Performance of concrete bridges in recent earthquakes

Abstract

It has been nearly 30 years since the 1971 San Fernando earthquake in California. This event was the genesis of the modern bridge seismic design specifications and construction details in the USA. Eleven years have passed since the Governor's Board of Inquiry into the cause of structure failures during the Loma Prieta earthquake of October 1989 issued its final report with the warning title ‘Competing Against Time’. The California Department of Transportation staff engineers, consulting firms, independent peer review teams, and university researchers have cooperated in an unprecedented, accelerated programme of research based bridge seismic design and retrofit strengthening to meet the challenge presented in that report. Performance of highway bridges in the January, 1994 Northridge earthquake provided reasonable assurance that those bridges which were designed or retrofitted to the new critertia and with the improved structural details can withstand expected earthquakes without collapse or serious damage. The major causes of damage in the earlier earthquakes have been separation of deck expansion joints, causing deck systems to collapse, and the failure of older non-ductile columns. Most damage in California has been suffered by bridges constructed prior to the 1971 San Fernando earthquake. Those bridges were designed for ground accelerations of 0·06 g (60 gau) with no consideration for the spectral response of the structure, for the performance of the foundation material in a seismic event, or for structural ductility. Damage to bridges in Kobe (1995), Turkey (1999) and Taiwan (1999) followed similar patterns because they were generally designed to meet older codes and had not been retrofitted to meet the latest codes. Since 1971, the California Department of Transportation (Caltrans) and the California Division of Mines and Geology have developed a comprehensive seismic map of the state which allows bridge designers to design for site specific peak rock acceleration. Research has been conducted to develop foundation soil response spectra so the seismic hazard and soil-structure interaction can be more accurately predicted. More stringent performance criteria have been adopted to prevent collapse or serious damage in major earthquakes. Soil liquefaction effects have been researched and appropriate mitigation techniques have been developed and are currently being implemented. The required confinement details have been developed to insure ductile performance in a seismic event, tested in half size laboratory models for confirmation of ductile performance, have been utilized in newer and retrofitted bridges, and field tested in three moderate earthquakes (Landers and Cape Mendocino 1992, Northridge 1994). At the national level the American Association of State Highway and Transportation Officials (AASHTO) has adopted seismic design specifications patterned after the West Coast specifications. The excellent performance of bridges utilising these newer design criteria and details gives bridge designers an indication that these structures can withstand a larger earthquake without collapse. Damage should be expected but it can be repaired in many cases while traffic continues to use the bridges.