A simplified procedure to numerically evaluate triggering of static liquefaction in upstream-raised tailings storage facilities

TL;DR

This paper introduces a simplified finite element procedure using the Hardening Soil model to assess static liquefaction in upstream tailings dams, emphasizing the importance of work input in failure processes.

Contribution

It proposes a calibration methodology for the HSS model to evaluate liquefaction triggering, addressing limitations of existing methods and applying it to a real TSF case.

Findings

Minor toe movements can induce strain-softening leading to progressive failure.

The calibration improves the model's ability to predict liquefaction.

Finite element analysis shows potential failure modes in upstream TSFs.

Abstract

The interest of the mining industry on the assessment of tailings static liquefaction has exacerbated after recent failures of upstream-raised tailings storage facilities (TSF). Standard practices to evaluate global stability of TSFs entail the use of limit equilibrium analyses considering peak and residual undrained shear strengths; thus, neglecting the work input required to drive the softening process that leads to progressive failure of susceptible tailings. This paper presents a simplified procedure to evaluate the static liquefaction triggering of upstream-raised TSFs by means of finite element models employing the well-known Hardening Soil model with small-strain stiffness (HSS). A calibration methodology is proposed to overcome the model limitation of not being implemented in a critical state framework, focusing on the stiffness parameters that control the rate of shear-induced plastic volumetric strains. A real TSF is modelled in Plaxis 2D to evaluate its vulnerability to liquefy due to an undrained lateral spreading at the foundation. Results show that minor movements near the toe induce the material into a strain-softening regime that leads to a progressive failure towards the structure crest.