Evaluating Empirical and Mechanics-based Pedestrian Stability Models

Floods are one of the most damaging natural hazards in human history, claiming lives and economic losses. While flood risk can be reduced by structural and non-structural flood protection measures, the need for evacuation in floods is not eliminated in disaster-resilient cities. Walking in flood water can be dangerous, and pedestrian stability models have been developed to evaluate the risk to a pedestrian walking in flood water. This paper evaluates an empirical and a mechanics-based pedestrian stability model. The incipient velocity (beyond which the pedestrian is unstable) for the models under different flood water depths were compared. Results showed that under high water depths, the empirical model gives lower incipient velocity compared to the mechanics-based model. Given the same flow velocity, the empirical model produces higher flood hazard ratings (FHRs), an indicator of risk levels. This point is confirmed by computing the FHRs for a simulated flood event in Keighley, Bradford, UK, in which the flood water depths were relatively high (beyond 0.6 m for an extended area) and the flow velocities were relatively low (generally about 0.1 m/s with small number of points reaching around 1 m/s). The mechanics-based model produced very low risk levels that did not justify an evacuation, while the empirical model produced significant risk levels for the flooded area. This paper shows that the stability model that gives more conservative risk levels is scenario dependent. Urban planners are recommended to choose pedestrian stability models to suit their intended applications.