Long-Lead-Time Prediction of Storm Surge Using Artificial Neural Networks and Effective Typhoon Parameters: Revisit and Deeper Insight

Abstract

Storm surge induced by severe typhoons has caused many catastrophic tragedies to coastal communities over past decades. Accurate and efficient prediction/assessment of storm surge is still an important task in order to achieve coastal disaster mitigation especially under the influence of climate change. This study revisits storm surge predictions using artificial neural networks (ANN) and effective typhoon parameters. Recent progress of storm surge modeling and some remaining unresolved issues are reviewed. In this paper, we chose the northeastern region of Taiwan as the study area, where the largest storm surge record (over 1.8 m) has been observed. To develop the ANN-based storm surge model for various lead-times (from 1 to 12 h), typhoon parameters are carefully examined and selected by analogy with the physical modeling approach. A knowledge extraction method (KEM) with backward tracking and forward exploration procedures is also proposed to analyze the roles of hidden neurons and typhoon parameters in storm surge prediction, as well as to reveal the abundant, useful information covered in the fully-trained artificial brain. Finally, the capability of ANN model for long-lead-time predictions and influences in controlling parameters are investigated. Overall, excellent agreement with observations (i.e., the coefficient of efficiency CE > 0.95 for training and CE > 0.90 for validation) is achieved in one-hour-ahead prediction. When the typhoon affects coastal waters, contributions of wind speed, central pressure deficit, and relative angle are clarified via influential hidden neurons. A general pattern of maximum storm surge under various scenarios is also obtained. Moreover, satisfactory accuracy is successfully extended to a much longer lead time (i.e., CE > 0.85 for training and CE > 0.75 for validation in 12-h-ahead prediction). Possible reasons for further accuracy improvement compared to earlier works are addressed.

An advanced spatio-temporal convolutional recurrent neural network for storm surge predictions

“…Several other similar studies have been carried out by Hashemi et al [32], Kim et al [33], Chao et al [34] and Das et al [35] for other geographical regions, using larger training datasets (with larger number of storms) to train neural network models. However, these efforts did not, once again, consider the temporal correlation of the storm data.…”

Know to Predict, Forecast to Warn: A Review of Flood Risk Prediction Tools

“…In contrast with other predictive models, ANN does not place constraints on input variables like knowing how distributed a variable is [88]. ANN has been applied in forecasting the likelihood of flooding at Dangola Station, Sudan [86], storm surge and typhoon floods in Longdong northeastern Taiwan [89]; determining river flow time series in Three Gorges and Gaochang, China [90], rainfall-runoff analysis in Hoshangabad Basin, India [91], streamflow forecasting in Western USA [92], flowing estimation in Mino-Sil and Segura water basins in Spain [93] and water quality prediction in Johor River, Malaysia [94].…”

Kết quả bước đầu thử nghiệm thuật toán XGBoost dự báo nước dâng do bão tại trạm Hòn Dáu

“…Dự báo nước dâng do bão theo hướng sử dụng phương pháp học máy đã được nhiều nhà khoa học trên thế giới nghiên cứu và phát triển mạnh mẽ trong thời gian gần đây. Mạng nơ-ron nhân tạo (ANN: Artificial Neural Networks) đã được sử dụng phổ biến trong dự báo độ cao nước dâng do bão [13][14][15][16][17][18][19].Với nguyên lý kết hợp các mô hình học tập có sai số cao thành một cây học tập mạnh hơn theo kiểu tuần tự nhằm mục đích xử lý bài toán học máy có giám sát với độ tin cậy cao mà mô hình XGBoost được ứng dụng nhiều trong dự báo liên quan đến lĩnh vực khí tượng thủy văn, quản lý rủi ro thiên tai, trong đó có dự báo nước dâng do bão.…”