UC San Diego

Current techniques for assessing the effects of liquefaction-induced lateral spreading on pile foundations are based on simplified analytical methods that potentially lead to estimates that vary within a wide range. This might lead to potential excessive design demands, with high expenses for pre-event mitigation. Conversely, underestimated design demands might lead to costly post-event damage remediation.

The conducted study is directed towards enhancements to the assessment of liquefaction induced lateral spreading effects on bridge foundation systems. Current simplified analysis techniques have been only been developed recently in preliminary form. In addition, quantitative data sets from large-scale experimentation are needed concerning the response of such ground-foundation scenarios.

An effort was undertaken to address the simplified method areas of applicability and potential for enhancements. Challenges in implementing the methodology are presented within a comparative scope contrasting results of a California bridge site from different studies. On this basis, insights are derived for improvement of the currently employed simplified analysis guidelines.

Furthermore, large scale shake table testing was performed on pile foundation-ground systems, under conditions of liquefaction-induced lateral spreading. A total of 7 different experiments were conducted with varying heights, ground inclination, soil profiles, pile material and cross-section. The tested models were densely instrumented, including strain gauges, total pressure and excess pore-pressure sensors, accelerometers and displacement pots. In addition, data from 4 different experiments conducted in the NIED Japan shake table facility, including single piles and pile groups and varying soil profiles were utilized to provide additional insights and characteristics.

In these tests, the laminar soil container was placed in a mildly-inclined configuration to allow for accumulation of the liquefaction-induced lateral deformations. Detailed instrumentation and data interpretation procedures enable measurement of the fundamental soil-pile interaction behavior. The loading mechanisms have large cyclic components that may act in-phase or out-of-phase along the pile embedded length.

The conducted heavily instrumented tests resulted in a wealth of quantitative response data sets, to be used for: i) drawing insights and recommendations of practical significance based directly on the observed response, ii) calibration of simplified and more elaborate computational analysis tools, and iii) enhancement of our design guidelines and practical assessment procedures. Monotonic pushover analysis based on newly derived p-y curves in this study is found to provide useful design estimates in good agreement with the observed experimental results.