A high-fidelity seismic intensity measure to assess dynamic liquefaction in tailings

Nicolas A. Labanda,Roberto J. Cier,Mauro G. Sottile

DOI: https://doi.org/10.48550/arXiv.2204.11393

IF: 3.264

2022-04-25

Geophysics

Abstract:Deformation analyses of tailings dams under dynamic conditions require using earthquake records as input loading. Moreover, these records must represent the local seismicity, expressed by ground motion power indicators denominated intensity measures (IM). The ability and accuracy to describe the characteristics of a seismic record play a fundamental role in earthquake engineering and damage assessment of geotechnical facilities. None of the existing IMs represents a robust enough predictor of a given seismic demand (e.g., residual displacements). Different signals may generate a wide spectrum of results, with diverse effects that could produce insignificant damage to global failure depending on the structure. Usual engineering procedures select a huge number of records to overcome this limitation and develop a large set of numerical simulations to bound the uncertainty of the results, which becomes a time-consuming approach. This paper presents a new high-fidelity seismic IM to perform more accurate ground motion {selection}, which captures the spectral properties of the record for the frequency content that the dam does not filter. This IM represents a way to estimate beforehand a seismic demand, expressed, for instance, in terms of displacements. The proposed IM is applied to a finite element model for an upstream tailings dam cross-section, using a constitutive model capable of capturing dynamic liquefaction. The obtained results show that our proposal gives highly reliable correlations with different selected demands. Comparisons with classical IMs are also discussed, showing that our proposal emerges as a practical solution to a large dated discussion within our community.