1. A Country Wide Evaluation of Sweden's Spatial Flood Modeling With Optimized Convolutional Neural Network Algorithms.
- Author
-
Panahi, Mahdi, Khosravi, Khabat, Rezaie, Fatemeh, Ferreira, Carla S. S., Destouni, Georgia, and Kalantari, Zahra
- Subjects
CONVOLUTIONAL neural networks ,FLOOD warning systems ,LANDSLIDE hazard analysis ,IMPERIALIST competitive algorithm ,ARTIFICIAL neural networks ,RAINFALL ,FLOODS - Abstract
Flooding is one of the most serious and frequent natural hazards affecting human life, property, and the environment. This study develops and tests a deep learning approach for large‐scale spatial flood modeling, using Convolutional Neural Network (CNN) and optimized versions combined with the Gray Wolf Optimizer (GWO) or the Imperialist Competitive Algorithm (ICA). With Sweden as an application case for nation‐wide flood susceptibility mapping, this modeling approach considers ten geo‐environmental input factors (slope, elevation, aspect, plan curvature, length of slope, topographic wetness index, distance from river, distance from wetland, rainfall, and land use). The GWO and ICA optimization improves model prediction by 12% and 8%, respectively, compared with the standalone CNN model performance. The results show 40% of the land area, 45% of the railroad, and 43% of the road network of Sweden to have high or very high flood susceptibility. They also show the aspect to have the highest input factor impact on flood susceptibility prediction while, for example, rainfall ranks only seven of the total 10 considered geo‐environmental input factors. In general, accurate nation‐wide flood susceptibility prediction is essential for guiding flood management and mitigation efforts. This study's approach to such prediction has emerged as well‐performing and cost‐effective for the case of Sweden, calling for further application and testing in other world regions. Plain Language Summary: Floods have the potential to negatively affect human wellbeing, infrastructures and the natural environment. Identifying regions prone to flooding is essential in avoiding such catastrophes. In the current study, standard and optimized convolutional neural network models are used to generate maps identifying the regions of Sweden with highest probability of flooding. Various topographic, hydrological, and anthropogenic factors are taken into account for the modeling. The analysis reveals numerous areas in Sweden prone to flooding, especially in the northern, central, and southeastern parts of the country. Malmo (the third largest city in Sweden) and some areas of Stockholm (capital) are the cities most susceptible to flooding. Additionally, considerable extensions of Sweden's roadways and railways might be impacted by floods. Accurate flood susceptibility mapping is required to assist policymakers and urban planners in implementing measures aimed at mitigating flood vulnerability. Furthermore, the findings show the ability of deep learning models for detecting flood‐susceptible areas, which can be implemented worldwide to improve flood management strategies and protect lives. Key Points: Deep learning model of convolutional neural network (CNN) was optimized and improved with gray wolf optimizer (GWO) and imperialist competitive algorithm (ICA) to detect flood‐prone areasDuring both the training and testing phases, the CNN‐GWO model demonstrated superior performance compared to CNN‐ICA and the standalone CNN modelThe results show 40% of the land area, 45% of the railroad, and 43% of the road network of Sweden to have high or very high flood susceptibility [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF