Journal of Geotechnical and Geoenvironmental Engineering

Two large-scale shake table experiments were conducted to study the seismic pile group–bridge soil system failure mechanisms and examine the role of soil crust in lateral spreading caused by liquefaction. Pile group–bridge systems were supported by two different ground profiles, inclined liquefiable soils with and without soil crusts. The test results are discussed in terms of soil acceleration, pore pressure ratio, and displacement response. In addition, the pile seismic failure mechanisms are depicted according to the acquired date, and the effects of kinematic and inertial interaction on the curvature of pile and pier are evaluated. It was found that during weak earthquakes, the crust did not have an apparent influence on the system response. However, during strong earthquakes, the soil bed without crust experienced larger lateral permanent displacement because the shallow soil showed dilatant response and triggered spikes in acceleration. Meanwhile, the soil bed with crust restrained the lateral bridge displacement. In addition, the lateral spreading caused by liquefaction transferred the damaged position of the pile group–bridge system from the pier bottom to the pile at the bottom of liquefied soil, while the crust shifted the damaged position from the bottom of the liquefiable soil to the top of pile. The results also revealed that the crust weakened the kinematic interaction on curvature at the pile head but enhanced the inertia effect during the strong earthquake, while the opposite was true for ground without crust; the kinematic interaction was stronger, and the inertia interaction was diminished.

The strength of partially saturated sand is critical for addressing engineering challenges such as beach trafficability or coastal erosion. Mapping strength remotely would allow for surficial geotechnical characterization despite reduced site access. A framework for estimating and mapping the bearing capacity of sandy beach soils from remotely sensed moisture contents and standard sand characteristics is presented. Toward this goal, the variability of relevant sediment parameters ($d_{60}$, $d_{10}$, dry unit weights, porosities, moisture contents, and angles of repose) were documented for the quartz sand beach located near the US Army Corps of Engineers Field Research Facility in Duck, North Carolina, and were assumed to follow normal distributions. Concurrently, in situ bearing capacity measurements were recorded using a portable free fall penetrometer (PFFP). Measured values were compared to bearing capacities derived using a model for partially saturated sands, resulting in a correlation coefficient, $\rho$, of 0.83 for stations where sediment samples were collected simultaneously. When sediment data was unknown, standard characteristics for medium sand were used to estimate the bearing capacity, yielding $\rho =0.71$ when compared to PFFP deployments. For stations where the estimated and measured bearing capacities did not match (residuals $>10\mathrm{kPa}$), impacts of upward-directed hydraulic gradients, rapid changes in moisture content, and unit weight estimates are discussed. Finally, the proposed framework is applied to spatially estimate bearing capacities using moisture contents derived from satellite-based data and medium sand characteristics to suggest the safest pathways for a test vehicle. Moisture contents were derived from an optical WorldView-4 image and an image from the X-band synthetic aperture radar Cosmo-SkyMed 2 satellite. The spatial estimates of moisture content and bearing capacity generally follow the expected trends for both images. Shortcomings of the framework for predicting the safest path are discussed, and an alternative analysis using moisture contents is presented.

A broad range of coarse-grained soils that vary in gradation uniformities, maximum particle sizes, and absolute densities are prone to liquefaction during earthquake shaking. However, clean, poorly graded sands form much of the liquefaction case-history database, with these soils often serving as the basis for analysis procedures. This paper presents a centrifuge testing program studying if the liquefaction triggering and deformation performance of embankment systems constructed with poorly graded sands universally applies to well-graded soils. Two soils were used in this study, named 100A and 25ABCD, had variations in maximum particle sizes, coefficients of uniformity (${\mathrm{C}}_{\mathrm{u}}$), and void ratio extremes. Dense arrays of in-situ porewater pressure transducers and accelerometers indicate that liquefaction was triggered at near unity in the different soils. The 25ABCD soil, with its larger ${\mathrm{C}}_{\mathrm{u}}$ and lower void ratio indices, exhibited stronger dilative tendencies, better preservation of the long period energy of the input motion, and more rapid dissipation of excess porewater pressures. The 25ABCD embankments had lower overall levels of strain at initial liquefaction triggering and accumulated less strain during cyclic mobility. While the two soils were pluviated to the same relative density and subjected to the same level of shaking, the slope surface displacements in the 25ABCD embankments were 60%–70% less than the displacements measured in the comparable 100A embankments. These results support the hypothesis that liquefaction and deformation behaviors depend on soil gradation, and findings from this experimental program can be leveraged for more accurate performance predictions of levees and earthen dams.

As a common vibration source, impact pile driving can adversely affect the surrounding environment. Predicting ground-borne vibration intensities and evaluating vibration impact should typically be performed before the start of pile-driving work. However, existing prediction methods based on empirical formulas are often too crude for the reliable analysis of ground-borne vibrations because they cannot reflect the inherent features of dynamic pile–soil interaction or lack a rigorous analytical basis. Furthermore, prediction methods based on sophisticated finite-element modeling and field experiments are too time-consuming to meet practical engineering needs. Given this research gap, this study proposed an efficient and rigorous prediction approach for ground-borne vibrations induced by impact pile driving, wherein hammer-driven closed-end circular concrete piles penetrating into homogenous soil ground were investigated. A simplified pile-driving analysis was developed to compute ground-radiated energy at the pile–soil interface. Wave attenuation equations were applied to predict vibration intensity variation with distance by considering the propagation nature of various types of waves. Finite-element simulations and field measurements verified the accuracy of the proposed prediction approach.

The development of underground spaces inevitably poses significant risks to nearby infrastructure due to construction-induced ground displacements. While our understanding of tunnel-induced ground movements is now relatively mature, there is a distinct lack of literature on large-diameter open caisson shafts. This paper fills this gap by describing results from a small-scale laboratory study exploring soil deformation mechanisms during caisson construction in dry sand. Results from seven tests are analyzed to identify the influence of key caisson geometric properties as well as the effectiveness of external cofferdams in minimizing soil displacement. The results show that the primary mechanisms driving ground movements are a compressive ‘bearing’ front beneath the cutting face and a ‘frictional’ contribution above the cutting face. The normalized radial settlement profile is also shown to be insensitive to the normalized caisson embedment depth, and the settlement zone of influence extends up to 0.25 diameters below the caisson cutting edge. Furthermore, the presence of an external cofferdam is shown to be highly effective in reducing soil settlements. Quantitative analysis reveals a significant decrease in soil settlement with an increase in cofferdam depth from 0.25 to 0.5 of caisson depth, with good consistency between results for different soil elevations. In addition, larger cofferdam diameters provide maximum benefits in minimizing ground displacements.

Abstract

With increasing demand for sustainable underground infrastructure and pressure to reduce embodied carbon (EC), microtunneling (MT) has become an increasingly popular trenchless method of installing buried utility tunnels. Life-cycle analyses have shown that trenchless methods cause lower emissions than traditional open-cut construction. However, existing literature specifically considering MT is limited and fails to consider the impact of the entire construction process. In this paper, an approach for calculating the EC of MT is presented. The proposed approach is applied to three recent case histories in the United Kingdom through the development of a bespoke MT EC database in collaboration with industry partners. Total emissions across all three projects (870 m of pipeline) total 1,005 ${\mathrm{tCO}}_2\mathrm{e}$. Production of materials and components is shown to account for an average of 68.5% of EC across the three projects, with most of these emissions coming from the key structural materials, namely concrete and steel. Sensitivity analyses demonstrate that the source and production method of steel products have a significant impact on EC. Site activities also make a significant contribution, accounting for an average of 20.5% of total EC. Normalization of the results suggests that MT produces less EC than open-cut pipeline installation and highlighted how increasing drive lengths and reducing the number of shafts can significantly reduce EC. One of the case studies is then used as an example to quantify how the reduction of intermediate launch/reception shafts can reduce overall EC.

Practical Applications

In this paper, an approach for calculating the embodied carbon (EC) of microtunneling (MT) is presented, defining a scope that will enable fair comparison of future projects. The presented methodology also provides a useful reference for readers to find sources for EC factors. The proposed approach is applied to three recent case histories in the United Kingdom through the development of a bespoke MT EC database in collaboration with industry partners. The presented results provide insights into the EC of MT projects in five key areas: (1) the relative contribution of different materials and different construction phases to overall EC, (2) the sensitivity of EC to the method of steel production, (3) the influence of on-site emissions, (4) the comparison of MT to traditional open-cut pipeline construction, and (5) quantifying the environmental benefits of minimizing the number of shafts.

Accurate prediction of the load transfer response of a geosystem requires an understanding of the micromechanical, or local, behavior of individual particles at the soil–structure interface. The difficulty in predicting local interface behavior is inhibited by a multitude of influential factors, including material properties, state parameters, and confinement conditions. While significant efforts have been made to investigate how load transfer occurs within the soil adjacent to the structural interface surface during monotonic shearing, there is a relative lack of local behavior studies focused on cyclic shearing and the resulting postcyclic monotonic response. This study presents the results of an investigation quantifying the localized soil–structure interface evolution during cyclic shearing using a modified direct interface shear box. Parameters investigated include relative density (initial void ratio), particle angularity, particle hardness, surface roughness, normal stress, and normal stiffness. Particle image velocimetry, a local nonintrusive measurement technique, is used to measure the local 2D particle displacement and interparticle strain distributions within the test specimens. The work extends a previous study focused on monotonic shearing using the same material properties and confinement conditions and further illustrates differences observed in the localized shear strain and volumetric strain zones adjacent to the structural surface. Insights provided by the detailed presentation of local data are linked to the global response to explain differences in precyclic monotonic and postcyclic monotonic behavior. The insights can be used to further understand the micromechanical response of sand interfaces for use in geotechnical design.

Slope stability prediction is an important task in geotechnical engineering which can be achieved through physics-based or data-driven approaches. Physics-based approaches rely on geotechnical knowledge from soil mechanics, such as limit equilibrium analysis and shear strength theories, to evaluate the stability condition of slopes, and they are often limited to slope-specific analysis. Data-driven approaches predict slope stability conditions based on learned relationships between influencing factors and slope stability conditions from past observations of slope failures (i.e., case histories); they rely on big data which are difficult to obtain. This study examines three easy-to-implement and effective methods to integrate geotechnical engineering domain knowledge into data-driven models for slope stability prediction: hybrid knowledge-data model, knowledge-based model initiation, and knowledge-guided loss function. These models were benchmarked against pure data-driven models and domain knowledge–based models, including a physics-based solution chart and a physics-based empirical model. A compilation of slope stability case histories from the literature was used as the benchmark database, and five-fold cross-validation was employed to evaluate model performance. The model validation results demonstrated that machine learning models outperformed domain knowledge–based models in terms of several evaluation metrics. The three proposed methods were found to outperform both domain knowledge–based models and pure data-driven models. Additionally, the hybrid knowledge-data models and knowledge-guided loss function were found to reduce discrepancies in the predicted slope stability conditions compared with reported factor-of-safety values, leading to a better alignment with the underlying physics related to slope stability. This study provides an initial assessment of the value of coupling domain knowledge and data-driven methods in geotechnical engineering applications using slope stability prediction as an example.

This paper compiles a cone penetration test (CPT) database, named Global-CPT/3/1196. It contains three CPT parameters (cone tip resistance, sleeve friction, and porewater pressure) from 1,196 CPT soundings taken at 59 sites worldwide. The spatially variable cone tip resistance corrected for porewater pressure ($q_t$) soundings in the database are analyzed by a uniform procedure to characterize their autocovariance parameters consisting of a coefficient of variation, scale of fluctuation, smoothness, and hole effect. This is the first study that characterizes spatial variability based on a more complete set of three parameters: scale of fluctuation, smoothness, and hole effect. Previous studies were based on a single parameter (scale of fluctuation). The statistics of the autocovariance parameters and their inter-relationships are presented as a function of soil behavior type. These new statistics may be valuable for a future design founded on insufficient site investigation data: they can be used to obtain first-order estimates for the autocovariance parameters based on soil behavior type. Applications include data-driven site characterization and random finite-element analysis.

This paper for the first time, to the best of our knowledge, reports 10 case histories of jack-up rig installation in different locations of the Gulf of Mexico consisting of seabed soil stratigraphy with weak surface (soft) clay over strong (stiff) clay or sand underlying layers. For each case, site investigation data are analyzed, leading to the selection of design undrained shear strength profile for the clay layers and relative density for the sand layers. A statistical averaging method recommended in the InSafeJIP guidelines (adopted in ISO 19905-1) is used to obtain the best fit of the undrained shear strength profile in the clay layers. The spudcan shape and dimensions and load-penetration profiles are presented. Measured load-penetration profiles are compared with the calculated profiles using the squeezing methods presented in ISO 19905-1, Clarom, and recently published papers. Large deformation finite element analyses were conducted for some sites to provide supporting insight into the penetration resistance profiles and potentially mobilized squeezing mechanisms. The lessons learned are noted, which will provide valuable insights for practitioners for estimating the behavior of jack-up installations in seabed soil consisting of surface weak-over-underlying strong soil stratigraphy.

Soil arching that is generated and develops within the ground behind the retaining wall in a multistrutted excavation is a soil stress transfer process related to the relative movement between the well-retained soil mass and the inward-moved soil mass. The lateral earth pressure behind the wall is redistributed induced by this soil arching, which has not been deeply investigated. In this paper, the wall deflection, lateral soil movement, ground surface settlement, strut forces, total lateral earth pressure, and the water level outside the excavation were monitored in a multistrutted excavation in soft soil. The measured results indicated that the lateral earth pressure redistribution was associated with the arching effect. A lateral earth pressure model considering the soil arching effect is proposed, and was verified using measurements obtained from this study and the literature. It was found that the coefficients of the proposed model are linearly related to the ratio of the soil volume loss to the gain (RLG), which can be used to characterize the soil arching process development. This study improves the understanding of the lateral earth pressure redistribution induced by soil arching, and the proposed model can be used to calculate more-realistic lateral earth pressure.