In practice, dynamic nonlinear simulations are performed considering a uniform hazard spectrum, and picking ground motions to match said spectrum over a period range, for a pair of moment magnitude and hypocentral distance. The latter are being picked based on a deaggregation analysis, which represents the contribution to the hazard of different scenarios defined by a given magnitude and distance. The selection of the seismic scenario and the period range over which the ground motion are scaled, has a significant impact on the selection of ground motions, and the results of the simulations. In Eastern North America (ENA), ground motion selection is further complicated by the lack of well-recorded ground motions at magnitudes and distances of interest. In ENA, it is common to use two different scenarios which are (1) low distances and magnitudes, controlling the hazard at short period and (2) high magnitudes and distances, which control the hazard at long periods. This paper discusses the influence of the selection of the seismic hazard scenario and ground motions for the simulation of the behavior of an earth dike with a short fundamental period, under seismic loading. To do so, six different seismic scenarios are simulated, and dynamic non-linear simulations are conducted using ground motions chosen to be consistent with the different scenarios. The results obtained are compared in terms of response spectra, lateral and vertical displacements, to determine the impact of the selection of seismic scenarios on the performance of an earth dike.

The Fundão tailing’s dam failure of November 2015 in Brazil is one of the deadliest and most environmentally damaging tailings dam breaches in recent history. Roughly 32 million cubic meters (Mm3) of iron mine tailings were accidentally released in this catastrophic collapse, claiming the lives of nineteen villagers and causing major environmental concerns after polluting local water systems. As part of the forensic investigation that followed, a finite difference analysis (FDA) numerical simulation using the NorSand constitutive model in FLAC software was conducted by the panel to test the hypothesis of lateral extrusion triggered failure. Given the numerical convergence limitations reported, the purpose of the present study is to simulate the static liquefaction failure of the Fundão dam using a finite element analysis (FEA) approach in which NorSand constitutive model is adopted in Rocscience RS2 software.

A selected number of the panel’s laboratory test results were simulated in a series of FEA to determine the validity of the numerical results and the strain-softening behaviour of the tailings. Both fully undrained and drained constant shear stress tests were simulated. In addition, a computer model of the failing section of the dam’s left abutment was subsequently generated for numerical analyses following depositional details provided in the panel’s report.

The results of the FEA are compared with the FDA shear stress-strain behaviour of the tailings reported by the panel on a laboratory scale and for the dam’s failing section supporting the hypothesis of a slope failure triggered by a lateral extrusion mechanism. The implications associated with the FEA are further discussed in this paper.

Some areas in the Lower Mainland of British Columbia rely on a system of dikes as the primary flood defense system. Past studies estimate that many dikes do not meet current provincial design standards, and that a major flood could result in losses as high as $22.8 billion. The economic and social consequences of flooding combined with the vulnerability of the flood protection infrastructure in the region highlight the need for a systematic approach for prioritizing upgrades and enhancing system resiliency. Reliability-based analyses have gained popularity as a means of assessing the performance of geotechnical systems, as they allow owners and policy makers to account for uncertainties in the design process and prioritize upgrades of the areas that pose higher risks.

This paper quantifies the influence of a dike’s length on the overall system reliability for a case study site in British Columbia. First, the probability of failure of a single cross-section is determined via stochastic slope stability analyses using the Random Finite Element Method (RFEM). Then, the number of independent reaches within the length of the dike is estimated. This is achieved by simulating Gaussian random processes over the dike’s length for various correlation lengths, and determining the equivalent number of independent random variables which lead to the same failure probability as the continuous case. Finally, the overall system reliability is calculated by treating the dike as a series system of these independent random variables. Recommendations regarding the length effect of dikes on system reliability are then made for design purposes.

The Deep Soil Mixing (DSM) technique is mainly applied for foundation support, retention systems, hydraulic cut-off walls, environmental remediation, and liquefaction mitigation. Generally, the latter application is done by forming a grid pattern of DSM walls with area replacement ratios typically ranging between 30% and 50% depending on the seismic excitation. The common design methodologies for DSM grids are based either on numerical modeling, simplified analytical procedures, case histories, or a combination of these approaches. The common ground between these design methods is mitigating liquefaction of the enclosed soil within the DSM grid cells. This paper proposes a rational design of the ground improvement based on a performance-based methodology, focused on the seismic performance of a structure supported by improved ground. Similar approaches have been considered by Yamashita et al. (2018), who analysed the performance of a DSM grid under seismic loading, and by Namikawa et al. (2007) who studied the performance of both the DSM grid and the untreated soils subjected to earthquake loading. Under a performance-based approach, the design of the DSM grid would be conducted such that the target performance is achieved, thus not necessarily imposing the no-liquefaction condition on the enclosed soil within the grid. This concept is explored for the case of embankments supported on square DSM grids. Three-dimensional, non-linear numerical analyses are conducted for a generic embankment founded on DSM-improved ground. The variability of the DSM strength on the embankment performance is evaluated for a given grid configuration. Conclusions and recommendations for this type of analysis are provided.

Yamashita et al. Seismic response analysis of piled raft with grid-form deep mixing walls under strong earthquakes with performance-based design concerns, Soils and Foundations, 58, 65-84.

Namikawa et al. Finite element analysis of lattice-shaped ground improvement by cement-mixing for liquefaction mitigation, Soils and Foundations, 47, 559-576.

Although earthquakes occur in all regions of Canada, certain areas have a higher probability of experiencing damaging ground motions caused by earthquakes. In these areas, the rigorous assessment of the dynamic behaviour of earth dams and their performance improvements are crucial to eliminate the risk of their failures under seismic loadings. A comprehensive understanding of the site condition, local seismic characteristics and dynamic behaviors of constituent materials of a simulated dam is of outmost importance to perform reliable dynamic assessment. A wide variety of laboratory apparatus (i.e, cyclic simple shear cyclic triaxial and cyclic triaxial simple shear apparatus, has been developed over the past few decades to replicate the seismic loading condition on reconstituted or intact samples. However, these laboratory apparatus may have some shortcomings to accurately characterize the dynamic behaviour of tested materials. In that case, the developed laboratory-based dynamic characteristics of a tested material may vary from one apparatus to another, which subsequently affect the seismic analyses outcomes of a simulated dam. To simplify the numerical model in this study, the simulated earth dam is considered as homogeneous in terms of the compactness and density, with uniform distribution of shear wave velocity. In first step, dynamic behaviours of constituent materials of the selected dam were determined using different cyclic laboratory apparatus. Thereafter, developed dynamic behaviors of tested materials obtained from each apparatus were implemented into the numerical model in FLAC to evaluate the behavior of the simulated dam under seismic loadings compatible with the site. At the end, the numerical results obtained for each set of the dynamic models were compared together. According to these comparisons, selecting materials’ suitable cyclic models is of outmost importance to perform reliable dynamic analysis. So that, an earth dam's seismic behavior is strongly influenced by the dynamic model of its constituent material.

In 2007, Tire Derived Aggregate (TDA) was used as lightweight fill to repair a very significant embankment failure of a four-lane divided highway leading to the Canada-U.S. border crossing in St. Stephen, New Brunswick. The highway embankment was under construction when it failed at a height of 12.3 m just short of the design height of 14 m. The cause of the failure was attributed to the rapid rate of construction and the intensity of the embankment loading on the low-strength foundation soils, consisting of 15 m of soft marine clay. The reconstruction strategy used TDA lightweight fill from 1.4 million scrap tires, and a system of prefabricated vertical drains installed through the marine clay over the original failure location. The reconstruction process was staged and controlled using geotechnical instrumentation and the observational approach. The reconstructed TDA embankment was successfully completed to the original design height in 2008 and continued to be monitored into 2009. From the perspective of TDA volume, this project was the largest TDA embankment in Canada and the second largest in North America at the time of construction. This TDA highway embankment has been open to the general public and in full operation since 2008.

In 2020, twelve years after construction, the owner and original designers have gone back to the site to assess the long-term performance of the TDA embankment and compare against the 2008/2009 performance data and the original design assumptions. This case study reviews the design, construction and long-term performance of the TDA highway embankment from the perspectives of the owner and the designers. The long-term performance results show that the TDA highway embankment and pavement continues to perform in general accordance with the original design assumptions.