More than 40% of the bridges in mountainous areas of Southwest China are constructed with piers having a height of over 40 m. Such piers are characterized by large structural flexibility and distributed mass. To investigate the effects of higher modes on the seismic performance of this class of bridges, shake table tests on two 1/7-scale, tall-pier models were conducted at Tongji University, Shanghai. This paper describes the design, instrumentation, and loading protocols for the tests and discusses and compares the results. Due to the higher-mode effects, the curvature at the pier base and displacement at the pier top of were found to be weakly correlated, indicating that displacement is not a reliable damage measure for tall piers. Moreover, results indicated that the contribution of higher modes can lead to the formation of an additional plastic region at midheight in the piers. However, current seismic design code guidelines are for short to medium-height piers, where the midheight region of piers is assumed to respond elastically; code guidelines are not provided for tall piers. This paper explores the effect of higher modes on the seismic performance of bridges with tall piers and suggests two methods to improve the seismic performance: (1) eliminating the midheight plastic response by including more longitudinal steel, and (2) using more confinement in the midheight region to improve pier ductility and prevent shear failure.

Numerous bridges with piers over 40 m are constructed in Southwest China, which is known as a region of high seismicity. In current research, this type of bridge is commonly simplified as a single-column system, and researchers mainly focused on how seismic performance is affected by the higher modes of columns. This study aims to investigate the seismic behavior of the entire system of tall-pier bridges subjected to near-fault ground motions, using probability-based fragility analysis. A numerical model of the entire bridge system is constructed incorporating the effects of higher-order modes of pier columns and the influence of various structural components as bearings, shear keys, and end abutments. Both pier columns and rubber bearings are considered as vulnerable components during fragility analysis. With probability seismic demand models (PSDMs) developed from extensive nonlinear time history analyses, both component-level and system-level fragility curves of the prototype bridge are constructed and compared. The results show that bridge performance is usually dominated by rubber bearings, rather than pier columns, and the relative vulnerability of components might change with the intensity of input motions. This fact implies that only focusing on the seismic demands of pier columns, as is usual in current investigations, is insufficient in estimating the damage states of tall-pier bridges; failure of bearings should be carefully considered in engineering practice as well. Furthermore, compared with fragility curves of individual components, higher seismic vulnerability is observed in those corresponding to the entire system, indicating that the seismic fragility of tall-pier bridges should be assessed at the system level rather than component level.

More than 40% of bridges in Southwest China have piers over 40 m height. The seismic behaviour of these bridges is critical, especially exposed to near-fault pulse-like ground motions. This paper conducts fragility analyses to assess the seismic vulnerability of tall pier bridges under near-fault motions. Fragility curves are developed for a typical 50-m tall pier and a 10 m conventional pier with probabilistic seismic demand models (PSDMs). When developing the PSDMs, curvature ductility at pier base is used as the demand parameter; the 5% damped spectral acceleration of the fundamental period (S_a(T₁, 5%)) is employed as an intensity measure of the motions. Results show that the 50 m pier is less vulnerable than the 10 m pier under both the near-fault and far-field motions and is unlikely to collapse in real earthquake events. Furthermore, the seismic fragility of the 50 m pier is found highly related to the pre-defined damage state capacities, indicating that the damage states should be carefully specified to properly estimate the vulnerability of tall piers. Finally, analyses are conducted for piers with heights of 40, 60, 70, and 80 m, extending the conclusions obtained from 50 m pier to a class of tall piers.