocity from linear-long and linear-Boussinesq approximations. We revealed that inclusion of dispersion effect is necessary to explain magnetic variations at a northwest Pacific seafloor site, ~1,500 km away from the epicenter, while linear-long approximation is enough at a seafloor site ~200 km east-northeast of the epicenter. Our simulations provided, for the first time, comprehensive views of spatiotemporal structures of tsunami-generated EM fields for the 2011 Tohoku tsunami, including large-scale electric current circuits in the ocean. Finally, subtraction of the simulated magnetic fields from the observed data revealed symmetric magnetic variations on the western and eastern sides of the epicenter for 30 min since the earthquake origin time. These imply a pair of southward and northward electric currents in the ionosphere that exist on the western and eastern sides of the source region, respectively, which was likely to be caused by tsunami-generated atmospheric acoustic/gravity waves reaching the ionosphere. |
|
Comment: |
|
|
|
|
