The influence of stress path on the stress‐strain behavior of a Florida sand subject to repetitive moving wheel loads is investigated in the laboratory with conventional traixial equipment. A conventional resilient modulus test with only cyclic varying compressive loads and a moving wheel stress path involving both extension and compressive loads determined from an elastic solution were examined at different initial confining pressures. The tests showed that excursions of applied extension loading followed by compression loading resulted in anisotropic material behavior. This, in turn, influenced both the cyclic permanent strain (rutting) build‐up as well as the resilient (distortion) behavior of the material for initial confining pressures below approximately 40 psi (276 kPa). It was concluded that, in order to model both the distortion and rutting characteristics of a sand, improved constitutive relationships over the simple linear elastic theory are warranted. A bounding surface plasticity model was subsequently developed based upon both the plastic shear dilation and volumetric behavior of the sand. The predicted model response agreed both qualitatively and quantitatively for the tests investigated.

A state‐of‐the‐art nonlinear finite element algorithm for pile‐soil interaction has been developed and implemented in a modular computer program to study the effects of predrilled oversize holes and layered soils on the vertical load carrying capacity of piles in integral abutment bridges where thermally induced lateral movements occur. A 40‐ft (12‐m) H pile embedded in very stiff clay is used to study the effect of a predrilled oversize hole with or without sand fill. Numerical results showed that the vertical load carrying capacity of H piles is significantly affected by predrilling. The study ol layered soils included the case where compacted soil (50 or more blow counts in standard penetration tests) is used as a fill on top of different layers of natural soil. For such cases, the numerical results indicated that the type of failure is greatly influenced by depth of the compacted soil.

A method to analyze reliability of soil slopes using the response surface method is described. The soil slope is modelled and analyzed by a finite element code as in prevalent deterministic studies. The simulation which is usually expensive, is repeated a limited number of times to give point estimates of the response corresponding to uncertainties in the model parameters. A graduating function is then fit to these point estimates so that the response given by the finite element code can be reasonably approximated by the graduating function within the region of interest. The approximating function, called the response surface, is used to replace the code in subsequent repetitive computations required in a statistical reliability analysis. The procedure is applied to a sample problem in slope stability involving uncertain soil properties. It is shown that the slope stability statistics from the response surface is within 1–9% of the statistics based on a direct Monte Carlo simulation using the finite element code. The merits of such an approach to reliability analysis are examined.

The potential of saturated deposits of cohesionless soil in downtown San Francisco to experience initial liquefaction due to seismically induced pore pressure is evaluated. Initial liquefaction, or the zero effective stress state, is used as the index of liquefaction potential because it provides the best available index for damage due to seismically induced pore pressures. Liquefaction potential is evaluated by comparing the conditional probability of liquefaction, or liquefaction susceptibility, to the expected intensity of seismic loading, or liquefaction opportunity. The probabilistic evaluation is made using a liquefaction hazard model developed by Chameau. Assuming the water table to be at the ground surface, results indicate that while no liquefaction is expected anywhere for an intensity with an annual probability of exceedance of 0.05, only the most resistant deposits will survive an event with an annual probability of exceedance of 0.02. For an event with an annual probability of exceedance of 0.01, initial liquefaction is expected to occur within all saturated, cohesionless soil deposits in the downtown San Francisco area. It must be emphasized that in dense soil deposits the consequences of initial liquefaction may be minimal because of their limited shear strain potential.

Residual soils are products of chemical weathering and thus their characteristics are dependent upon environmental factors of climate, parent material, topography and drainage, and age. These conditions are optimized in the tropics where well‐drained regions produce reddish lateritic soils rich in iron and aluminum sesquioxides and kaolinitic clays. Conversely, poorly drained areas tend towards montmorillonitic expansive black clays. Andosols develop over volcanic ash and rock regions and are rich in allophane (amorphous silica) and metastable halloysite. The geological origins greatly affect the resulting engineering characteristics. Both lateritic soils and andosols are susceptible to property changes upon drying, and exhibit compaction and strength properties not indicative of their classification limits. Both soils have been used successfully in earth dam construction, but attention must be given to seepage control through the weathered rock. Conversely, black soils are unpopular for embankments. Lateritic soils respond to cement stabilization and, in some cases, lime stabilization. Andosols should also respond to lime treatment and cement treatments if proper mixing can be achieved. Black expansive residual soils respond to lime treatment by demonstrating strength gains and decreased expansiveness. Rainfall induced landslides are typical of residual soil deposits.

A new method of indexing the grain and aggregate properties of sand using electrical parameters is described. Correlations are established between these parameters and relative density, $D_r\text{,}$ cyclic stress ratio, $\tau /{\sigma }_o^{\prime }\text{,}$ and $K_2{}_{\max }\text{.}$ An electrical probe, used to predict these parameters from in‐situ electrical measurements, is described. Evaluations are made of $D_r$ and $\tau /{\sigma }_o^{\prime }\text{,}$ which are compared with values measured independently from controlled laboratory tests. Reasonable agreement is found between predicted and measured values. The potential applicability of the electrical probe in the field is shown by evaluation of liquefaction and nonliquefaction at sites affected by the 1906 San Francisco, Niigata and Tangshan earthquakes.

A previously developed technique, used mainly in seismology and geophysics, has been adopted to determine the shear modulus from soil displacements measured during wave tank experiments. This iterative inversion technique allowed the determination of the shear modulus and Coulomb specific loss (damping) with depth within a bentonite clay profile. It also allowed the observation of both the modulus reduction and damping curves for large shear strain amplitudes (up to 7%) for the first time. There was fairly good agreement between the modulus reduction curve determined by the inversion method and values found in the literature. Refinements to the inversion technique are suggested to improve determination of shear modulus. An alternate procedure to provide fairly reliable and quick estimates of the shear modulus is also presented.

A series of laboratory vane shear tests was performed on sedimented samples of kaolinite and slaked Pierre shale to determine the effects of shear rate and vane dimensions on measured strength. Results of the tests were expressed in terms of shear strength on a vertical plane and were plotted versus average shear rate. For kaolinite samples vane shear strength decreased as shear rate increased because of partial drainage occurring at slower shear rates. However, for the highly plastic Pierre shale vane, shear strength increased as shear rate increased because of undrained creep occurring at slower rates of shear. Test results for kaolinite and Pierre shale are compared to published information from other studies.

Volume measuring systems for triaxial tests are susceptible to changes in ambient temperature, and the volume change due to temperature can represent 10% or more of the measured value, depending on the magnitude of the total volume change. The automatic volume change logging system showed a clear dependence on temperature, a dependence related to the thermal expansion of the water in the system, the apparatus itself and any entrapped air. By using regression analysis on data accumulated over several tests, a simple temperature correction formula for volume change is derived.