TY - JOUR
T1 - Detection of Coastal Damage using Differences in Automatically Generated Digital Surface Models (DSMs)
T2 - Application to the 2011 off the Pacific Coast of Tohoku Earthquake Tsunami
AU - Udo, Keiko
AU - Takeda, Yuriko
AU - Koshimura, Shunichi
N1 - Funding Information:
ALOS PRISM image was provided by JAXA. This research was supported by the Special Project Research of IRIDeS (International Research Institute of Disaster Science, Tohoku University).
Publisher Copyright:
© Coastal Education and Research Foundation, Inc. 2018.
PY - 2018/5/1
Y1 - 2018/5/1
N2 - The detection of coastal damage due to the 2011 Off the Pacific Coast of Tohoku Earthquake Tsunami on the southern Sendai Coast was attempted via state-of-the-art remote sensing technology using ALOS PRISM satellite images with a spatial resolution of 2.5 m. Digital surface models (DSMs) were generated semi-automatically, and their accuracies were confirmed via comparisons with digital elevation models generated from aerial laser measurements. The differences between the DSMs prior to and after the tsunami were analyzed statistically for each type of land cover: residential districts, rice fields, coastal forests, and beaches. It was demonstrated that the damage to the coastal forests and residential districts could be detected though the damage in the low land area was difficult to be detected because of its small undulations in topography. Low pass filtering was employed to remove the jitter noise; however, the major characteristics such as damage in the coastal forests and residential areas were not changed by the filtering; that is, the major statistical characteristics could be detected without filtering.
AB - The detection of coastal damage due to the 2011 Off the Pacific Coast of Tohoku Earthquake Tsunami on the southern Sendai Coast was attempted via state-of-the-art remote sensing technology using ALOS PRISM satellite images with a spatial resolution of 2.5 m. Digital surface models (DSMs) were generated semi-automatically, and their accuracies were confirmed via comparisons with digital elevation models generated from aerial laser measurements. The differences between the DSMs prior to and after the tsunami were analyzed statistically for each type of land cover: residential districts, rice fields, coastal forests, and beaches. It was demonstrated that the damage to the coastal forests and residential districts could be detected though the damage in the low land area was difficult to be detected because of its small undulations in topography. Low pass filtering was employed to remove the jitter noise; however, the major characteristics such as damage in the coastal forests and residential areas were not changed by the filtering; that is, the major statistical characteristics could be detected without filtering.
KW - Automatic DSM generation
KW - coastal forest damage
KW - tsunami disaster
UR - http://www.scopus.com/inward/record.url?scp=85051395566&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85051395566&partnerID=8YFLogxK
U2 - 10.2112/SI85-140.1
DO - 10.2112/SI85-140.1
M3 - Article
AN - SCOPUS:85051395566
SN - 0749-0208
VL - 85
SP - 696
EP - 700
JO - Journal of Coastal Research
JF - Journal of Coastal Research
ER -