Director, O.H. Hinsdale Wave Research Laboratory

Oregon State University

Recent failures of coastal highway bridges during storm events have highlighted the need for improved analysis of wave-bridge superstructure interaction.  Large-scale physical models are an effective method of determining the wave forces on the complex geometry of a highway bridge while minimizing scale effects that can be enhanced by air entrainment, turbulence, and other  processes.  This work represents one of the first large-scale physical model tests for wave-in-deck loads using realistic bridge geometries and random wave forcing. Three laboratory experiments were conducted to examine realistic wave forcing on a large-scale model of a highway bridge superstructure in the 104 m-long Large Wave Flume at the O.H. Hinsdale Wave Research Laboratory at Oregon State University.  A 1:5 scale model of a typical section of the I-10 Bridge over Escambia Bay, Florida that failed during Hurricane Ivan in 2004 was used as the test specimen.  The reinforced concrete specimen consisted of six AASHTO Type-III girders with diaphragms, a deck, and railing details. The model was fitted with six tension-compression load cells to measure horizontal and vertical forces as well as overturning moments.  A unique feature of this model was its roller and rail system which allowed the specimen to move freely along the axis of wave propagation to simulate the dynamic response of the structure for two of the experiments (the first experiment was for a rigid structure).  In addition to the load cells, twelve pressure transducers were embedded in both the exterior and interior girders and along the underside of the bridge deck to record the pressure distribution profiles.  The associated hydrodynamics were measured with ten surface piercing wave gages to provide the incident, reflected, and transmitted wave heights.  The model was subjected to both periodic and random waves over a range of water depths. The data are analyzed to study the relative importance of the impulse load versus the sustained wave load, the magnitudes of the horizontal to vertical forces and their time histories to identify the modes of failure.  The data are compared to the empirical methods outlined in Cuomo et al. (2007), Kaplan et al. (1995), and proposed U.S. Federal Highway standards.   Confidence limits are provided for forces due to periodic waves, and the exceedance probabilities of normalized forces are calculated for random waves.  The authors propose a new method that calculates horizontal and vertical forces for random waves based on exceedance probabilities.  The method will calculate a reference force based on wave height, period, and water depth.  The reference force is then factored by a multiplier based on the desired exceedance probability.  This method will help engineers to make decisions regarding the retrofit of existing bridges and the design of new bridges.