Journal of Geotechnical and Geoenvironmental Engineering

Seven commercially available needle punched nonwoven geotextiles were evaluated in three outdoor locations and the results were contrasted against results from three different accelerated laboratory weathering test devices. The outdoor tests were conducted in Pennsylvania, South Carolina, and Texas, and were conducted for 12 months' duration. The laboratory tests were carbon arc, xenon arc, and UV-fluorescent methods conducted for 500 hours' duration. Responses of various nonwoven geotextiles varied from little strength and strain loss to significant degradation of the material. Outdoor versus accelerated results are contrasted and practical conclusions are given based on information gained from this investigation.

A theoretical approach is described for back analysis to determine shear strengths from slope failures. This approach systematically accounts for uncertainties encountered in practice, including uncertainties in empirical information, laboratory test results, field measurements, and the slope stability model used in the back analysis. The approach is applied to a slope failure in the Kettleman Hills landfill to demonstrate its use. Insight is gained from this case history concerning uncertainties in (1) calculated factors of safety from two-dimensional, limit equilibrium methods and (2) mobilized strengths in strain-softening materials.

Numerous investigations have been conducted in recent years to predict the load-bearing capacity of driven piles in cohesionless soils. The problem is extremely complex owing to the large number of factors that affect the behavior of piles. The available methods either oversimplify the nature of the problem or improperly consider the effect of certain governing factors. In this paper, a general regression neural network (GRNN) is used for predicting the capacity of driven piles in cohesionless soils. Predictions of the tip, shaft, and total pile capacities are made for piles with available corresponding measurements of such values. This is done using four different procedures as well as the GRNN. Comparisons of capacity component predictions (i.e., tip and shaft capacities) are only achieved for instrumented piles. The actual measurements of tip and shaft pile capacities were adjusted to account for residual stress using the wave equation analysis. For the remaining piles, the predicted total capacities are compared only with the measured value. The GRNN provides the best predictions for the pile capacity and its components. It is also shown that the GRNN is applicable for all types of conditions of driven piles in cohesionless soils.

Experimental evidence is produced using two soils that support the hypothesis that small-scale models frozen on a centrifuge for several hours can simulate correctly full-scale soil freezing response developing in the field over one or more winters. This is not the case in small 1g models. The conclusion is based on three observations: first, that model data produce self-consistent results from tests conducted at different scales; second, that model thin sections show general scaling of ice formation; and third, that heave developing in the field under broadly similar conditions shows broadly similar heave characteristics. A small series of tests to investigate effects on heave of various temperature regimes that might exist in the field also is included.

This paper presents results of 11 centrifuge model tests of liquefaction and earthquake-induced lateral spreading in sand using a laminar box. The centrifuge experiments simulate a horizontal or sloping, 10-m thick stratum of water-saturated homogeneous coarse sand of infinite lateral extent and relative density 40–45%, placed on an impervious rigid base. Such homogeneous, clean, coarse stratum represents some field liquefaction situations, but not others involving fine and/or silty sands or strong layering of strata of different permeabilities. The centrifuge tests presented provide significant general insight into mechanisms and parameters influencing the lateral spreading phenomenon. The layer was subjected to lateral base shaking with a prototype peak acceleration ranging from 0.17 to 0.46g, a frequency of 1–2 Hz, and a duration of approximately 22 cycles in all cases. The slope angle simulated in the field α_field ranged from 0 to 10°. Detailed discussions and comparisons of the 11 centrifuge test results are included. The observed effects of prototype slope angle α_field, input peak acceleration, a_max, and frequency f on the following measured parameters are summarized: excess pore pressure u, thickness of liquefied soil H_l, soil acceleration a, lateral displacement D_H, permanent shear strain γ, and surface settlement S. It is found that if H_l is considered, D_H depends only on α_field and H_l, with no influence of a_max and f. System identification analysis of the shear stress–strain response reveals significant dilative behavior of the liquefied soil at large cyclic downslope strains, characterized by large acceleration and strength increases and simultaneous pore pressure drops that arrest the downslope lateral displacement.

Examples of periodic loads applied on a soil layer appear in practice in the filling and emptying of a silo or a tank. This loading is not dynamic loading. This paper develops a solution to the consolidation equation with boundary conditions that are cyclic with time. Measurements of settlements of the Ca' Mello silo are compared with a case of sinusoidal loading. Good agreement is observed for the settlements. It appears that the pore pressure at the center of the soil layer is significantly larger than the amplitude of the applied load.

Triaxial compression tests were carried out to evaluate the effect of randomly distributed fiber reinforcement and cement inclusion on the response of a sandy soil to load. Cemented specimens were prepared with cement contents of 0% and 1% by weight of dry soil and cured for seven days. Fiber length was of 12.8 mm, in the contents of 0% and 3% by weight of dry soil-cement mixture. Test results indicated that the addition of cement to soil increases stiffness, brittleness, and peak strength. The fiber reinforcement increases both the peak and residual triaxial strength, decreases stiffness, and changes the cemented soil's brittle behavior to a more ductile one. The triaxial peak strength increase due to fiber inclusion is more effective for uncemented soil. However, the increase in residual strength is more efficacious when fiber is added to cemented soil. Peak strength envelopes indicate that the friction angle is increased from 35° to 46° as a result of fiber inclusion. The cohesion intercept is affected slightly by fiber addition, being basically a function of cementation.

The steady-state line relates the effective confining pressure (p′) with the void ratio (e) of a soil during steady-state deformation. The line is commonly plotted in an e–log p′ plane and is treated as a straight line in that plane. However, it has long been observed that the steady-state lines for cohesionless soils are in general curved in the e–log p′ plane, and treating them as straight lines is done only for simplicity. As an alternative, this note suggests the steady-state lines of cohesionless soils be plotted in the e–(p′/p_a)^α plane, where p_a is the atmospheric pressure serving as a reference pressure for normalization, and α is a number around 0.7 that can be fine tuned for optimization. It is shown in the note that the steady-state lines of sand are nearly straight in the e–(p′/p_a)^α plane, and the straightness is insensitive to the parameter α. This note presents the consideration behind the choice of the plane and demonstrates the straightening effects by examples.