Journal of Geotechnical and Geoenvironmental Engineering

Buried water reservoirs are increasingly being built to replace open aboveground municipal water supply reservoirs in urban areas to enhance water quality and utilize their surface footprint for other purposes such as public parks or placement of solar arrays. Many of these lifeline structures are in seismically active regions and, as such, need to be designed to remain operational after severe earthquake shaking. However, evaluating their seismic response is challenging and involves accounting for the interaction of the structure with the stored fluid and the retained soil; in other words, accounting for fluid–structure–soil interaction (FSSI). This paper presents a combined experimental–numerical study on the seismic behavior of buried water reservoirs while considering FSSI. Two series of centrifuge model tests were performed at different reservoir orientations to investigate one-dimensional (1D) and two-dimensional (2D) motion effects under full, half-full, and empty reservoir conditions. Corresponding numerical models were developed whereby the structure and the soil were represented by continuum Lagrangian finite elements, while the fluid was modeled via Arbitrary Lagrangian Eulerian formulation. Soil–structure and fluid–structure interface parameters were calibrated using the experimental measurements. The simulations successfully captured the measured reservoir responses in terms of accelerations, bending moment increments, and water pressures. The study found that the common assumption of plane strain is not applicable for reservoirs because their behavior was found to be truly three-dimensional (3D) whereby stresses accumulated at the corners. Furthermore, the full reservoir resulted in the highest seismic demands in the reservoir walls and roof while the empty reservoir yielded the highest base slippage. The study demonstrates that the complex reservoir seismic response is best captured by carrying out a 3D FSSI numerical simulation.

The particle shape plays an important role in the macroscopic response of granular materials. Based on discrete element method (DEM) simulation results of triaxial tests on samples with different particle shapes, this study focuses on the microcharacteristics of contact network and anisotropy in granular media during shearing by a quad-partition method. Numerical results show that the interlocking is enhanced, and stronger contact force will be generated along the loading direction as the deviation of particle shape from a sphere, causing the increase of shear strength. Sliding mainly occurs in weak contacts, and the microscopic dissipative behavior in weak contact sliding plays a key role in the transformation from contraction to dilation. The samples with different particle shapes all reach the maximum proportion of sliding contacts at the phase transformation state that is the transition points of volumetric change. The anisotropy of samples increases as the deviation of particle shape from a sphere, and the contribution of the normal contact force anisotropy to the shear strength is the largest one. With the increase of the particle axial ratio, the mean coordination number and geometrical anisotropy increase due to the oriented alignment occurring at the critical state. Compared with superball particles, the ellipsoid particles have lower compressibility with a smaller axial strain and volumetric strain at the peak state.

A two-phase liquid/gas flow numerical model has been used to investigate the presence of elevated pore water pressures in a 20-m-deep municipal solid wastes (MSW) landfill underlain by a fully drained leachate collection layer. Monitoring of leachate levels in the landfill using piezometers located at different discrete levels found water table type conditions to within 10 m of the surface and established a downward hydraulic gradient at an infiltration rate of $\sim 400\mathrm{mm}/\mathrm{year}$. Short-duration falling-head piezometer tests indicated landfill hydraulic conductivities ($K_h$) between $1\times {10}^{-4}$ and $1\times {10}^{-5}\mathrm{m}/\mathrm{s}$, with a general reduction in $K$ with depth. Several different hypotheses to explain the high pore water pressures in the landfill were investigated using a one-dimensional configuration of the landfill degradation and transport model LDAT. It was assumed that the unsaturated properties of the landfilled wastes can be bounded by two sets of van Genuchten parameters. Comparing the values of $K_v$ required to create a match between observed and modeled leachate heads with the measured $K_h$ values at the site led to a tentative conclusion that landfill-scale anisotropy could be as high as $\sim 1:\mathrm{1,000}$. The introduction of a distributed landfill gas (LFG) source term into LDAT at a rate of $0.61{\mathrm{m}}_{\mathrm{LFG}}^3{t}_w^{-1}{\mathrm{year}}^{-1}$, similar to the gassing rate at the site, increased the adopted saturated permeability relationship in LDAT by a factor of between $\sim 3$ and $\sim 7.5$ compared with a no-gassing scenario. Introducing even moderate gas generation rates ($5.6{\mathrm{m}}_{\mathrm{LFG}}^3{\mathrm{t}}_w^{-1}{\mathrm{year}}^{-1}$) into models simulating low infiltration rates of $50\mathrm{mm}/\mathrm{year}$ can result in a significant depth of waste where pore water pressures are more than 1 kPa (10-cm water head). This results in apparent below-water-table type conditions because water will enter piezometers installed into such wastes, even though the gassing reduces the degree of saturation to below one.

Abstract

The effect of high overburden pressure above 1 atm on sand liquefaction potential is typically evaluated based on cyclic undrained testing, with the overburden pressure correction factor, $K_{\sigma }<1.0$ and $K_{\sigma }$ decreasing as the pressure increases. Recent centrifuge experiments of a prototype 5 m-thick clean sand layer having a permeability $1.2\times {10}^{-4}\mathrm{m}/\mathrm{s}$, with free drainage at the top and subjected to 1 and 6 atm overburden pressures, show that a high overburden pressure may increase partial drainage. As a result, the measured field overburden pressure factor, $(K_{\sigma }{)}_{\mathrm{field}}$ was estimated to be >1.0 instead of $<1.0$ in these centrifuge tests. A parametric study is presented here that extends the centrifuge results for a relative density, $D_{\mathrm{r}}=45\%$ and free top drainage, utilizing a high-fidelity, calibrated numerical model (P2Psand in FLAC 3D). A stepped acceleration base input is used that ensures a uniform cyclic shear stress amplitude at the elevation of maximum pore pressure ratio. The main parameters varied in the numerical study are overburden pressure, ${\sigma }_{\mathrm{v}0}^{\prime }$ (1-12 atm); sand permeability, $k$ (${10}^{-6}$ to ${10}^{-3}\mathrm{m}/\mathrm{s}$); and sand layer thickness, $H$ (2-10 m). A new drainage factor, $K_{\mathrm{dr}}\ge 1.0$, is proposed to separate the usual undrained $K_{\sigma }$ from the effect of partial drainage. The recommendation is to evaluate the overburden pressure factor to be used in liquefaction charts, $(K_{\sigma }{)}_{\mathrm{field}}$, as the product of two factors, $(K_{\sigma }{)}_{\mathrm{field}}=K_{\sigma }\times K_{\mathrm{dr}}$. The study shows that for constant ${\sigma }_{\mathrm{v}0}^{\prime }=6$ atm, $K_{\mathrm{dr}}\approx 1.4–1.7$ when $k={10}^{-4}$ to ${10}^{-5}\mathrm{m}/\mathrm{s}$, even for a very thick sand layer of $H=10\mathrm{m}$. Still for a constant ${\sigma }_{\mathrm{v}0}^{\prime }=6$ atm, $K_{\mathrm{dr}}$ decreases considerably to values close to 1.0 for a low $k\approx {10}^{-5}\mathrm{m}/\mathrm{s}$, when the layer thickness is $H=7$ or 8m or greater. And for constant $k\approx {10}^{-4}\mathrm{m}/\mathrm{s}$ and ${\sigma }_{\mathrm{v}0}^{\prime }=12$ atm, $K_{\mathrm{dr}}\approx 1.4$ if $H=5–10\mathrm{m}$, increasing to $K_{\mathrm{dr}}\approx 1.8$ if the layer thickness decreases to $H=2\mathrm{m}$. The parametric study provides insight on when the current State-of-Practice of using only undrained $K_{\sigma }$ becomes too conservative and should be supplemented with consideration of partial drainage through use of factor $K_{\mathrm{dr}}$. Ready-to-use charts are provided to evaluate $K_{\mathrm{dr}}$ and $(K_{\sigma }{)}_{\mathrm{field}}$ for a $D_{\mathrm{r}}=45\%$ sand layer with free top drainage.

Practical Applications

This research challenges conventional notions of sand liquefaction risk assessment. Traditionally, higher overburden pressures were thought to increase this risk, as indicated by the factor $K_{\sigma }<1.0$. However, recent findings, supported by centrifuge experiments and advanced numerical modeling, reveal a more nuanced reality. Under specific field conditions, especially with elevated overburden pressures, partial drainage in clean sand deposits can enhance the layer stability, resulting in $(K_{\sigma }{)}_{\mathrm{field}}$ values greater than 1.0. This means that high overburden pressures may not necessarily lead to increased liquefaction vulnerability, contrary to previous assumptions. To better guide practical applications, a novel drainage factor, $K_{\mathrm{dr}}$, is introduced. When combined with $K_{\sigma }$, it provides a more accurate assessment of liquefaction potential. For engineers and practitioners, this means considering both $K_{\sigma }$ and $K_{\mathrm{dr}}$ when evaluating the liquefaction potential of loose sand layers with at least one drainage boundary. This study supplies valuable insights into when the current practice of relying solely on undrained $K_{\sigma }<1.0$ may be overly conservative. It offers ready-to-use charts to assess $K_{\mathrm{dr}}$ and $(K_{\sigma }{)}_{\mathrm{field}}$, empowering practitioners to make more informed decisions regarding liquefaction risk. It must be noted that an actual field case history is expected to require some engineering judgment due to soil stratigraphy and the usual spatial variability associated with field deposits.

Assessment of earthquake-induced liquefaction is an important topic in geotechnical engineering due to the significant potential for damage to infrastructure. Some of the most significant infrastructure damage occurs due to differential settlement of the ground, including due to liquefaction. Postliquefaction deformations commonly are assessed using one-dimensional empirical models, which inherently assume laterally homogeneous soil layers. Numerical models offer the potential to examine the effects of ground motion variability and spatially variable soil properties on liquefaction-induced deformations. This study explored the postliquefaction reconsolidation settlement for a site in Hollywood, South Carolina, which was characterized using a three-dimensional (3D) geostatistical model and simulated using the numerical platform FLAC and constitutive model PM4Sand. The effects of ground motion characteristics on mean and maximum differential settlements were investigated. The physical mechanisms associated with postliquefaction responses such as excess pore pressures, shear strains, and volumetric strains also were examined. The efficacy of uniform models assuming representative percentile soil properties to represent the stochastic mean settlement was investigated. The inherent inability of uniform models to capture differential settlements and therefore the need for using stochastic models is discussed.

Abstract

The widespread use of rubble-mound (RM) breakwaters along coasts across the world highlights the importance of understanding their behavior during natural disasters such as tsunamis. The failure of these breakwaters during tsunamis can have far-reaching consequences, potentially causing damage to coastal infrastructure and loss of life. Many breakwaters failed during past tsunamis. Despite this, studies on the behavior of RM breakwaters during tsunamis are minimal. The present study thus attempts to elucidate the behavior of RM breakwater subjected to a tsunami. Furthermore, efforts were made to develop effective countermeasures that can safeguard the breakwater against tsunamis. To the end, a novel technique of using geogrids for reinforcing the RM is proposed. This study could be a pioneering application of geogrids as reinforcing elements in RM breakwaters to mitigate damages from tsunamis. Geogrid layers are provided on both the seaside and harborside to mitigate tsunami-induced damage to the breakwater. In addition, a crown wall (with shear keys) is also introduced to prevent the scouring of the crest and sheet piles from preventing excess seepage through the seabed. Physical model tests, analytical studies and numerical simulations were carried out to assess the performance of the proposed countermeasures by comparing it with the behavior of conventional RM breakwater during the tsunami. The tsunamis can overflow the breakwater, potentially exceeding its design limits. Hence, provision was made in the study for overflow, where the breakwater may overflow by the tsunami. It was observed that excess seepage through the body of the breakwater and the scouring of the crest were significant factors that led to the failure of RM breakwaters under tsunami overflow. A novel reinforced model was proposed to address these issues. This model effectively withstood tsunami-induced damages without significant deformations, demonstrating its potential as a reliable solution.

Practical Applications

In the past, tsunamis have caused significant damage to several rubble-mound breakwaters, leading to devastating impacts on coastal communities. The 2004 Indian Ocean tsunami and the 2011 Great East Japan tsunami are two well-known examples of the destructive power of tsunamis. A significant number of breakwaters across the globe are of the rubble-mound type, and finding a practical solution to make them resilient against tsunamis could protect many coasts against future tsunamis. This study proposes novel reinforcing techniques to make the rubble-mound breakwater tsunami-resilient. These techniques include adding geogrid layers on either side, providing a crown wall (with shear key) at the top, and inserting two sheet piles into the seabed at either end of the rubble-mound breakwater. Perhaps, this is the first-time geogrids were used on rubble-mound breakwaters to mitigate tsunami-induced damages. The reinforcing elements were chosen for their practicality in the real world. These elements can be practically applied to make existing and future breakwaters tsunami resilient. The study found that the proposed reinforced model, particularly Reinforced model III, remained intact without significant deformation even under a 15-m high tsunami.

Rigorous modeling of the stress-induced anisotropy of soils with different stress histories and loading conditions typically requires advanced constitutive models. However, calibration of state-of-the-art constitutive models can be expensive due to a large number of parameters and can encounter convergence issues when implemented in finite element codes. To circumvent these limitations, this study combines the well-known modified Cam-Clay (MCC) model with a machine learning-based multifidelity training framework, which is distinctive compared to current modeling approaches. A ‘low-fidelity’ neural network is first trained on synthetic data generated by the MCC model to ‘learn’ the model’s interpretations of critical state soil mechanics. A ‘high-fidelity’ neural network is subsequently trained using limited experimental data to fine-tune predictions of soil behavior. The proposed framework is applied to the prediction of stress-induced anisotropy of lower Cromer till (LCT) clay. The results show that the mechanical behavior of LCT under drained and undrained triaxial compression/extension with different consolidation histories can be accurately predicted by the model. The model is also shown to be insensitive to the exact composition of the synthetic data set, specifically, the base constitutive model and parameter set used. It also shows an ability to generalize unseen data outside of the calibration space due to the underpinning soil mechanics training. Finally, explicit consideration of prediction uncertainty increases the interpretability and reliability of the proposed model toward increasing the likelihood of industry take-up.

This paper presents the results of a study to determine the drained strength parameters of overconsolidated clays at low effective stresses. Overconsolidated clays experience dilation when sheared under undrained conditions in conventional triaxial compression (TC) and direct simple shear (DSS) tests. Moreover, the effective stresses at failure can be significantly larger than the preshear stresses. Consequently, interpretation of the results of conventional triaxial and DSS tests to define the failure envelope at low effective stresses involves considerable extrapolation of the data. Therefore, obtaining reliable interpretations of the drained strength parameters at low effective stresses (${\sigma }_{\mathrm{v}}^{\prime }<100\mathrm{kPa}$) is difficult. In this study, a new approach in performing TC and DSS tests, which involves shearing anisotropically consolidated specimens by increasing the pore water pressure at a constant shear stress, is presented. This procedure enables a reliable interpretation of the drained strength parameters at effective stresses as low as 10 kPa. This paper presents the conceptual test methodology, typical results of TC and DSS tests, and statistical summaries of the drained strength parameters of inorganic clays determined from numerous tests performed in this study. The results of the tests performed in this manner show consistent mobilization of significant effective cohesion, ${\mathrm{c}}^{\prime }$, which is related to the preconsolidation stress, ${\sigma }_{\mathrm{p}}^{\prime }$. Effective cohesion values normalized with respect to the preconsolidation stress (${\mathrm{c}}^{\prime }/{\sigma }_{\mathrm{p}}^{\prime }$) generally range from 0.3% to 6.1% for lean clays and from 1.7% to 7.1% for inorganic fat clays. The mean ${\mathrm{c}}^{\prime }/{\sigma }_{\mathrm{p}}^{\prime }$ values determined from triaxial tests were 2.5% for lean clays and 3.4% for fat clays. Similarly, the mean values from the DSS tests were 2.8% for lean clays and 3.5% for fat clays. The effective friction angles for lean and fat inorganic clays ranged between 23° and 37°, with mean values of 31° and 30° from the triaxial and DSS tests, respectively.

When drying from a fully saturated state, soil transits from saturation to unsaturation, activating the effects of capillarity, adsorption, and osmosis. Traditionally, these effects have been elusive in modeling soil shrinkage behavior. This paper addresses this gap of knowledge by reexamining the underlying mechanisms for the drying-induced shrinkage of soil. To this end, the concept of intergranular stress is introduced to lump all the effects of osmosis, capillarity, and adsorption into a unique effective stress tensor, and a simple constitutive model is developed based on the modified Cam-Clay (MCC) model. The proposed model inherits the simplicity of the MCC model while ensuring a smooth transition from saturation to unsaturation. It is shown that the drying-induced intergranular stress includes two components, accounting for capillary and adsorptive effects, respectively, intertwined with the osmotic effect. During a drying process, the adsorptive component of intergranular stress constantly increases and surpasses the capillary component at low water content, while the capillary component increases first and then decreases at the dry end. It is revealed that, for active soils, the effect of adsorption remains significant through the whole drying process, in contrast to nonactive soil where the effect of adsorption is significant only at low water content. As water content decreases, the deformation of soil is first elastic, then elastoplastic, and finally elastic. The elastic compression at the dry end can be attributed to the evaporation of the tightly adsorbed pore water. Comparison of simulations with experimental data shows that the proposed model captures very well the main features of the drying-induced shrinkage behavior of soil, implying that the intergranular stress tensor can be effectively used to address the hydro-chemo-mechanical behavior of soil under complex loading conditions.

As a common natural phenomenon in clayey soil, drought-induced desiccation cracks impose danger to soil structures by degrading soil mechanical strength and hydraulic properties. In this study, through a series of field tests, we evaluated the effect of microbially induced calcite precipitation (MICP) technology on the suppression of desiccation cracks in field clayey soil and examined the impacts of MICP treatment methods (two-phase and one-phase methods) and cementation solution (CS) concentrations (0.5 and 1.0 M) on soil desiccation cracking behaviors and soil microstructure variation. The results revealed that MICP is capable of suppressing desiccation cracks in field soil. The surface crack ratio, average crack width, total crack length, and average crack depth decreased with the increase of treatment cycles and CS concentration. The two-phase MICP method performed better on soil crack suppression than the one-phase MICP method. Further ${\mathrm{CaCO}}_3$ distribution and soil microstructure characterizations indicated that the ${\mathrm{CaCO}}_3$ crystals generated by the MICP process contributed dominantly to the soil crack suppression by filling crack space, bonding soil particles, and reducing soil water evaporation. This study demonstrates that MICP is expected to serve as an eco-friendly and sustainable technology for soil desiccation crack suppression at the field scale.

This study conducted field tests to investigate the thermomechanical response of an energy wall under practical intermittent and experimental monotone operation modes. Sensors and thermometers were utilized to measure inlet/outlet water temperature, air temperature, wall temperature, and strain. The heat-exchange performance and three-dimensional thermomechanical behavior of the energy wall were closely monitored and analyzed. The test results revealed that the intermittent heating operation demonstrated superior heat-exchange performance compared to the monotone heating operation, resulting in a reduced magnitude of wall temperature change. Moreover, the study highlighted the influence of physical and thermal boundary conditions on thermally induced stress within the wall. The response of the wall to thermal loading was found to be influenced by various factors, including location, temperature change, and the heating history, particularly in the intermittent operation.

The evaluation of erosion is often conducted by dedicated computer software, such as Hydrologic Engineering Center’s River Analysis System (HEC-RAS), which considers several input parameters such as the flow conditions, river geometry, and soil properties. It considers the soil condition primarily based on the mean particle size (${\mathrm{D}}_{50}$). Research showed that grain size mainly controls the erosion characteristics of coarse particles, while additional factors such as the particle shape, size, and interparticle microelectrical forces control the erosion characteristics of fine particles, resulting in higher erosion resistance as the particle size becomes smaller. The borderline between these different behaviors is approximately at the mean particle size of 0.2 mm. Therefore, it needs a careful approach in evaluating the erosion depth of fine riverbed materials based on HEC-RAS and similar software. This study conducted erosion tests on four different riverbed soils in Nebraska using the Mini-JET. Then, it correlated the mean grain size to the critical shear stress and developed the logic of the “equivalent sand particle diameter method for fine particles” to rationally convert the particle diameter of fine particles to that of coarse ones with the same critical shear stress. HEC-RAS was then used with the equivalent sand particle diameter concept. It found that this method predicted the erosion depth for fine soils rationally well, opening the possibility of easy assessment for the erosion of fine-grained riverbed soils.