Torsional Ground Motion Effects on the Seismic Response of Continuous Box-girder Highway Bridges

Highway bridges are particularly vulnerable to seismically-induced deformations and forces due to their complex dynamic characteristics. Therefore, many studies have been conducted to provide insight into their seismic response characteristics under various types of ground motions. Undoubtedly, investigations relating to both structural components as well as the strong ground motion characteristics play major roles in providing reliable design guidelines for highway bridges. It is generally accepted that ground motion characteristics such as spatial variability may result in differential excitations at the supports of relatively longer span highway bridges which may induce adverse interactions. In addition, the propagation characteristics of seismic waves along different wave paths result in rotational deformations on the ground surface which remains unaccounted, however, studies on implications for the seismic response of structural systems are rare. Therefore, the primary objective of the present study is to investigate force and deformation demands on highway bridge components particularly due to torsional components of earthquake ground motions, which are not considered explicitly in design codes. For this purpose, continuous concrete box-girder highway bridges were considered and torsional ground motion effects on these bridges were investigated in two phases. Firstly, the upper and lower bound effects of torsional ground motions (TGMs) on the seismic response of highway bridges were investigated through simplified computational models with varying skew angles, eccentricity between the centers of mass and rigidity, gap sizes between the deck and abutment, and overall dynamic characteristics. It is noted that highly nonlinear impact elements were included explicitly in the computational models to ensure that consistent deck rotations are accounted for. Furthermore, the translational ground motions accompanying with TGMs were applied with different incident angles. The results were compared for the cases with and without TGM and observations revealed that larger deck rotations due to TGMs lead to larger impact forces, which further amplifies the deck rotations. The observations from the first phase warranted a more comprehensive investigation that utilized three-dimensional finite element (3D FE) models of typical highway bridges. In this second phase of the study, the inelastic response characteristics of bent columns and abutment components such as shear keys, bearings, piles as well as the impact between deck and abutment, and abutment soil-structure interaction were explicitly included in OpenSees models. Furthermore, 3D FE models with a range of standard dimensions were altered to carry out an investigation on bridges with varying skew angles, number of bent columns, and column height-to-diameter ratios. Inelastic seismic responses of bridges subjected to only translational and both translational and torsional ground motions were compared. The most unfavorable TGM effects were observed when TGMs resulted in uneven and asymmetric shear key failure, in which, an instantaneous eccentricity was induced when the deck comes in contact with the shear key(s). Subsequently, the deck rotations amplified directly due to TGMs as well as due to more frequent occurrence of impact and larger impact forces. Furthermore, it was noted that TGM-induced torsional moment demand combined with axial-flexure-shear interactions may result in complex failure modes, high shear stresses, and reduction in lateral deformation and flexural capacity of the bridge columns. Finally, the observations relative to the shear key failure mechanisms suggested that superstructure displacement demands can be restrained by either preventing or delaying the failure of shear keys. In this regard, a design method that ensures effectiveness of shear keys while preventing asymmetric failure mechanisms to form, in mitigating seismic response of highway bridges was proposed. The proposed method follows the conventional approach to ensure that substructure components are capacity-protected, however formalizes a practical procedure to specify desired deformation limits associated with i) gap size between the superstructure and shear keys, and ii) ultimate deformation capacity of shear keys. The efficacy of the method was demonstrated through nonlinear response history analyses of a series of benchmark bridges. It was demonstrated that excessive in-plane deck rotations and extent of damage can be limited when the effectiveness of shear keys is maintained throughout the duration of seismic excitation