Citation

Latcharote P, Leelawat N, Suppasri A, Thamarux P, Imamura F. Int. J. Disaster Risk Reduct. 2018; 29: 37-54.

Copyright

(Copyright © 2018, Elsevier Publishing)

DOI

10.1016/j.ijdrr.2017.06.024

PMID

unavailable

Abstract

The 2011 Great East Japan earthquake caused a wide-ranging and devastating tsunami that reached a maximum height of 40m and caused 19,000 fatalities, particularly along the Tohoku coast of Japan. The purpose of this study is to present a new estimation of the fatality ratios based on tsunami arrival times and evacuation perspectives and to present lessons learned for future tsunami loss assessments in other areas. In addition, this study investigates influencing factors, such as age, gender, and two coastal topography types (Sanriku ria-coast and Sendai plain). The fatality ratio is calculated by the number of fatalities divided by the population at the town scale along the shoreline, and the tsunami arrival time is calculated using the TUNAMI model with nesting grids of 1350m, 450m, 150m, and 50m. Then, linear and nonlinear regression analyses are performed to develop a relationship between the fatality ratios and tsunami arrival times. For different topography types, different distributions of fatality ratios with tsunami arrival times were observed; the fatality ratios of the Sendai plain were generally higher than those of the Sanriku ria-coast for the same arrival time. Based on the results, a strong inverse correlation between the fatality ratios and the tsunami arrival times was found in the Sendai plain, while the Sanriku ria-coast must be divided into two groups to obtain this correlation. Furthermore, other influencing factors, such as age and gender, contributed to differences in estimating the fatality ratios.

Language: en

Keywords

2011 Great East Japan tsunami; Fatality ratio; Regression analysis; Survey data; Tsunami arrival time