A physically-based conceptual model calibrated on experimental data for estimating the flood instability hazard for pedestrians in urban area

Abstract Urban flood events pose significant risks to pedestrian safety. Analyzing pedestrian stability, during those events, enables the correlation of hazard levels with flood dynamics, which helps to identify critical areas and suggest evacuation routes. The present study introduces a novel approach for determining instability-thresholds for pedestrians during urban flooding. The methodology is based both on instability unit strength and specific energy—two fundamental hydraulic parameters that influence the balance of a body affected by fluid flow actions. The approach employs a physically-based formula that incorporates three calibration factors to balance the static and dynamic components of pedestrian stability, resulting in a conceptual instability threshold curve optimized by fitting a dataset of 417 experimental literature cases. By combining flow velocity and water depth, we identified a graph that identifies three hazard levels. These levels were obtained by minimizing the model’s prediction errors and show areas where pedestrians are at the highest risk of being swept away. The proposed identification of critical thresholds was applied to the urban areas of Matera in the Basilicata Region of South Italy. This case study aimed to demonstrate the method’s effectiveness and its ability to provide valuable insights for flood management. By pinpointing flood-prone roads, warning areas, and safe evacuation routes, this approach supports emergency response efforts, enhances flood preparedness, and ultimately improves public safety during flooding events.