Evaluation and remediation measures for shield tunnels under soil liquefaction

Abstract

In liquefiable soils, shield tunnels constructed with shallow soil cover tend to be influenced by soil liquefaction during earthquake. Some reports have been published for the 1995 Hyogoken-Nanbu earthquake showing the damages of utility pipes and sewage tunnels due to soil liquefaction. However, the conventional seismic design for shield tunnels considers the racking deformation as well as the induced stress only. The mechanism and behavior of shield tunnels under soil liquefaction situation are not known well. This paper focuses on the evaluation and remediation measures for shield tunnels with the consideration of soil liquefaction on the designer's point of view. The whole evaluation process consists of three major questions to be answered. First, will the surrounding soil be liquefied and affect the shield tunnels? If it does, then secondly, how does the shield tunnel react? Will the segmental linings and connecting bolts survive due to the deformation? The third, what can a designer do to remedy such a problem? The first question can be answered by some published evaluation methods. For example, the commonly used simplified empirical methods using SPT-N results such as Tokimatsu and Yoshimi (1983) method, Seed (1985) method, JRA (1990) method, NJRA (1996) method and NCEER (2001) method. However, 178 soil liquefaction cases and 128 non-liquefaction cases from 1999 Chi-Chi Taiwan earthquake have concluded that the NCEER (2001) may be most feasible to Taiwan area. The reaction of tunnel should consider the relative position of the structure and the liquefied soil. It is noted that liquefaction occurring to the soil stratum above the tunnel springline has minor effect on the tunnel. The key factor to evaluate the influences is on the quantitative extent of exited pore water pressure. Equation reduced from ''Design Standard for Railway Structure, Seismic Design (1999)'' by Japanese Transportation Railway Department is used to compute the exited pore water pressure. By the assumptions that the soil is liquefied in some specific region along the tunnel axis and the soils at the two sides of the region remain intact, the whole analysing process can therefore be: (1) computing the exited pore water pressure; (2) calculating the external force and the bending moment acting on the tunnel; and (3) computing the stress and deformation of the tunnel. Comparing the resulted values and the tolerable maintenance values, the designer can determine whether the remediation is necessary or not. Since the computation procedure indicates that greater stress and deformation may occur with a larger liquefied zone, the stress and deformation can be restrained in tolerable values while the liquefied zone is restricted. Considering the case, a practical remediation measure is suggested to grout the underneath soil every specific interval from tunnel inside. Example results show that the exited pore water pressure may cause the tunnel structure to sustain extra stress. Consequently, the alignment may deform and affect tunnel operation. The remediation design is to grout the soil beneath tunnel springline in an interval along tunnel alignment equal to 20 m. The grouting range shall reach some low liquefaction potential stratum to ensure the grouted zone is non-liquefied during earthquake. (A). Reprinted with permission from Elsevier. For the covering abstract see ITRD E124500.