Earthquake behavior of geosynthetic-reinforced retaining structures

Abstract

Geosynthetic-reinforced soil walls have so far shown a very good performance during earthquakes. Nevertheless, additional tests like reduced-scale shaking table testing can be useful in understanding the effects of various parameters. Eight different reduced-scale models were tested using the shaking table facility at the Kandilli Observatory and Earthquake Research Institute (KOERI) of Boğaziçi University in the scope of this study. A woven geotextile was used as reinforcement and concrete blocks were used as wall facing. Four tests were conducted using 1:2 scale models of two meters height, one test involved a 1:4 scale model with 1 meter height, and the remaining three models were 1:4 scale two meters high walls. The models were instrumented with eight optical laser distance sensors to measure face displacement, ten accelerometers to measure accelerations on face and top of wall, and eight special transducers to measure the strain in geotextiles. The effects of peak ground acceleration, reinforcement length and spacing, model scale, and treatment of top two rows of facing blocks on amplification of acceleration, maximum displacements during shaking, permanent displacements and geotextile stresses were investigated. Maximum accelerations observed during shaking on the wall face increased from bottom to top and increased linearly with increasing table acceleration. Geotextile length and spacing did not affect the amplification factors for acceleration and affected maximum face displacements during shaking only slightly as long as the geotextile length was meeting the minimum requirements of FHWA design procedure for seismic loading. No noteworthy permanent displacements were observed. Measured geotextile stresses were higher than the design values calculated and the difference was more pronounced in walls with short reinforcements. It is concluded that for the tested type of geosynthetic reinforced soil wall with purely frictional reinforcement-block connection, determining the length and spacing of reinforcement using the pseudo-static design approach suggested by FHWA provides satisfactory performance during seismic loading, but geotextile stresses higher than those calculated in design may be encountered.