UC Berkeley

Sediment ejecta mechanism contributes significantly to the severity of liquefaction-induced ground failure (e.g., excessive land subsidence). Estimating the amount of ejected sediment is a key step to assess the severity of ground failure; however, procedures to quantify it are currently lacking. Sediment ejecta is a post-shaking phenomenon resulting from the migration and redistribution of excess-pore-water-pressure (ue) generated during earthquake shaking. The dissipation process of residual ue can trigger intense upward seepage-induced heaving or a localized piping condition in the crust layer. With sufficient artesian pressure, it can produce artesian flow while transporting the liquefied sediment to the ground surface. As more sediment is transported to the ground surface, additional ground subsidence is produced. This research proposes a new way to quantify the quantity of sediment ejecta and hence the severity of post-shaking liquefaction consequences due to sediment ejecta for level ground.

The characteristics of liquefiable sites that did and did not produce sediment ejecta manifestation during the 2010-2011 Canterbury earthquake sequence in Christchurch, New Zealand, remain unclear. The severity of liquefaction-induced ejecta manifestation for the 2010-2011 Canterbury earthquakes was overestimated or underestimated using liquefaction vulnerability indices, such as the Liquefaction Potential Index (LPI) or Liquefaction Severity Number (LSN), at several sites in Christchurch. Firstly, nonlinear effective stress analyses (ESA) of two representative level ground sites are performed to investigate the causes of these misestimations. The ESA solves a fully coupled u-p formulation using the fast OpenSees finite element code and robust PM4Sand and PM4Silt constitutive models to model the cyclic behavior of liquefiable materials. The ESA focuses on the influence of the site’s impedance contrasts and its profile of vertical hydraulic conductivity (kv) to estimate the site’s potential to produce ejecta. The ESA results show a thick, clean sand site can develop high-gradient upward seepage that is sustained after strong shaking ends to trigger seepage-induced secondary liquefaction. The upward seepage can flow rapidly within highly permeable deposits without significant restriction from low-k¬v layers and develop high artesian pressure at shallow depths to sufficiently blow the liquefied sediment. Conversely, the stratified silty soil site develops high excess pore water pressures (ue) in isolated liquefiable layers, but the overlying low-k¬v layers impede the upward seepage, so the amount and rate of upward seepage are insufficient to produce ejecta.

The Artesian Flow Potential (AFP) and Ejecta Potential Index (EPI) concepts are formulated to capture this post-shaking hydraulic mechanism. AFP estimates the required artesian pressure at a specified time step to induce upward piping in the crust layer and produce ejecta. EPI, which is the integral of AFP over time, estimates the severity of sediment ejecta by tracking the duration in which the excess hydraulic head (hexc) exceeds the critical head required for artesian flow (hicr). The excess hydraulic head profile with depth that develops during and after earthquake shaking determines the potential of upward seepage-induced piping that produces sediment ejecta. The thick, clean sand site can develop high-gradient seepage flows that are sustained after strong shaking ends to produce severe ejecta. The stratified silty soil site develops high pore water pressures in isolated soil layers, but the amount and rate of upward seepage is insufficient to produce ejecta. The proposed Ejecta Potential Index captures key aspects of the hydraulic mechanisms of liquefaction manifestation. EPI estimates the severity of sediment ejecta by tracking the duration in which the excess hydraulic head exceeds the critical head required for artesian flow. The EPI values computed from the simulations of the two sites capture the observed trends of liquefaction manifestations during the Canterbury earthquakes.

The AFP and EPI concepts are then evaluated further by applying them to 45 well-documented liquefaction field case histories. The calculated LSN, LPI, and EPI values are compared to the observed ejecta manifestation. The computed EPI values correlate well to observed ejecta amount as opposed to LSN and LPI values, which do not. The range EPI values of the sites with None, Minor, Moderate, Severe, and Extreme ejecta severity are 0–5, 15–30, 45–105, 105–300, and 360–410, respectively. The calculated EPI value is influenced by (1) hexc generated during shaking, which is predominantly determined by the location of Artesian Flow Potential depth (zAFP) and GWL, soil relative density, and ground shaking intensity; (2) site dynamic response, which depends on the properties of and impedance contrast between soil layers; and (3) the advection process, which is governed by the distribution of hexc and the kv profile of the deposit. The Ejecta Potential Index captures these key aspects of the hydraulic processes of liquefaction manifestation. EPI can be sensitive to variables such as hydraulic conductivity and groundwater level. However, it proves to be a useful index that correlates well to the ejecta manifestation observed in the field case histories. Sites with severe ejecta have high EPI values, and sites without ejecta have low EPI values.

A CPT-based procedure to estimate liquefaction-induced ejecta potential is developed by incorporating the AFP concept. The procedure is derived to estimate ejecta severity by quantifying the liquefaction ejecta demand (ED) and crust resistance (CR) in its formulation. ED quantifies the amount of upward seepage pressure due to elevated hexc during shaking. The CR is calculated based on the thickness of the nonliquefiable layer and its equivalent shear strength. The formulation is examined using 176 well-documented liquefaction field case histories consists of thick sand and partially-to-highly stratified sites. The box-and-whisker plot of the ED value from 176 cases indicates a strong correlation between the computed ED value and observed ejecta severity. Although there is scatter, the ED tends to increase systematically as ejecta severity increases as opposed to other liquefaction demand indices such as LSN, LPI, LPI¬ISH, which do not. Importantly, ED outperforms LSN, LPI, LPI¬ISH in estimating the lack of ejecta produced at partially-to-highly stratified silty soil sites. The finding leads to the development of a liquefaction severity chart (ED–CR), which shows that Severe-Extreme cases have high ED and low CR values, and the Moderate-None cases tend to be on the lower part of the chart with low ED and high CR values. Lastly, the limitations and recommendations for using the ED–CR chart to estimate ejecta severity in practice are discussed.