Intelligent Methods for Estimating the Flood Susceptibility in the Danube Delta, Romania

Over 65% of the Danube Delta (DD) is highly or very highly susceptible to flooding, according to a study that combined multicriteria decision-making and artificial intelligence techniques. Researchers used data from previously flooded regions alongside eight geographical factors to train models including deep learning neural networks, support vector machines, and multilayer perceptron. All models demonstrated excellent performance with an area under the curve (AUC) exceeding 0.908 in receiver operating characteristic (ROC) analysis. This finding underscores the significant vulnerability of the DD to flooding, providing a critical basis for implementing mitigation measures.

Costache, R., Crăciun, A., Ciobotaru, N., & Bărbulescu, A. (2024). Intelligent Methods for Estimating the Flood Susceptibility in the Danube Delta, Romania. Water, 16(23), 3511. https://doi.org/10.3390/w16233511