Journal of Geotechnical and Geoenvironmental Engineering

Study of several field case histories has shown that the difference between two- and three-dimensional factors of safety is most pronounced in cases that involve a translational failure. Two- and three-dimensional slope stability analyses of field case histories and a parametric study of a typical slope geometry revealed that commercially available three-dimensional slope stability programs have a number of limitations with respect to (1) accounting for the shear resistance along the sides of the sliding mass; (2) modeling the stress-dependent nature of failure envelopes of the materials involved; and (3) considering the internal forces in the slide mass. These limitations can significantly affect the calculated factor of safety for a translational failure mode. A technique is presented to overcome some of these limitations and provide a better estimation of the three-dimensional factor of safety. Field case histories are presented to show the importance of using a three-dimensional analysis in back-calculating the mobilized shear strength of the materials involved in a slope failure and in the design of slopes with complicated topography, shear strength conditions, and/or pore-water pressures.

A 4.88-m-high retaining wall test facility was constructed to test tire shreds as retaining wall backfill. The front wall of the facility could be rotated outward away from the fill and was instrumented to measure the horizontal stress. Measurement of movement within the backfill and settlement of the backfill surface during wall rotation allowed estimation of the pattern of movement within the fill. Tests were conducted with tire shreds from three suppliers. The horizontal stress at a rotation about the base of 0.01H was about the same for tire shreds from the three suppliers. Moreover, horizontal stress at this rotation for tire shreds was about 35% less than the active stress expected for conventional granular backfill. Design parameters were developed using two procedures; the first used the coefficient of lateral earth pressure and the other was based on equivalent fluid pressure. The inclination of the sliding plane with respect to horizontal was estimated to range from 61° to 70° for the three types of shreds.

This paper describes an experimental program aimed at a close examination of the dynamic properties (shear modulus and damping ratio) of a normally consolidated dry sand as affected by the previous vibration. Tests were performed using a newly developed energy-injecting virtual-mass resonant column system. The new system allows the strain as well as the strain rate be kept constant concurrently throughout a test (hours of continuous vibration), irrespective of changes in soil properties. During the tests, modulus and damping values were monitored continuously without interrupting vibration. The test variables include strain amplitude, number of previbration cycles, confining pressure, and relative density. The results confirm that the damping ratio of dry sand is evidently dependent on the number of loading cycles when the shear strain exceeds a threshold, while the variations of shear modulus as number of loading cycles increases are relatively moderate. The test results also show that there exists an unequivocal signature of vibration history in the measured sand properties. This new finding may lead to certain applications.

This paper presents a method to analyze the problem of lateral extrusion of sandwiched weak soil in a slurry trench. This problem is modeled as weak material compressed between rough plates. Theoretical solutions are derived on the basis of the limiting plasticity theory. The arching effect due to the finite length of the trench is taken into account in the computation of vertical pressure being exerted upon the weak soil. The derived theoretical solutions are reduced to simple forms that can be used in practice. The stability of sandwiched weak soil is evaluated using a factor of stability that represents a measurement of the effectiveness of slurry pressure on preventing weak soil from lateral extrusion. This factor is defined as a ratio of slurry pressure to lateral extrusion pressure at the level of weak soil. Validity of the proposed method is examined using a field case study.

This paper presents a case study for a dredge disposal site called Spilmans Island, located along the Houston-Galveston Ship Channel, east of Houston. Initially classified as a sand bar in the San Jacinto River, Spilmans Island evolved in recent years with the construction of perimeter levees to contain the flow of materials produced from dredging operations. These levees were often constructed on soft dredged sediments, and as the levees were raised, occasionally slope failures occurred. The objectives of this paper are to illustrate the importance of reconstructing the history of a site as a basis for geotechnical analyses, and to demonstrate the significance of keeping accurate records of past investigations, construction activities, slope failures and subsequent remedial measures. The results of the geotechnical investigation described in this paper offer a clear example of how such data can be used to provide reliable predictions on the stability conditions of raised levees.

The development of localized strains (shear banding) in soils near peak stress levels and the subsequent continued deformation within shear bands during observed softening render important issues such as critical state difficult to investigate. The determination and overall validity of critical state behavior in sand is of considerable importance, as it provides the basis both for failure criteria/postfailure behavior of many constitutive models and for stability analysis. A series of drained plane-strain experiments on sand specimens with detailed local analysis was carried out to investigate the evolution of stress state and void ratio, as well as the uniqueness of critical state. Persistent shear bands form at the peak effective stress ratio; subsequent strain softening behavior occurs in concert with localized deformation with the shear band. A unique critical stress state was found to exist for a given confining stress; however, the test results indicate that there is no unique relationship between void ratio and confining stress at the critical state.

The objective of the present study was to evaluate the mechanical properties of tire chips and soil-tire chip mixtures relevant to geosynthetic-reinforced earthworks. Tests were conducted to evaluate shear strength and pull-out capacity with a woven geotextile and two geogrids. Soil-tire chip mixtures made with clean sand and sandy silt were tested. These properties were then used to assess the potential advantages of using soil-tire chip backfills for geosynthetic-reinforced retaining walls and embankments. The test results show that the geosynthetic pull-out force in tire chip and soil-tire chip backfills increases with displacement—i.e., no peak pull-out force is generally obtained, at least for displacements ≤100 mm. Pull-out interaction coefficients for tire chip backfills are typically greater than 1, whereas for soil-tire chip backfills they typically range between 0.2 and 0.7, even though the pull-out capacity for soil-tire chip backfills is generally similar to or greater than the pull-out capacity in a soil backfill. The higher strength, lower unit weight and good backfill-geosynthetic interaction obtained with soil-tire chip backfills can result in walls requiring less geosynthetic reinforcement than walls backfilled with soil. In addition, embankments can potentially be constructed with steeper slopes and a smaller volume of material when soil-tire chip fill is used, while providing greater resistance against lateral sliding and foundation settlement.

Approximately 100,000 m³ of atmospheric fluidized bed combustion (FBC) ash and stoker ash were used as structural fill in the construction of a large roadway embankment. The embankment is ∼200 m long and 10 m high, and it supports an extension of a street across a gravel quarry in West Lafayette, Ind. The use of coal combustion by-products on this project was motivated by the need to find cost-effective alternatives to the disposal of these materials in solid-waste landfills. Similar projects have been completed using the more common bottom ash and fly ash. However, little information is available related to the performance of FBC and stoker ashes in this application. An overview of the project and construction operations is described, and the results of geotechnical laboratory tests and field monitoring are presented. Instruments used in the monitoring of fill behavior include settlement plates, vertical and horizontal inclinometers, seismic cross-hole tests, and postconstruction standard penetration tests. The compacted FBC material exhibited significant swell stresses and strains and is still expanding 2 years after construction.

Cyclic lateral loading centrifuge tests were performed on a model of an embedded foundation to study its lateral response at various displacement levels up to ultimate lateral load capacity. The tests simulated a reinforced concrete square, 1.14 m × 1.14 m × 0.84 m high foundation embedded in a dry sand of 75% relative density. The results are relevant to shallow foundations and embedded pile caps. The tests' objective was to evaluate the relative contributions of the base, shearing sides, and active/passive sides of the foundation to the total lateral response as well as any possible interaction between these partial contributions. Lateral response parameters studied included secant stiffness and material damping ratio at various displacement amplitudes and ultimate lateral capacity of the foundation. The lateral stiffnesses measured are compared with available elastic solutions. Three-dimensional static nonlinear finite-element analyses are used to predict the measured lateral load-displacement curves of two of the tests, as well as to confirm analytically the relative contributions of base, shearing sides, and active/passive sides. Thus calibrated, the finite-element model is used to predict the lateral stiffness response of a broad range of embedded footing geometries in sand.

When there is an estimate of the probability of static failure for a set of slopes, based on historical records, maps, and subjective judgment, it is often necessary to estimate the additional hazard presented by earthquakes. Combining the procedures of reliability analysis with conventional recurrence relations for horizontal earthquake accelerations leads to a simple formulation that is easily implemented on a spreadsheet. For simple failure modes the dynamic factor of safety is a simple function of the static factor of safety, the amplified acceleration, the slope of the failure surface, and the friction angle. Application of the technique to the slopes of the Montes de Caneja in Puerto Rico indicates that the additional risk due to earthquakes is about one tenth of the static probability of failure, but other sets of input parameters can give probabilities of failure during earthquakes that are comparable to the static values.