TY - JOUR
T1 - The 2011 Great East Japan earthquake
T2 - fragility of Japanese buildings to ground shaking
AU - Ioannou, I.
AU - Bertelli, S.
AU - Chandler, R. E.
AU - Ohno, S.
AU - Shibayama, A.
AU - Suppasri, A.
AU - Rossetto, T.
N1 - Publisher Copyright:
© The Author(s) 2025.
PY - 2025/6
Y1 - 2025/6
N2 - The quality of post-disaster data directly impacts the reliability of empirical fragility curves. Many seismic studies rely on estimated, rather than measured, ground motion intensity due to the limited number of recording stations. However, this study utilizes a unique database of over 7,817 buildings surveyed near 37 ground motion stations in the prefectures of Fukushima, Miyagi and Iwate Prefectures following the 2011 Great East Japan earthquake. This research examines damage data inland of the tsunami inundation zone, which has received little attention. A flexible Bayesian framework is employed to construct fragility curves, accounting for data overdispersion and uncertainty in the actual level of ground motion intensity experienced by the buildings. Despite large uncertainties, the results reveal that post-2000 buildings demonstrated superior seismic performance, while pre-1981 buildings were the most vulnerable. Contrary to observations from the 1995 Kobe earthquake, heavy steel frame buildings emerged as the most susceptible to damage, whereas wooden buildings performed similarly to reinforced concrete and light steel frame structures. Additionally, a sensitivity analysis indicates that the difference between the recorded ground motion intensity and the level actually experienced by the buildings surveyed in the station’s vicinity has a negligible impact on the fragility curves.
AB - The quality of post-disaster data directly impacts the reliability of empirical fragility curves. Many seismic studies rely on estimated, rather than measured, ground motion intensity due to the limited number of recording stations. However, this study utilizes a unique database of over 7,817 buildings surveyed near 37 ground motion stations in the prefectures of Fukushima, Miyagi and Iwate Prefectures following the 2011 Great East Japan earthquake. This research examines damage data inland of the tsunami inundation zone, which has received little attention. A flexible Bayesian framework is employed to construct fragility curves, accounting for data overdispersion and uncertainty in the actual level of ground motion intensity experienced by the buildings. Despite large uncertainties, the results reveal that post-2000 buildings demonstrated superior seismic performance, while pre-1981 buildings were the most vulnerable. Contrary to observations from the 1995 Kobe earthquake, heavy steel frame buildings emerged as the most susceptible to damage, whereas wooden buildings performed similarly to reinforced concrete and light steel frame structures. Additionally, a sensitivity analysis indicates that the difference between the recorded ground motion intensity and the level actually experienced by the buildings surveyed in the station’s vicinity has a negligible impact on the fragility curves.
KW - Bayesian regression
KW - Construction age
KW - Empirical fragility assessment
KW - K-Net and KiK-net stations
KW - Ordered probit model
KW - Probit model
KW - Seismic damage
KW - Structural system
UR - http://www.scopus.com/inward/record.url?scp=105004361802&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=105004361802&partnerID=8YFLogxK
U2 - 10.1007/s10518-025-02174-1
DO - 10.1007/s10518-025-02174-1
M3 - Article
AN - SCOPUS:105004361802
SN - 1570-761X
VL - 23
SP - 3173
EP - 3201
JO - Bulletin of Earthquake Engineering
JF - Bulletin of Earthquake Engineering
IS - 8
ER -