Flood susceptibility mapping plays a critical role in flood risk management and spatial planning, particularly in regions frequently affected by hydrological extremes. This study evaluates the performance of hybrid machine learning models that integrate a multi-layer perceptron (MLP) neural network with three optimization algorithms: Artificial Bee Colony (ABC), Elephant Herd Optimization (EHO), and Differential Evolution (DE). The models were applied to the Putna River basin, Romania, using a flood inventory of historical events and fourteen flood-conditioning factors derived from topographic, hydrological, geological, and land-use data within a GIS environment. Model performance was assessed using accuracy, precision, recall, F1-score, and the area under the ROC curve (AUC). Among the tested approaches, the DE–MLP model showed the most robust and generalizable performance, achieving an accuracy of 0.982 and an AUC value of 0.985 on the testing dataset. The ABC–MLP and EHO–MLP models also produced reliable results but with slightly lower predictive capability, while the single MLP model exhibited comparatively reduced performance. Variable importance analysis indicates that slope, elevation, distance to river, and rainfall exert the strongest influence on flood susceptibility in the study area. The results demonstrate that optimization-based hybrid models, particularly DE–MLP, can significantly enhance flood susceptibility prediction and provide valuable decision-support tools for flood risk mitigation and territorial planning.
Flood susceptibility mapping is a critical tool for flood risk mitigation and land-use planning, particularly in regions prone to recurring hydrological extremes. This study, published in Natural Hazards in April 2026, evaluates the performance of hybrid machine learning models that integrate a multi-layer perceptron (MLP) neural network with three metaheuristic optimization algorithms — Artificial Bee Colony (ABC), Elephant Herd Optimization (EHO), and Differential Evolution (DE) — for flood susceptibility assessment. The models were applied to the Putna River basin in Romania using a comprehensive set of fourteen flood-conditioning factors derived from topographic, hydrological, geological, and land-use data within a GIS framework.
The results demonstrate that the DE-MLP hybrid model consistently outperforms both the standalone MLP and the other hybrid approaches, achieving the highest predictive accuracy of 0.982 and an AUC of 0.985 on the testing dataset. The ABC-MLP and EHO MLP models also yield reliable results, albeit with slightly lower performance. Variable importance analysis further identifies slope, elevation, distance to rivers, and rainfall as the most influential factors driving flood susceptibility in the study area. The study concludes that optimization-enhanced neural networks, particularly DE-MLP, can effectively capture complex, non-linear spatial patterns and provide highly accurate, interpretable flood susceptibility maps. This framework offers a scalable and robust decision-support tool for flood risk mitigation, early warning prioritization, and sustainable territorial planning, with potential transferability to other flood-prone regions worldwide.
Sources:
Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards, Volume 122, Article number 406 (2026)
https://doi.org/10.1007/s11069-026-08162-1 .
Provided by the IKCEST Disaster Risk Reduction Knowledge Service System
Comment list ( 0 )
