Abstract

In 2022, a new set of probabilistic shear wave velocity (𝑉𝑆) based liquefaction triggering curves was developed for gravelly soil by Rollins et al., using a dataset of 96 liquefaction and 78 no liquefaction case histories from 17 earthquakes in seven countries. Although these curves provide liquefaction assessment based on direct field performance, they suffer from the fact that there are relatively few case histories for high 𝐶𝑆𝑅 and high 𝑉𝑆 values to define the shape of the upper branch of the triggering curves. Thus, we made shear wave velocity measurements at three sites in Valdez, Alaska where liquefaction did not occur in the Mw 9.2 1964 Great Alaska earthquake. The Multi-channel Analysis of Surface Wave (MASW) technique was used to develop several median Vs profiles at each site that account for uncertainty in the experimental dispersion data and inversion parameterizations. 𝑉𝑆-based liquefaction evaluations were then made at each site, using the 𝑉𝑆 profiles derived from each solution. Results from previous Dynamic Cone Penetrometer (DPT) tests were then used in selecting the most reasonable velocity interpretation. Based on this VS profile, the layer most likely to liquefy was selected and used to define 𝑉𝑆1 and 𝐶𝑆𝑅7.5 at the middle of this critical layer, obtaining three points of no liquefaction, that could change the shape of the upper branch of the existing VS-based liquefaction triggering curves. These preliminary results suggest that it might be necessary to shift the triggering curves to the left or steepen their slope to provide better agreement with observed performance.

Abstract

Major challenges are encountered in characterizing and performing liquefaction assessment of gravelly soils since performing traditional in-situ tests, such as the Cone Penetration Test (CPT), in gravelly soils can be challenging, and since gravelly soils are not well-represented in existing case history databases used to develop semi-empirical methods for liquefaction assessment. This has been the primary reason for the ongoing efforts to develop methods based on other invasive tests, such as shear-wave velocity (Vs), and penetration tests with larger probes and greater energy delivered to the rod, such as the Dynamic Cone Penetration Test (DPT). While CPT-based liquefaction analyses have been comprehensively studied, limited research has been conducted comparatively evaluating the performance of the other two methods. This paper performs liquefaction triggering assessment of gravelly reclamations using data from the welldocumented case history of the port of Wellington (New Zealand) using CPT, DPT, and Vs measurements. Results show that while the Vs measurements overestimate liquefaction resistance, the DPT and CPT agreed relatively well with observations from past earthquakes due to the greater sensitivity of these test methods to soil density and stratigraphy

Abstract

Liquefaction of loose saturated soil poses a significant threat to civil infrastructure during major earthquake events. Although liquefaction is most common in loose saturated sands, numerous liquefaction events in gravelly soil profiles have been reported. Assessing liquefaction resistance in gravelly soils is challenging because large particle sizes can interfere with the standard penetration test (SPT) and the cone penetration test (CPT). To address this challenge, recent efforts have focused on developing liquefaction triggering curves based on a large diameter (74 mm) dynamic cone penetrometer (DPT) blow count and normalized shear wave velocity, Vs1, which are less affected by gravel-sized particles. While based on field case histories, the curves are poorly constrained in some areas; additional case histories continue to be highly desirable. This paper describes an investigation of six gravel sites that liquefied in the 2020 Mw6.4, Petrinja, Croatia earthquake. At each site, boreholes were completed to define the soil profile, accompanied by DPT soundings and shear wave velocity profiling using the Multi-channel Analysis of Surface Waves (MASW) approach. At some sites, the DPT blow count increased through a silty clay surface layer even though the CPT cone resistance remained constant in this layer. This increase was thus attributed to side friction on the drill rods during penetration. Subsequent DPT tests performed after casing through the silty clay eliminated the rod friction. The measured blow count and shear wave velocities in the critical layers at these sites correctly predicted liquefaction using recent probabilistic DPT- and Vs1-based triggering curves