Local scour at bridge piers in nonuniform sediment under dynamic conditions

Scour of the streambed near bridge piers is a problem of national concern. The Federal Highway Administration has implemented a program to assess every bridge over a scourable stream for vulnerability to scour. Our understanding of scour processes and the equations for estimating the scour depth near bridge piers are based primarily on laboratory research. Laboratory research is limited by the range of hydraulic conditions typically tested, is conducted primarily under steady flow conditions, and often uses uniform bed material. Previously, field measurements on scour at bridges lacked the necessary data to study the effect of nonuniform sediment on scour at piers under dynamic hydraulic conditions. However, the U.S. Geological Survey in cooperation with the Federal Highway Administration has collected and compiled field data that permit pier scour to be studied using field data.;None of the published equations satisfy the criteria for a good design equation. Although the field data display patterns similar to some of the laboratory relations, these relations are unable to accurately describe the variability of the field measurements. Analysis of the field data shows the coarse size fractions of the bed material to be important for determining the scour depth. A coefficient based on the analysis of the residuals from the current design equation was developed to account for the effect of the coarse size fraction of the bed material. This correction coefficient improved the accuracy of scour predictions for coarse bed material. The correction coefficient is successfully validated using independent field data.;The results of published predictive equations are compared with field measurements of local pier scour to evaluate the adequacy of the equations for design. Variables commonly reported from laboratory investigations are evaluated to determine if they adequately explain processes affecting pier scour in the field. Analysis of field data is complicated by the potential variation in all explanatory variables. Partial residual analysis and multiple linear regression are used to isolate the effect of bed-material and other hydraulic parameters on scour depth. An improved design equation is developed based on this analysis.