In a groundbreaking study published in the Journal of Hydrology: Regional Studies, a team of researchers led by Mahdi Panahi from the Department of Computer Engineering at Chosun University in South Korea, has unveiled a powerful new approach to predicting global flood hotspots. The study, which assessed the efficiency of three machine learning approaches, including support vector regression (SVR) and its optimized versions through combination with grey wolf optimizer (GWO) and whale optimization algorithm (WOA), offers a significant leap forward in flood susceptibility mapping.
Floods are among the most catastrophic and dangerous natural calamities globally, causing irreparable damage to human lives and property, and environmental degradation. The study utilized data from 6682 historical flood events, covering eight flood-related geo-environmental factors to generate accurate flood susceptibility maps. The maps were evaluated based on root mean square error (RMSE), mean squared error (MSE), standard deviation, and area under the receiver operating characteristic curve (AUC).
The findings reveal that the SVR-GWO model has the best performance in predicting flood-prone areas worldwide based on AUC, RMSE and MSE. “The SVR-GWO model outperformed other models in terms of accuracy and reliability,” Panahi stated. “This is a significant step forward in our ability to predict and mitigate the impacts of floods on a global scale.”
The study indicates that approximately 17.14% of the global land area is highly and very highly susceptible to flood occurrence. Flood hot-spot countries were identified as the United States of America (7.75%), Indonesia (6.33%), India (6.31%), Brazil (5.33%) and Nigeria (4.08%). Countries with the lowest probability of flood occurrence were the Russian Federation, Canada, Greenland, the United States of America and China. This information is crucial for the energy sector, as many power plants and infrastructure are located in flood-prone areas. By identifying these hotspots, energy companies can better prepare and protect their assets, reducing the risk of costly disruptions and damage.
The study also highlights the potential for incorporating additional satellite-based environmental data to further enhance the model’s accuracy. This could lead to more precise and reliable flood predictions, enabling better management strategies and resource allocation. “Incorporating additional satellite-based environmental data could further enhance the model’s accuracy,” Panahi explained. “This approach sets a foundation for future research in tailoring flood prediction models to regional scales, addressing the diverse challenges posed by different geographic and environmental settings.”
The implications of this research are far-reaching. By providing a more accurate and reliable tool for flood susceptibility mapping, the study paves the way for improved disaster management and mitigation strategies. This is particularly important for the energy sector, where floods can cause significant damage to infrastructure and disrupt power supply. With this new approach, energy companies can better assess their risk and take proactive measures to protect their assets.
The study, published in the Journal of Hydrology: Regional Studies, represents a significant advancement in the field of flood modeling. By leveraging machine learning and optimization techniques, the researchers have developed a powerful tool for predicting flood-prone areas and enhancing disaster management strategies. As the impacts of climate change continue to be felt around the world, this research offers a timely and valuable contribution to the field of hydrology and disaster management.