Liquefaction susceptibility based on dissipated energy

R O Davis,J B Berrill

doi:10.5459/bnzsee.29.2.83-91

Abstract

    This paper presents a novel approach to characterization of liquefaction susceptibility for deposits of saturated cohesionless soils. The method we propose is based on an assumed relationship between pore pressure increase and dissipated energy density within the soil layer. Use of dissipated energy is not new. What makes the present work different is our approach to the energy calculation. Earlier analyses used simple attenuation models based on earthquake magnitude and epicentral distance to determine the dissipated energy and hence the pore pressure increase within a sand deposit. In this work, instead of magnitude and distance, we will use the response spectrum for surface motion at the site as input information. This permits us to carry out liquefaction susceptibility analyses which are more closely aligned with other types of analyses such as structural response. In particular, we can employ code-prescribed spectral loads exactly as are used by structural designers. This leads to an analysis of liquefaction which is consistent with other earthquake engineering practice in New Zealand.   

Highlights

The purpose of the work described below is to create a methodology for analysis of liquefaction susceptibility which is consistent with other earthquake engineering design practice currently in use in New Zealand
In this paper we have attempted to construct a liquefaction susceptibility model which is consistent with other design practice in New Zealand
Using a transfer matrix approach, the calculation of dissipated energy density in a multilayered soil profile due to specified surface motion has been considered in detail

Summary

INTRODUCTION

Note that the acceleration spectrum given ,by equation (3) corresponds to 5 percent structural damping While this is appropriate for structure loads, our objective is different: The undamped spectrum is desired for calculation of dissipated energy. Displacements will be bounded and we will set a limiting Fourier displacement for low frequencies equal to 2 m x sec corresponding to the value obtained from Brune's attenuation model [12,13) for a 7.5 magnitude earthquake with hypocentral distance of 12 km With these corrections, equation (4) becomes rI I U0 (ro) = i.s1/ro 2 3,90/ro Z.4 15.86/ro 3 for ro s 0.967 sec-1 for 0.967 sro s·6.28 sec-1 for 6.28 s ro s 10.47 sec-I (5) for 10.47sec-1 sro.

87 Free surface

CONCLUSION