Local scour around bridge foundations has been recognized as one of the main causes of bridge failures. The objective of this study is to investigate the relationships among field, laboratory, and numerical data for the purpose of improving scour prediction methods for complex bridge piers. In this study, three field sites in Georgia were selected for continuous monitoring and associated laboratory models were fabricated with physical scale ratios that modeled the full river and bridge cross sections to consider the effect of river bathymetry and bridge geometry. Three different sizes of sediment and several geometric scales of the bridge pier models were used in this study to investigate the scaling effect of relative sediment size, which is defined as the ratio of the pier width to the median sediment size. The velocity field for each bridge model was measured by the acoustic Doppler velocimeter (ADV) to explain the complicated hydrodynamics of the flow field around bridge piers as guided by the results from a numerical model. In each physical model with river bathymetry, the comparison between the results of laboratory experiments and the measurements of prototype bridge pier scour showed good agreement for the maximum pier scour depth at the nose of the pier as well as for the velocity distribution upstream of each bridge pier bent. Accepted scour prediction formulae were evaluated by comparison with extensive laboratory and field data. The effect of relative sediment size on the local pier scour depth was examined and a modified relationship between the local pier scour depth and the relative sediment size was presented. A useful methodology for designing physical models was developed to reproduce and predict local scour depth around complex piers considering Froude number similarity, flow intensity, and relative sediment size.

No description available.