15:30 〜 15:45
[AHW29-01] Empirical flood depth damage functions based on insurance data from multiple flood events in Japan
キーワード：洪水リスク評価、洪水被害関数、洪水被害の発生確率、被害推定の不確実性
Flood damage estimation is an important factor in estimating the cost-benefit of public works and in making decisions regarding flood mitigation measures. In internationally standardized methods, a flood depth-damage function that expresses the relationship between the depth of inundation and damage ratio is applied to asset value, but this damage function usually has a large uncertainty. One of the causes of this uncertainty is that there are cases where there is no actual damage even at locations that appear to be flooded in inundation maps or inundation simulation models. This is thought to be due to the ground or floor being locally raised, or because the house-owners has taken flood prevention measures. In order to model these non-damaged process, we have to obtain a dataset of assets that have not been damaged, however it is difficult to collect data on house assets themselves in general. Therefore, previous flood depth-damage functions have not explicitly provided the relationship between the probability of flood-damage occurrence and inundation depth. In this study, we addressed this issue by using insurance policy data to model cases where there is no damage in flooded areas.
In this study, empirical flood depth damage functions for detached residential buildings were developed using 340 insurance data, whose contracts were in 14 inundation areas under the three major flood events between 2018 and 2020 in Japan. The developed flood depth damage functions have two components: the probability of damage occurrence with a sigmoid function and the conditional damage ratio with a cumulative distribution function of the standard normal distribution. The developed functions accurately reproduced the observed aggregate losses. Two-fold cross-validation confirmed that the proposed method shows a 13% error to the observed loss in NRMSE, which is 3% larger than estimated by the Ministry of Land, Infrastructure, Transport and Tourism (MLIT). Focusing on the 0.45–1.0 m inundation depth, the MLIT function, which does not consider the probability of no damage, overestimates the aggregated damage loss, but our method considering the probability of no damage has good estimation accuracy. At shallow inundation depths with some probability of no damage, the accuracy of damage estimation may be improved by considering the probability of damage. The proposed method is particularly suitable for estimating damage at shallow inundation depths because it considers the probability of no damage and is extensible to existing damage functions.
In this study, empirical flood depth damage functions for detached residential buildings were developed using 340 insurance data, whose contracts were in 14 inundation areas under the three major flood events between 2018 and 2020 in Japan. The developed flood depth damage functions have two components: the probability of damage occurrence with a sigmoid function and the conditional damage ratio with a cumulative distribution function of the standard normal distribution. The developed functions accurately reproduced the observed aggregate losses. Two-fold cross-validation confirmed that the proposed method shows a 13% error to the observed loss in NRMSE, which is 3% larger than estimated by the Ministry of Land, Infrastructure, Transport and Tourism (MLIT). Focusing on the 0.45–1.0 m inundation depth, the MLIT function, which does not consider the probability of no damage, overestimates the aggregated damage loss, but our method considering the probability of no damage has good estimation accuracy. At shallow inundation depths with some probability of no damage, the accuracy of damage estimation may be improved by considering the probability of damage. The proposed method is particularly suitable for estimating damage at shallow inundation depths because it considers the probability of no damage and is extensible to existing damage functions.