Extreme rainfall in northern Italy in 2023 caused severe flooding and multiple levee failures along the Senio River. This study used a high-resolution computer model to reconstruct how the flood spread and how the levees collapsed. The results show that breaches can strongly interact with each other, sometimes causing floodwater to flow back into the river. These processes can significantly affect flood extent and timing and should be considered in flood risk assessment and emergency planning.

We proposed an innovative model to improve flood prediction and management, mitigating the impact of extreme inundations. Using advanced deep learning techniques, the proposed model accurately forecasts long-lasting river flood events in negligible computational time. The results show significant improvements in prediction accuracy, providing communities and policymakers with a valuable tool for flood management and early warning systems.

Seagrass meadows are an emblematic coastal habitat. Their sensitivity to environmental change means that it is essential to monitor their evolution closely. However, high costs make this endeavor a technical challenge. Here, we used machine learning to map seagrass meadows in 148 satellite images in the Venice Lagoon, Italy. We found that adding information such as depth of the seabed and known seagrass location improved our capacity to map temporal change in seagrass habitat.