The project investigated the use of conventional and drained driven timber piling to provide a cost-effective liquefaction mitigation strategy for transportation infrastructure. The goal was to evaluate the magnitude of densification in contractive soils by draining driving-induced excess pore pressures, as well as to investigate the possibility of draining excess pore pressures generated during earthquake-induced ground motions. The work was organized in two stages. Work on Stage 1 focused on the development of a drained timber pile prototype and the evaluation of its effectiveness in improving soil densification at a test site, controlling for such factors as spacing, drainage, and time. Work on Stage 2 focused on a controlled blasting program to compare the effectiveness of various timber pile configurations to reduce excess pore pressures and ground settlements as compared to an unimproved, liquefiable soil. Long-term observation of the cone penetration resistance, measured over a period of one-half to eight months following pile installation, occurred due to relaxation of locked-in driving stresses. Controlled blasting of the unimproved and improved ground was conducted to make comparisons of the improvement in performance to raised excess pore pressures of sufficient magnitude to liquefy the unimproved ground. First, blasting of the unimproved ground was conducted to determine the amount of explosives required to liquefy the ground as well as to serve as a control on the improved ground. Following dissipation of excess pore pressures, the control zone settled 200 mm. In comparison, soil in between the piles settled approximately 20 to 80 mm, and piles that were tipped within a dense layer settled approximately 20 mm, on average. All excess pore pressures in the improved ground were lower than those in the unimproved ground and exhibited dilationary responses (i.e., behaved in a “dense” manner) at the end of the blasting cycle. A freely available coupled fluid-mechanical finite element program, termed FEQDrain, developed for use with earthquake drains, was calibrated using the laboratory test and blast-induced excess pore pressure data. Adjustments to soil resistance parameters to reflect the greater densification showed that the pore pressure response of the ground, improved with conventional piles, could be predicted accurately. However, the simulation of blasting for the drained piles showed that the drainage elements did not work during blasting as would have been expected from the idealized finite element model. Therefore, the findings indicate that drains with a larger discharge capacity would be required to help reduce excess pore pressures during strong ground motion. Future efforts are to be aimed towards accelerating the implementation of this technology in practice. This would be pursued through the development of design methods and outreach efforts at regional and national workshops. The IDEA researcher is seeking DOT partnering for inclusion of this technology in a demonstration project for a bridge and approach fills founded in liquefiable soils, possibly leveraging the “Every Day Counts” AID demonstration funds.
