Geotechnical Engineering Online Library

The objective of this paper is to investigate the effects of initial particle gradation and rock content on the crushing behavior (i.e. grain size before and after crushing) of weathered phyllite fills. Compaction tests were conducted on weathered phyllite fills with rock contents of 35%, 45%, 55%, 65% and 75% (by weight). First, the particle size distributions (PSDs) were observed before and after compaction, and then the particle breakage of weathered phyllite fills was analyzed by fractal dimension. Relative fractal dimension was proposed to evaluate the effects of initial rock content and initial gradation on the particle breakage. It was found that the fractal dimension method can well characterize the crushing behaviors of the weathered phyllite fills. The finer the fills were, the more they were compacted. That is, after the first compaction, the relative fractal dimension of the weathered phyllite fills increased as the rock content increased, reaching the values of 0.013, 0.016, 0.024, 0.037 and 0.08, respectively. After the second compaction, these relative fractal dimension values, dominated by the initial particle gradation, became 0.059, 0.072, 0.052, 0.095 and 0.118, respectively. In conclusion, the weathered phyllite fills with 55% rock content exhibited the least breakage and were most suitable for filling the subgrade. Findings in this paper will provide significant guidance for the construction of weathered phyllite filling subgrade in future projects.

In this context, we experimentally studied the anisotropic mechanical behaviors of rough-walled plaster joints using a servo-controlled direct shear apparatus under both constant normal load (CNL) and constant normal stiffness (CNS) conditions. The shear-induced variations in the normal displacement, shear stress, normal stress and sheared-off asperity mass are analyzed and correlated with the inclination angle of the critical waviness of joint surfaces. The results show that CNS condition gives rise to a smaller normal displacement due to the larger normal stress during shearing, compared with CNL condition. Under CNL conditions, there is one peak shear stress during shearing, whereas there are no peak shear stress for some cases and two peaks for other cases under CNS conditions depending on the geometry of joint surfaces. The inclination angle of the critical waviness has been verified to be capable of describing the joint surface roughness and anisotropy. The joint surface is more significantly damaged under CNS conditions than that under CNL conditions. With increment of the inclination angle of the critical waviness, both the normal displacement and sheared-off asperity mass increase, following power law functions; yet the coefficient of determination under CNL conditions is larger than that under CNS conditions. This is because the CNS condition significantly decreases the inclination angle of the critical waviness during shearing due to the larger degree of asperity degradation.

Mechanical cutting provides one of the most flexible and environmentally friendly excavation methods. It has attracted numerous efforts to model the rock chipping and fragmentation process, especially using the explicit finite element method (FEM) and bonded particle model (BPM), in order to improve cutting efficiency. This study investigates the application of a general-purpose graphic-processing-unit parallelised hybrid finite-discrete element method (FDEM) which enjoys the advantages of both explicit FEM and BPM, in modelling the rock chipping and fragmentation process in the rock scratch test of mechanical rock cutting. The input parameters of FDEM are determined through a calibration procedure of modelling conventional Brazilian tensile and uniaxial compressive tests of limestone. A series of scratch tests with various cutting velocities, cutter rake angles and cutting depths is then modelled using FDEM with calibrated input parameters. A few cycles of cutter/rock interactions, including their engagement and detachment process, are modelled for each case, which is conducted for the first time to the best knowledge of the authors, thanks to the general purpose graphic processing units (GPGPU) parallelisation. The failure mechanism, cutting force, chipping morphology and effect of various factors on them are discussed on the basis of the modelled results. Finally, it is concluded that GPGPU-parallelised FDEM provides a powerful tool to further study rock cutting and improve cutting efficiencies since it can explicitly capture different fracture mechanisms contributing to the rock chipping as well as chip formation and the separation process in mechanical cutting. Moreover, it is concluded that chipping is mostly owed to the mix-mode I-II fracture in all cases although mode II cracks and mode I cracks are the dominant failures in rock cutting with shallow and deep cutting depths, respectively. The chip morphology is found to be a function of cutter velocity, cutting depth and cutter rake angle.

Numerical methods are helpful for understanding the behaviors of geotechnical installations. However, the computational cost sometimes may become prohibitive when structural reliability analysis is performed, due to repetitive calls to the deterministic solver. In this paper, we show how accurate and efficient reliability analyses of geotechnical installations can be performed by directly coupling geotechnical software with a reliability solver. An earth dam is used as the study object under different operating conditions. The limit equilibrium method of Morgenstern-Price is used to calculate factors of safety and find the critical slip surface. The commercial software packages Seep/W and Slope/W are coupled with StRAnD structural reliability software. Reliability indices of critical probabilistic surfaces are evaluated by the first- and second-order structural reliability methods (FORM and SORM), as well as by importance sampling Monte Carlo (ISMC) simulation. By means of sensitivity analysis, the effective friction angle (ϕ′) is found to be the most relevant uncertain geotechnical parameter for dam equilibrium. The correlations between different geotechnical properties are shown to be relevant in terms of equilibrium reliability indices. Finally, it is shown herein that a critical slip surface, identified in terms of the minimum factor of safety (FS), is not the critical surface in terms of the reliability index.

Prefabricated Vertical Drains (PVDs) are typically used for embankment construction over saturated soft cohesive soil deposits to accelerate consolidation and reduce construction time in the field. PVDs accelerate consolidation of thick soil deposits by reducing the drainage path from tens of meters to 1-2 meters depending on PVD spacing in the field. Current design methodologies typically consider the increase of shear strength due to accelerated consolidation, but still use undrained shear strength for the entire cohesive soil layer even after PVD’s are installed. However, for cases in which PVDs are closely spaced, which allows excess pore water pressure to dissipate relatively fast, the assumption of undrained conditions for design may be overly conservative and, in some cases, this assumption may render an embankment construction unfeasible, unless additional ground improvement techniques are used to significantly enhance the foundation strength. This paper presents a Hybrid Drained-Undrained (HDU) model for construction of embankments over soft soils that accounts for the improved soil drainage conditions after installation of PVDs in the assessment of the shear strength used for design. A field case study is presented where the HDU methodology was used for the design of a 2.4-km long MSE berm constructed over a PVD-improved soft soil site, allowing for significant cost savings. The HDU approach was implemented using limit equilibrium models during the design stages to analyze the global stability of the MSE berm at different stages. Finite element models calibrated using field monitoring data collected during construction showed factors of safety comparable with that calculated using the HDU approach, which further supports the suitability of the HDU approach for PVD design.

In arid to semi-arid regions alluvial soil deposits can range from collapse-susceptible, variably cemented to indurated. In these soils, reliable estimation of soil properties based on penetration resistance indicators such as N-values from Standard Penetration Tests (SPTs) is difficult. Varying content of gravels and cobbles provides additional complexity. In such conditions, sample recovery from SPTs is limited and refusal N-values occur. This paper presents a database of 74 pressuremeter tests along portions of Interstate 10 (I10) through Tucson, Arizona, where soil conditions noted above were encountered. The results are used to evaluate the subsurface stratigraphy. Estimated side and base resistance for drilled shafts are presented and compared with recommendations by the American Association of State Highway and Transportation Officials (AASHTO) and the Federal Highway Administration (FHWA). Comparisons of estimated side and base resistances from pressuremeter tests with those from a large-scale load test are presented. Evaluation of the collapse potential of soils using pressuremeter tests is also discussed and the results compared with collapse potential based on 1-D laboratory tests.

A parametric study of undrained stability of a spherical cavity in clays is investigated by finite element limit analysis with an axisymmetric condition. Influences of cover depth ratio of cavity and dimensionless overburden factor on predicted failure mechanisms and dimensionless load factor are examined. It is found that a previously recommended and up-to-date lower bound solution to the problem was significantly inaccurate for practice use. Thus, an accurate approximate solution to the problem is proposed from nonlinear regression analysis of the computed average bound solutions. New cavity stability factors for the soil cohesion and soil unit weight are proposed. New findings are revealed for the three-dimensional effect of the cavity shape on these factors between the axisymmetric and plane strain conditions, and their applications to the undrained stability evaluation of cavity problems in practice are described.

Damage smear method (DSM) is adopted to study trans-scale progressive rock failure process, based on statistical meso-damage model and finite element solver. The statistical approach is utilized to reflect the mesoscopic rock heterogeneity. The constitutive law of representative volume element (RVE) is established according to continuum damage mechanics in which double-damage criterion is considered. The damage evolution and accumulation of RVEs are used to reveal the macroscopic rock failure characteristics. Each single RVE will be represented by one unique element. The initiation, propagation and coalescence of meso-to macro-cracks are captured by smearing failed elements. The above ideas are formulated into the framework of the DSM and programed into self-developed rock failure process analysis (RFPA) software. Two laboratory-scale examples are conducted and the well-known engineering-scale tests, i.e. Atomic Energy of Canada Limited's (AECL's) Underground Research Laboratory (URL) tests, are used for verification. It shows that the simulation results match with other experimental results and field observations.

Forensic analysis of past failures is valuable to improve our understanding of levee behavior. In this article a new systematic approach of forensic analysis for levee failures is proposed and applied to the Breitenhagen levee breach that occurred along the river Saale in Germany in 2013. The purpose of this study is to identify the cause of the breach based on the proposed approach, even though limited data is available. Based on the information prior, during and after the breach of the levee, a slope stability model is developed for the entire event. First, results from this model are obtained based on the expected values of the uncertain parameters and the best estimates of the situation. Uncertainty of the model is included in the calculation subsequently by defining possible failure scenarios. The most likely failure scenarios are derived from the data and included into the model so that it is possible to eliminate or validate all possible causes by means of a sensitivity calculation. It is concluded that the levee breach is likely caused by locally weak soil conditions, unexpected high water pressures due to a connection between a pond and the aquifer and unexpected saturation of the levee. These conditions are associated with the occurrence of a previous breach at this location.

Internal erosion through or beneath levees frequently occurs during flood events, resulting in sand boils. Determining the extent of damage from the seepage channels that caused the sand boils remains difficult, and many sand boils are simply monitored in subsequent floods. There is a need for a nondestructive methodology to map the voids caused by seepage channels so that needed repairs can be identified prior to the next flood. The objective of this study is to identify the electrical response of internal erosion pathways with electrical resistivity tomography (ERT). A case study using ERT on the Hager Slough Levee, which has experienced sand boils since the 1990s, is presented. Three ERT surveys were conducted along the levee in an area of apparent damage (i.e., crown settlement and sand boils) and where nearby boring logs were available. Laboratory resistivity tests provide support for the field ERT interpretations. The results indicate that ERT may be used to identify zones of damage within and beneath a levee. If used in practice, ERT has the potential to improve remediation of levees by identifying needed repairs.

Slope failures are a common occurrence in tropical regions with a high intensity of rainfall. Tropical areas such as Singapore are normally covered with residual soils whose behaviour does not follow the principles of classical saturated soil mechanics because these soils are often unsaturated in nature. The negative pore-water pressure in unsaturated soil is highly influenced by the changes in the flux boundary conditions, resulting from the variation in climatic conditions. On the other hand, the negative pore-water pressure contributes additional shear strength to the unsaturated soil. As water infiltrates into the slope, pore-water pressure in the slope increases (matric suction decreases), and the additional shear strength due to matric suction will decrease, causing the slope to be more susceptible to failure. Singapore is a land scarce country with a critical need to optimize land utilization. Steepening slopes or cutting back slopes and supporting them using a retaining structure is one way to create new spaces. In this study, a new type of retaining structure, Geobarrier System (GBS) is proposed. A GBS is a man-made three-layer cover system designed as a vegetative layer combined with a two-layer unsaturated system, which harnesses the distinct difference in unsaturated hydraulic properties between a fine-grained layer and a coarse-grained layer. GBS consists of recycled materials and does not use steel or concrete and is hence more cost effective, thereby making it economical for use in urban areas. Geobag for vegetative layer is supported by specially designed pockets for planting different types of sustainable plant species. The paper presents the design, construction procedures, material selection and field performance of a GBS constructed at an inclination angle of 70o in response to rainfall infiltration. In addition, the results of the finite element seepage and slope stability analyses of the GBS subjected to extreme rainfalls are also presented. The results from field instruments and numerical analyses showed that GBS was able to protect the slope from rainfall infiltration; therefore, the stability of the slope retained by GBS was not affected by the rainfall.

Rainfall-induced slope failures commonly occur within residual soil slopes. One of the possible systems for slope preventive measure is capillary barrier system. A capillary barrier is a two-layer system of distinct hydraulic properties that is used to prevent water infiltration into the soil below the capillary barrier system by utilizing unsaturated soil mechanics principles. This paper presents the numerical simulation of the capillary barrier system as a slope preventive measure against rainfall-induced slope failures. The capillary barrier was constructed on a slope which experienced several shallow failures due to rainfall. In this study, the capillary barrier system was designed to repair the slope and at the same time to provide preventive measures for further failures due to heavy rainfall conditions of the tropics. The capillary barrier system was constructed using fine sand as the fine-grained layer and granite chips as the coarse-grained layer. Both layers were contained in geocells. The slope was instrumented with tensiometers and piezometers. The tensiometers were installed at different depths from about 0.5 m to 2.0 m below the slope surface. In addition, the adjacent original slope without the capillary barrier system was also instrumented using tensiometers in order to investigate the performance and effectiveness of the capillary barrier system in reducing rainwater infiltration and maintaining negative pore-water pressure in the slope. Results of field measurements from one-year monitoring period and numerical analyses of slope with and without capillary barrier system are presented in the paper. The results of numerical analyses of the slopes with and without capillary barrier system indicated that the capillary barrier system performed well in minimizing rainwater infiltration into the underlying soil layer. In addition, the numerical results showed that the factor of safety of a slope with a capillary barrier system was significantly higher than that of an original slope without the capillary barrier system. The field measurement and numerical analyses results were in good agreement, demonstrating the successful application of unsaturated soil mechanics principles in the design and construction of a capillary barrier system.

This paper presents a description of post-flood reconnaissance and geotechnical investigation of four earthen dams that were breached or damaged following an extreme flooding event in South Carolina in 2015. As a result of unprecedented rainfall and flooding that occurred during a five-day period at the beginning of October in 2015, a total of 51 earthen dams failed and nearly 200 were damaged. Many of these dams failed due to overtopping that led to a breach of the dam. Among the four dams investigated, full breach was observed at three dams; two of which were overtopped, and one was not. The fourth dam was not breached but was severely damaged. For each of these dams, the paper documents background information, pre-flood conditions, post-flood field observations and measurements, and laboratory testing results of collected soil samples. Impacts of vegetation on the dams and the effects of dam failure on critical infrastructure are also presented. The detailed descriptions and geotechnical investigations of the dam failures presented herein serve as case histories that can be used for dam breach modeling and risk assessment of dam failure.

A numerical method, PEI (Pile behavior in Expansive soil upon Infiltration) developed at the University of Ottawa can be used as a tool for the prediction and interpretation of the mechanical behavior of piles in expansive soils taking account of the influence of matric suction changes associated with water infiltration. This program is employed in back analyses of the in-situ single pile test results for a case study at the Colorado State University (CSU) field test site, which has expansive soil deposits. Four single piles were drilled into the expansive shale of the Pierre Formation west of Fort Collins, Colorado. The piles were 350 mm in diameter and were installed to a depth of 7.6 m. During installation, slope indicator-type VS embedment strain gauges were embedded on the piers at a depth of 1.8 m, 3.1 m, 4.3 m and 5.5 m below the ground surface. Water content variations of the surrounding soil along the pile depth were measured periodically from 1995 to 2004, in addition to other measurements, which include pile axial force distribution and pile head displacement. Extensive analysis regarding this case study with respect to the mechanical behavior of pile associated with changes in water content is available in the literature. In this paper, this case study is illustrated in an innovative way to understand the influence of matric suction on the mechanical behavior of the pile using the program PEI. In addition, in-situ measurements of both pile head displacements and pile axial force distribution are compared with the results of numerical simulations using the PEI. Analyses of the results show that measured in-situ and numerical results are in good agreement. The results of this study are of significant interest for the practicing engineers who routinely design pile foundations in expansive soils.