Application of SIND model for the Prediction of Flooding in Coastal Cities

Abstract

Flooding damage is increasing due to abnormal climates. In particular, in the case of coastal cities, it is necessary to simultaneously consider inundation caused by marine disasters such as storm surge and tsunami as well as inundation due to urbanization. In order to respond to disasters in coastal cities, flood prediction must be preceded. In Korea, the intensity of disasters has been classified by frequency, and flood hazard maps have been produced accordingly. The map is generally derived through numerical simulation, which is based on a scenario, so there is a lot of uncertainty, and it is difficult to predict a disaster for which a scenario has not been established. Because the target range of the coast is wide, the calculation speed of numerical simulation is slow, and a considerable amount of time is required for inundation prediction. It is practically difficult to predict a disaster such as torrential rain in a short period of time. Therefore, in this paper, by using the SIND model, a scientific interpolation model proposed by Kim et al. (2018), the ability to predict the inundation of coastal cities was reviewed and the method of using the model was presented. The SIND model is a short-term prediction model of urban inundation for a desired scenario within the range by using a pre-established inundation forecast map, and can be used for short-term inundation prediction such as torrential rain. To examine the applicability, the accuracy of the flood hazard map derived from the SIND model for coastal cities was analyzed. As a result, it was confirmed that the shape similarity suggested by Kim et al. (2019) was about 0.7 or higher, and it was judged to be appropriate in terms of shape similarity. If the shape similarity technique used for model validation is improved to suit the urban flooding characteristics, the use of the SIND model is expected to increase.