The Toolbox for Bridge Scour Analysis Technology: Review and Roadmap

This review paper provides a comprehensive synthesis of bridge-scour evaluation methods in the United States, tracing their evolution from the post-Schoharie Creek collapse development of HEC-18 to contemporary deterministic, field-based, numerical, stochastic, and machine-learning approaches. It examines the foundational empirical equations derived from laboratory flume studies, contrasts them with large-scale field investigations and region-specific envelope curves, and highlights recent advances in turbulence-resolved hydraulics, computational fluid dynamics modeling, and sediment-transport theory. The review also evaluates emerging probabilistic and data-driven frameworks, including Bayesian inference, decision trees, ensemble learning, and uncertainty-quantified scour prediction, that offer improved accuracy and risk-informed decision-making compared with traditional methods. Collectively, the findings underscore both the progress achieved and the persistent limitations in current scour appraisal practice, emphasizing the need for integrated, multidisciplinary, and probability-risk-based methodologies to support more reliable and cost-effective bridge-foundation design and system-level asset management. Finally, a technology roadmap is proposed to guide bridge administrators, engineers, the transportation industry, and researchers in advancing solutions to the persistent challenges of bridge scour.