Open AccessA high‐resolution large‐scale flood hazard and economic risk model for the property loss insurance in J apan
Author(s) -
Kobayashi K.,
Takara K.,
Sano H.,
Tsumori H.,
Sekii K.
Publication year - 2016
Publication title -
journal of flood risk management
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.049
H-Index - 36
ISSN - 1753-318X
DOI - 10.1111/jfr3.12117
Subject(s) - tributary , drainage basin , hydrology (agriculture) , flood myth , environmental science , surface runoff , hazard , flood risk management , scale (ratio) , vulnerability (computing) , flooding (psychology) , geography , computer science , geology , cartography , geotechnical engineering , psychology , ecology , chemistry , archaeology , organic chemistry , computer security , psychotherapist , biology
This paper presents the development of a large‐scale (e.g. several thousand km 2 ) high‐resolution (e.g. 250 m) distributed rainfall–runoff/flood inundation simulation ( DRR / FI ) model and an economic loss estimation model considering the confidence interval within what is called the K yoto, K obe U niversity‐ SJNK ( KKU‐SJNK ) model. The DRR / FI model can simulate rainfall–runoff, dike‐breaks, and inland flood inundation processes simultaneously in a seamless/integrated manner with practical accuracy. The river network for the modelling includes most of the catchment main and tributary rivers; thus, DRR / FI can simulate all of the discharge/water levels of the rivers in the catchment. Data processing was carried out with ArcGIS , which handles large data sets as one sees them with a graphical interface. The coordinate system of the model is appropriately set up; thus, the model can interact with other models such as weather, climate, evacuation, vulnerability and financial models. This also makes it possible to use extensive GIS data from all over the world. Moreover a vulnerability model, what we call the KKU ‐ SJNK model, was developed. The KKU ‐ SJNK model yields the damage ratios and thus economic loss of buildings due to flooding considering the confidence interval. The models are applied to the Y odogawa R iver catchment (8240 km 2 ), the 7 th largest river catchment in J apan, which crosses six prefectures. Though the catchment size is not necessarily very large compared with continental rivers, there is seldom seen such a detailed high‐resolution large‐scale runoff‐inundation model in J apan. To validate the model, data from 1997 and 2009 floods in the Y odogawa R iver catchment was used. The results of the model exhibited the potential effectiveness of the DRR / FI + KKU ‐ SJNK model for risk management toward property loss insurance, though it also identified some difficulties. The paper presents these results.