Tsunami and storm wave impacts on coastal bridges

Abstract

Wave loads on coastal bridges due to tsunami and storm waves are studied through a set of laboratory experiments and numerical calculations. Effects of wave nonlinearity and entrapped air on wave loading under conditions where the bridge may be partially or fully inundated are of particular interest. In addition, effects of compressibility and scaling are investigated through numerical calculations. With the destruction of bridges during recent events such as the 2011 Tohoku tsunami and hurricanes Katrina in 2005 and Ivan in 2004, this highlights the importance of this research in understanding the mechanisms of failure during such events to prevent future coastal bridge failures. Destruction of these bridges is not only financially costly, but can prevent emergency services from reaching coastal communities, thus potentially contributing to loss of life. Along with the bridges, this research is applicable to other coastal and offshore structures, such as piers, submerged breakwaters and offshore platforms, in which wave loading or entrapped air is of concern. To investigate this problem, an extensive set of experiments is conducted on a flat plate, bridge model with girders, and a bridge model with varying percentages of trapped air that serves as a valuable benchmark for understanding wave loads on coastal structures, and bridges in particular. The wave cases tested include an extensive set of water depths, wave amplitudes and wave periods (for periodic waves) to cover a wide range of solitary and shallow-water to intermediate-water depth cnoidal waves. A range of model elevations was also tested to cover a range where the entire model is fully submerged below the water surface, to where the model is fully elevated above the water surface, and in the case of the model with girders, the girders are fully elevated as well. Experiments were conducted in a two-dimensional wave flume located in the University of Hawaii at Manoa's Hydraulics Lab in the Civil and Environmental Engineering Department. The models used were representative of a 1:35 scale model of a two-lane coastal bridge, typical in an island community. To study the effects of entrapped air, a series of experiments is conducted where side panels were added to each side of the model to trap air between the girders. Then different percentages of air-relief openings are added to the panels to capture a range of cases where no air can escape between the girders, to where all the air can escape and the wave can freely interact with the bottom of the bridge deck. Data from these experiments show the largest vertical uplift forces and forces in the direction of wave propagation on a flat plate and a bridge model occur when the structure is near the still-water level (SWL). For the cases where air is trapped, the addition of air relief openings significantly reduces uplift forces. Many current empirical relations estimating wave loads on coastal bridges only take hydrostatic effects into account. When compared with empirical estimations, data from these experiments show both hydrostatic and hydrodynamic forces must be taken into consideration. Comparison is also made with numerical calculations solving incompressible Euler's equations by use of the CFD software OpenFOAM, discussed in Hayatdavoodi (2013), Seiffert, Hayatdavoodi & Ertekin (2014), and Hayatdavoodi, Seiffert & Ertekin (2014b) with excellent agreement. Effects of compressibility and scaling are tested numerically by solving compressible and incompressible Euler's equations. Numerical calculations show that the effects of compressibility on the long duration forces are small. Calculations also show that when Froude scaling is applied to forces on the model scale, the forces agree well with force calculations at the prototype scale. These results have important design implications for bridge engineers.