Journal of Geotechnical and Geoenvironmental Engineering

This paper presents the results of a systematic laboratory investigation on the static behavior of silica sand containing various amounts of either plastic or nonplastic fines. Specimens were reconstituted using a new technique suitable for element testing of homogeneous specimens of sands containing fines deposited in water (e.g., alluvial deposits, hydraulic fills, tailings dams, and offshore deposits). The fabric of sands containing fines was examined using the environmental scanning electron microscope (ESEM). Static, monotonic, isotropically consolidated, drained triaxial compression tests were performed to evaluate the stress-strain-volumetric response of these soils. Piezoceramic bender element instrumentation was developed and integrated into a conventional triaxial apparatus; shear-wave velocity measurements were made to evaluate the small-strain stiffness of the sands tested at various states. The intrinsic parameters that characterize critical state, dilatancy, and small-strain stiffness of clean, silty, and clayey sands were determined. All aspects of the mechanical behavior investigated in this study (e.g., stress-strain-volumetric response, shear strength, and small-strain stiffness) are affected by both the amount and plasticity of the fines present in the sand. Microstructural evaluation using the ESEM highlighted the importance of soil fabric on the overall soil response.

The undrained remolded shear strength of soft clays is of importance in geosystem design, particularly for offshore structures. Common methods to estimate remolded shear strength, such as correlations with cone penetration data, direct measurement with an in situ field vane shear device, and laboratory measurements, produce varied results and can be particularly costly and time consuming. Full-flow penetrometers (T-bar and Ball) provide an alternative rapid method to estimate remolded shear strength and soil sensitivity through remolding soil by repeated cycling of the penetrometer up and down over a given depth interval. The cyclic penetration resistance degradation curve inherently contains information regarding remolded strength and sensitivity. The objective of this paper is to assess the ability of full-flow penetrometers to predict remolded strength and soil sensitivity, and to develop a suite of predictive correlations in which these properties can be estimated in the absence of complementary laboratory or in situ test data. To accomplish this, full-flow penetration profiles and cyclic tests were performed at five well characterized soft clay sites, which together represent the broad range of soils in which the penetrometers will be often used. A previously developed model for the reduction in penetration resistance with cycling is modified to predict the entire degradation curve, including the remolded penetration resistance using only measurements obtained during initial penetrometer penetration and extraction. Using field vane shear strength as the reference measurement, correlations are developed to predict soil sensitivity and remolded shear strength based solely on full-flow penetrometer data, which is particularly useful in site investigation programs where site specific data are not yet available or are sparse. Finally, the usefulness of these relationships is demonstrated by implementing them for two additional soft clay sites.

The particle-size distribution of soil with mean particle size and fines content are used not only in soil classifications but also in a number of other soil property relationships. In this study, two simple relationships (hyperbolic $\left[\text{tan}h\left(x\right)\right]$ and $S$ -curve) were investigated to represent the particle size distribution of soils. The parameters of the hyperbolic model were correlated to various soil parameters such as the mean particle size, particle size range, and fines content. There was no direct correlation between Fredlund (four-parameter model) and $S$ -curve model parameters and the soil parameters. The predictions of the two (hyperbolic) and three ( $S$ -curve) parameter models were compared to the four-parameter model (unimodal) using limited soil data from the literature and the agreements were good. The hyperbolic model was used to map the Unified Soil Classification System. A recent study had quantified the relationship between the grouting pressure and the fines content in nonplastic soils. Also in the current practice, upper and lower particle-size distribution limits are used in determining the groutability of soils. In this study, the relationship between grouting pressure and fines contents of the soil was generalized using the hyperbolic particle-size distribution model and verified with a groutability study using an acrylamide grout. Based on limited data in the literature, the groutability of soils was defined using a new set of parameters, grouting pressure, fines content, and mean particle size diameter of the soil.

Grouted soil nails are widely used in slope stabilization. The influence of both grouting pressure and overburden stress on the soil-nail pullout interface shear resistance is still not well understood due to the complex of soil-grout interactions. A series of laboratory soil-nail pullout tests have been carried out on a completely decomposed granite soil in nearly saturated condition under a combination of different grouting pressures and overburden stresses. The pullout tests simulate the real construction process of a soil nail, including establishment of initial soil stresses in a soil slope, drilling a hole with stress release, grouting, and soil-nail pullout when the slope is sliding. The pullout box is well instrumented. Test data are collected automatically by a data logger. Typical test results are presented, explained, and discussed in this paper. The soil-nail interface shear resistance data from all tests are analyzed and interpreted. The study shows that the grouting pressure and overburden stress have interactional influence on the soil-nail pullout resistance. Based on the test results, a new empirical liner equation with two grouting pressure dependent parameters is proposed for calculation of soil-nail pullout resistance considering both grouting pressure and overburden stress. New understandings and findings from the study are presented.

This paper explains and evaluates the potential and limitations of conducting cross-well radar (CWR) in sandy soils. Implementing the experiment and data collection in the absence of any scattering object, and in the presence of an acrylic plate [a representative of dielectric objects, such as dense nonaqueous phase liquid (DNAPL) pools, etc.], as a contrasting object in a water-saturated soil is also studied. To be able to image the signature of any object, more than one pair of receiving and transmitting antennas are required. The paper describes a method to achieve repeatable, reliable, and reproducible laboratory results for different transmitter-receiver combinations. Different practical methods were evaluated for collecting multiple-depth data. Similarity of the corresponding results and problems involved in each method are studied and presented. The data show that the frequency response of a saturated coarse-grained soil is smooth due to the continuous and dominant nature of water in saturated soils. The repeatability and potential symmetry of patterns across some borehole axes provide a valuable tool for validation of experimental results. The potential asymmetry across other borehole axes is used as a tool to evaluate the strength of the perturbation on the electromagnetic field due to hidden objects and to evaluate the feasibility of detecting dielectric objects (such as DNAPL pools, etc.) using CWR. The experimental simulation of this paper models a real-life problem in a smaller scale, in a controlled laboratory environment, and within homogeneous soils that are uniformly dry or fully water saturated, with a uniform dielectric property contrast between the inclusion and background. The soil in the field will not be as homogeneous and uniform. The scaling process takes into consideration that as the size is scaled down; the frequency needs to be scaled up. It is noteworthy that this scaling process needs to be extensively studied and validated for future extension of the models to real-field applications. For example, to extend the outcome of this work to the real field, the geometry (antenna size, their separation and inclusion size) needs to be scaled up back to the field size, while soil grains will not. Therefore, soil, water, and air coupling effects and interactions observed at the laboratory scale do not scale up in the field, and may have different unforeseen effects that require extensive study.

Close agreement between theory and experiment is critical for adequate understanding and implementation of the cross-well radar (otherwise known as cross-borehole ground penetrating radar) technique, mentioned in a previous paper by the authors. Comparison of experimental results to simulation using a half-space dyadic Green’s function in the frequency domain requires development of transfer functions to transform the experimental data into a compatible form. A channel transfer function (CTF) was developed to avoid having to model the transmitting and receiving characteristics of the antennas. The CTF considers electromagnetic wave propagation through the intervening media only (soil in this case) and hence corresponds to the simulation results that assume ideal sources and receivers. The CTF is based on assuming the transmitting antenna, soil, and receiving antenna as a cascade of three two-port microwave junctions between the input and output ports of the vector network analyzer used in the experimental measurements. Experimentally determined CTF results are then compared with computational model simulations for cases of relatively dry and saturated sandy soil backgrounds. The results demonstrate a reasonable agreement, supporting both the model and CTF formulation.

This paper presents an experimental investigation of the effect of clay consolidation on parameters that govern the advective-dispersive transport of an inorganic solute. Batch, diffusion, dispersion, and solute transport tests were conducted using kaolinite clay and dilute solutions of potassium bromide (KBr). Batch tests produced the highest levels of ${\text{K}}^{+}$ sorption and indicated that equilibrium sorption was achieved in approximately 10–30 min. The increase in sorption observed in the batch tests, as compared to the dispersion or solute transport tests, reflects the significantly lower solids-to-solution ratio and more efficient mixing process. By comparison, kaolinite consolidation had little effect on sorption due to the relatively small change in porosity. Values of hydrodynamic dispersion coefficient $\left(D_h\right)$ , effective diffusion coefficient $\left(D^{\ast }\right)$ , and apparent tortuosity factor decreased with decreasing porosity. Values of $D^{\ast }$ obtained for ${\text{Br}}^{-}$ were generally larger than for ${\text{K}}^{+}$ , whereas $D_h$ values for ${\text{Br}}^{-}$ were significantly smaller than for ${\text{K}}^{+}$ . Values of longitudinal dispersivity $\left(\alpha \right)$ were larger for ${\text{K}}^{+}$ than ${\text{Br}}^{-}$ and showed no clear trend with decreasing void ratio. In general, the experimental results suggest that changes in $D^{\ast }$ and $D_h$ should be taken into account during clay consolidation whereas the sorption isotherm and $\alpha$ may be considered as unchanged during the consolidation process.

This paper presents an experimental and numerical investigation of consolidation-induced solute transport. Diffusion and large strain consolidation tests were performed on composite specimens of kaolinite clay consisting of an upper uncontaminated layer and a lower layer contaminated with potassium bromide. Experimental measurements of effluent concentration, solute mass outflow, and final concentration profiles were obtained for a variety of initial, boundary, and loading conditions, including unload/reload. Numerical simulations were conducted using a computational model in which solute transport occurs by advection, dispersion, and sorption and is consistent with temporal and spatial variations of porosity and seepage velocity in the consolidating layer. Large strains were taken into account as well as variation of effective diffusion coefficient with porosity and nonlinear nonequilibrium sorption effects. The numerical simulations are in good to excellent agreement with the experimental measurements. Results indicate that, depending on conditions, diffusion and consolidation-induced advection can make important contributions to solute transport and mass outflow. Thus, both mechanisms should be considered for transport analyses involving soft contaminated clays undergoing large volume change. Results also indicate that nonequilibrium sorption effects were not significant for the materials and test conditions used in this study.

Based on water chemistry data, a rockfill dam in South Korea was investigated to identify major parts of substantial dam water leakage. Sinkholes in the dam crest and abnormal water leakage through the dam were observed since the first impounding of the dam. Previous investigations, including observational drillings, hydraulic tests, geophysical surveys, and tracer tests, revealed the development of a high permeability zone in the clay core and an abnormally high phreatic line zone in the left embankment. Sequential tracer tests provided further evidence of locations of preference and rapid water leakage pathways. In this study, we examined the usefulness of water chemistry data to confirm previous findings of water leakage. Factor analysis of the water chemistry data revealed major processes controlling the chemical composition, which include the wash-out and successive dissolution of fill materials in the dam interior, interaction between percolating water and dam core materials, and contamination of seeping water by grouting materials and previous tracer tests. Distinct spatial distribution of similar chemical groups was illustrated by cluster analysis and confirmed the location of major water leakage at the left abutment adjoining the spillway.

Spudcan foundations for offshore mobile drilling rigs are large saucer-shaped foundations that can penetrate several tens of meters into soft sediments. The penetration depth is typically predicted by considering a wished-in-place foundation at different depths and following traditional bearing capacity approaches to assess the depth at which the estimated capacity matches the applied loading. However, the geometry of the spudcan and its progressive mode of penetration lead to soil failure mechanisms that differ markedly from those relevant to onshore practice. This paper presents a new rational design approach for assessing spudcan penetration in single layer clays based on a study combining centrifuge model testing and large deformation finite-element (FE) analysis. The design approach takes account of the evolving failure mechanisms in the soil, which start with cavity formation and surface heave at shallow penetration, gradually transforming to backflow of soil over the spudcan. A detailed FE parametric study has explored the relevant range of normalized strength, strength nonhomogeneity, and spudcan base roughness, with results validated against centrifuge model test data. The penetration response curves are presented in terms of profiles of bearing capacity factors, forming nondimensional design charts along with simplified expressions for convenient use in practice. Comparisons with approaches suggested in the SNAME design code suggest an urgent need to update current practice.

Capacity-based design of structures limits the soil-structure interaction mechanism to the determination of the bearing capacity of a pile group. However, in many cases the criterion for the design of piles to resist lateral loads is not the ultimate lateral capacity but the deflection of the piles. Many procedures exist for estimating the response of single piles and pile groups under lateral loading, ranging from application of empirical relationships and simple closed-form solutions to sophisticated nonlinear numerical procedures. With the aim of investigating the effect of cracking, disregarded by most of the above-mentioned methods, a three-dimensional (3D) nonlinear analysis that accounts for cracking is presented. Response prediction correlates well with the experimental data from a full-scale pile load test. Interesting conclusions have also been drawn regarding the discretization of the computational domain and the combination of 3D numerical nonlinear analysis and the structural beam theory.

Tapered piles in comparison to cylindrical piles can be beneficial in terms of the load capacity. In this paper, estimation of the load capacity for tapered piles using cone penetration test (CPT) resistance was investigated. Fourteen calibration chamber load tests using different pile types and six CPTs were conducted under various soil conditions. From the calibration chamber test results, the total, base, and shaft load capacities were analyzed in terms of soil conditions and taper angle. To evaluate CPT-based load capacity of tapered piles, normalized base and shaft resistances were obtained from normalized unit load-settlement curves. Based on the normalized base and shaft resistances, design equations that can be used to evaluate the base and shaft resistances of tapered piles were proposed. The proposed method is valid for sands of medium to dense conditions, while it may result in unconservative predictions for loose sands. To check the accuracy of the proposed method, field load tests using both cylindrical and tapered piles were conducted and compared with the predictions using the proposed method. A simplified approach using an equivalent cylindrical pile was also investigated and compared.

The seismic sequence that occurred in October and November 2002 in the Molise region (Southern Italy) was characterized by two $M_w=5.7$ earthquakes within 24 h followed by one month long aftershocks series. The mainshocks caused substantial structural damage in the village of San Giuliano di Puglia. The damage distribution was highly non uniform. Heavy and widespread damage was observed to all buildings constructed in the recently developed part of the village, where subsoil conditions are characterized by a bowl-shaped basin filled with stiff clays, whereas in the historical center, built on an adjacent rock outcrop, many buildings showed no or light damage. Several accelerograms were recorded during the aftershocks sequence by a temporary network installed on two sites in the San Giuliano village, located on rock and soil, respectively. The geological, seismological, geotechnical, and structural relevant information of the earthquakes are presented in the first part of the paper. The second part of the paper investigates the possible role of site effects in the observed pattern of damage by one-dimensional (1D) and two-dimensional (2D) numerical site response analyses. First, the computed ground surface motions were compared to the aftershocks recordings. It was found that 1D analyses considerably underpredicted dynamic response while 2D modeling provided a better understanding of the amplification phenomena. Further, based on the calibration site response study performed with the aftershock records, the ground response simulation of October 31, 2002, mainshock was carried out. The results of 2D numerical analyses led to average ground surface motion characteristics consistent with the observed distribution of damage throughout the village.

A probabilistic analysis of a shallow circular tunnel driven by a pressurized shield in a frictional and/or cohesive soil is presented. Both the ultimate limit state (ULS) and serviceability limit state (SLS) are considered in the analysis. Two deterministic models based on numerical simulations are used. The first one computes the tunnel collapse pressure and the second one calculates the maximal settlement due to the applied face pressure. The response surface methodology is utilized for the assessment of the Hasofer-Lind reliability index for both limit states. Only the soil shear strength parameters are considered as random variables while studying the ULS. However, for the SLS, both the shear strength parameters and Young’s modulus of the soil are considered as random variables. For ULS, the assumption of uncorrelated variables was found conservative in comparison to the one of negatively correlated parameters. For both ULS and SLS, the assumption of nonnormal distribution for the random variables has almost no effect on the reliability index for the practical range of values of the applied pressure. Finally, it was found that the system reliability depends on both limit states. Notice however that the contribution of ULS to the system reliability was not significant. Thus, SLS can be used alone for the assessment of the tunnel reliability.

This paper studies the accuracy of the three-dimensional finite-element predictions of a displacement field induced by tunneling using new Austrian tunneling method (NATM) in stiff clays with high $K_0$ conditions. The studies are applied to the Heathrow express trial tunnel. Two different constitutive models are used to represent London Clay, namely a hypoplastic model for clays and the modified Cam-clay (MCC) model. Good quality laboratory data are used for parameter calibration and accurate field measurements are used to initialize $K_0$ and void ratio. The hypoplastic model gives better predictions than the MCC model with satisfactory estimate for the displacement magnitude and slightly overestimated width of the surface settlement trough. Parametric studies demonstrate the influence of variation of the predicted soil behavior in the very-small-strain to large-strain range and the influence of the time dependency of the shotcrete lining behavior.

Pseudostatic slope stability procedures are commonly used in engineering practice. However, the selection of the seismic coefficient employed in the analysis is often based on precedence without due consideration of the amount of seismic displacement that constitutes satisfactory performance for each particular project and without incorporating the vastly different seismic exposure for sites around the world. In this Note, a rational basis for selecting the seismic coefficient is presented. The proposed procedure requires that the engineer establishes the project-specific allowable level of seismic displacement. The seismic response characteristics of the slope are represented by the fundamental period of the potential sliding mass, and the site-dependent seismic demand is characterized by the 5% damped elastic design spectral acceleration at the degraded period of the potential sliding mass. The level of uncertainty in the estimates of the seismic demand and displacement can be handled through the use of different percentile estimates of these values. With the proposed equations, the engineer can properly incorporate the amount of seismic displacement judged to be allowable and the seismic hazard at the site in the selection of the seismic coefficient.

The capacity of a pile-soil system can be obtained by pushover analysis using the beam-on-nonlinear-Winkler foundation model. When the pile section enters the plastic state, the plastic hinge method is usually adopted to model its postyielding flexural behavior. The propagation of the plastic zone in a pile can be traced using a series of plastic hinges distributed along the pile that is embedded in the soil. This study proposes a proper model for arranging the distributed plastic hinges along the depths of the pile that are prone to yielding. A modified definition of plastic curvature is proposed herein in order to correctly simulate the nonlinear flexural behavior of a pile. Examples are presented to show the effectiveness of the proposed model.

The treatment of soils with cement is an attractive technique when the project requires improvement of the local soil for the construction of subgrades for rail tracks, as a support layer for shallow foundations and to prevent sand liquefaction. As reported by Consoli et al. in 2007, a unique dosage methodology has been established based on rational criteria where the voids/cement ratio plays a fundamental role in the assessment of the target unconfined compressive strength. The present study broadened the research carried out by Consoli et al. in 2007 through quantifying quantifies the influence of voids/cement ratio on the initial shear modulus $\left(G_0\right)$ and Mohr-Coulomb effective strength parameters $\left(c^{\prime },{\varphi }^{\prime }\right)$ of an artificially cemented sand. A number of unconfined compression and triaxial compression tests with bender elements measurements were carried out. It was shown that the void/cement ratio defined as the ratio between the volume of voids of the compacted mixture and the volume of cement is an appropriate parameter to assess both initial stiffness and effective strength of the sand-cement mixture studied.