Abstract
Early and accurate flood forecasting and warning for urban flood risk areas is an essential factor to reduce flood damage. This paper presents the urban flood forecasting and warning process to reduce damage in the main flood risk area of South Korea. This process is developed based on the rainfall-runoff model and deep learning model. A model-driven method was devised to construct the accurate physical model with combined inland-river and flood control facilities, such as pump stations and underground storages. To calibrate the rainfall-runoff model, data of gauging stations and pump stations of an urban stream in August 2020 were used, and the model result was presented as an R2 value of 0.63~0.79. Accurate flood warning criteria of the urban stream were analyzed according to the various rainfall scenarios from the model-driven method. As flood forecasting and warning in the urban stream, deep learning models, vanilla ANN, Long Short-Term Memory (LSTM), Stack-LSTM, and Bidirectional LSTM were constructed. Deep learning models using 10-min hydrological time-series data from gauging stations were trained to warn of expected flood risks based on the water level in the urban stream. A forecasting and warning method that applied the bidirectional LSTM showed an R2 value of 0.9 for the water level forecast with 30 min lead time, indicating the possibility of effective flood forecasting and warning. This case study aims to contribute to the reduction of casualties and flood damage in urban streams and accurate flood warnings in typical urban flood risk areas of South Korea. The developed urban flood forecasting and warning process can be applied effectively as a non-structural measure to mitigate urban flood damage and can be extended considering watershed characteristics.
Highlights
Urban flooding is one of the most damaging natural disasters to human life and property worldwide, and the frequent occurrence of flood damage in recent years has highlighted the need to prevent and reduce urban flooding [1,2,3]
The major urban areas in South Korea are experiencing a rapid increase in flood damage due to increased urbanization and torrential rainfall caused by climate change, resulting in casualties and property damage [10,11]
Important research trends for establishing the optimal urban flood forecasting and warning method are the previous artificial intelligence-based time-series learning using a neural network theory that first began with McCulloch and Pitts [18], and based on this research, the backpropagation learning algorithm developed for the nonlinear signal processing by Lepedes [19,20,21], which initiated time-series future value forecasting studies
Summary
Urban flooding is one of the most damaging natural disasters to human life and property worldwide, and the frequent occurrence of flood damage in recent years has highlighted the need to prevent and reduce urban flooding [1,2,3]. Important research trends for establishing the optimal urban flood forecasting and warning method are the previous artificial intelligence-based time-series learning using a neural network theory that first began with McCulloch and Pitts [18], and based on this research, the backpropagation learning algorithm developed for the nonlinear signal processing by Lepedes [19,20,21], which initiated time-series future value forecasting studies. The Thames Estuary 2100 (TE 2100 project) provides a 36-h flood risk forecasting and warning by connecting weather stations and weather satellites It establishes flood management plans phased according to climate change for the Thames River basin and specific areas in London and provides situational codes of conduct. Water 2022, 14, 187 the rainfall intensity was analyzed for warning of flood damage Based on this wor forecasting and warning criteria were calculated, various deep learning models w structed, and time-series data from gauging stations were trained to forecast an water levels of flood-prone areas.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
