PUSHOVER ANALYSIS OF MASONRY PIERS

Masonry towers/piers have been used in bridge construction for hundreds of years. They are present in the construction of all types of bridge systems. The behavior of these masonry towers during seismic events is of concern to bridge owners, as well as bridge engineers. A distinctive characteristic of masonry structures is their composition of two distinct materials: the masonry blocks and the mortar. Masonry blocks can be natural stones, such as granite or limestone, oven-baked clay, or concrete blocks. Reinforcement has been used in the past to add strength, however, its detailing may not be sufficient to add ductility. Any of these materials can exhibit very different engineering properties. The composition and properties of mortar also vary considerably. When dealing with older masonry towers, the effects of aging can also make the properties of blocks and mortar even more varied. The tensile strength of the mortar is much weaker than the compressive strength of both the mortar and the masonry blocks. This inherent nonlinearity of the masonry towers/piers makes it particularly difficult to approximate the seismic behavior of the masonry system using an equivalent linear (elastic) approach. For a realistic seismic analysis and safe, yet cost effective, retrofit measures, an analysis method that recognizes the basic material properties of both mortar and masonry blocks is needed. Pushover analyses have been utilized in the recent past to address the structural nonlinearities during seismic events, in a simple manner. This paper investigates the use of pushover analysis method for the seismic analysis of masonry towers/piers. First, a baseline, high-resolution analysis method is utilized to investigate the effects of different factors on the pushover behavior of masonry piers. Second, two simplified analytical methods are introduced and their results are compared with the baseline method. It is shown that even though the computational effort of the simplified methods is much less than the high-resolution method, the simplified methods produce realistic and accurate results.