[J04-5] Geohazard early warning systems V

[J04-5-04] Improvement of tsunami-forecasting method based on tsunami inversion: small-size and large-amplitude tsunamis

^○Hiroaki Tsushima (Meteorological Research Institute, Japan Meteorological Agency, Tsukuba, Japan)

For more accurate forecasting of small-size and large-amplitude tsunamis, I improve the tFISH algorithm, which is a tsunami forecasting method based on offshore tsunami-waveform inversion for initial sea-surface height distribution. In the tFISH, initial sea-surface height distribution is expressed by a linear superposition of elementary tsunami sources each of which size is about 20 km. The size is small enough to resolve tsunami sources of moderate to large interplate earthquakes, but tsunamis that affect coastal communities are sometimes generated by a spatially small-size event such as a 2016 Nov. Mw 6.9 intraplate normal-faulting earthquake, occurring off Fukushima prefecture, northeastern Japan, and resulting in the 1.4-m-height tsunami at a coastal tide gauge. Because the source is comparable or smaller than that of an elemental source in the tFISH, the tFISH may not resolve the source and decrease accuracy of tsunami forecasting. To solve the problem, this study proposes a two-steps inversion. At the first step, the conventional tFISH inversion is performed. Then, we go to the second-step inversion when the following criterion is satisfied: the estimated source is expressed by extremely small number of elemental sources (e.g., one source only), meaning that the source may be unresolved image of the true source. At the second step, we perform tsunami inversion with elemental sources whose sizes are much smaller than the original. The elemental sources are distributed to sea area where tsunami source is imaged in the first-step inversion. Finally, we decide whether the second-step solution is chosen as the final solution using the following criterion: the comparison between observed tsunami waveforms and the calculations in the inversion shows the better agreement than in the first-step inversion. Numerical test assuming the 2016 off Fukushima earthquake shows the good performance of the new approach to avoid the unresolved tsunami source image.