Building ANN-Based Regional Multi-Step-Ahead Flood Inundation Forecast Models

Abstract

A regional inundation early warning system is crucial to alleviating flood risks and reducing loss of life and property. This study aims to provide real-time multi-step-ahead forecasting of flood inundation maps during storm events for flood early warnings in inundation-prone regions. For decades, the Kemaman River Basin, located on the east coast of the West Malaysia Peninsular, has suffered from monsoon floods that have caused serious damage. The downstream region with an area of approximately 100 km2 located on the east side of this basin is selected as the study area. We explore and implement a hybrid ANN-based regional flood inundation forecast system in the study area. The system combines two popular artificial neural networks—the self-organizing map (SOM) and the recurrent nonlinear autoregressive with exogenous inputs (RNARX)—to sequentially produce regional flood inundation maps during storm events. The results show that: (1) the 4 × 4 SOM network can effectively cluster regional inundation depths; (2) RNARX networks can accurately forecast the long-term (3–12 h) regional average inundation depths; and (3) the hybrid models can produce adequate real-time regional flood inundation maps. The proposed ANN-based model was shown to very quickly carry out multi-step-ahead forecasting of area-wide inundation depths with sufficient lead time (up to 12 h) and can visualize the forecasted results on Google Earth using user devices to help decision makers and residents take precautionary measures against flooding.

A Numerical Framework for Evaluating Flood Inundation Hazard under Different Dam Operation Scenarios—A Case Study in Naugatuck River

“…This study used The Coupled Routing and Excess Storage-Soil-Vegetation-Atmosphere-Snow (CREST-SVAS) model driven by NLDAS forcing data to simulate flows in basins of Northeastern United States [13].The objective of this study is to demonstrate a numerical framework for evaluating flood vulnerability in terms of inundation at a site of interest in the Naugatuck River basin featuring critical utility infrastructure and different operation scenarios for an upstream dam. In the past, artificial intelligence (AI) has been used for the forecast of flood inundation [16,17] and dam-controlled reservoir water level [18]. Applying the AI techniques to assess flood vulnerability at this site of interest is however difficult because of the lack of necessary long-term observations at the site of interest and the existence of a major flood control dam upstream.…”

Self-organizing maps of typhoon tracks allow for flood forecasts up to two days in advance

“…We found that the predicted flood hydrographs of the 10 test events generally matched the actual flood hydrographs providing a long lead time (e.g., several days), with acceptable variation in the timing and volume of peak flows. This is a major improvement of existing prediction modeling approaches (e.g., physical, conceptual, and data-driven) that focus on rainfallrunoff mechanisms providing a short lead time (e.g., one-to sixhour) [36][37][38] . Figures 7-9 present the predicted flood hydrographs of the three test typhoon events (i.e., Typhoon Fitow, Typhoon Soulik, and Typhoon Dujuan, respectively) under three TD schemes implemented with the two FCC selection strategies.…”

Flood Risk Assessment of Global Watersheds Based on Multiple Machine Learning Models

“…With the continuous development of machine learning algorithms, their applications in the field of hydrology are becoming more and more extensive [27][28][29][59][60][61]. Flood susceptibility maps, as an important basis for watershed planning and management, have also evolved from traditional human judgment to statistical analysis methods based on big data.…”