Based on Maxwell's pioneering work, on the electrical conduction through heterogeneous media a fundamental approach utilizing the electrical properties of soils to characterize their grain (particle shape) and aggregate (porosity and fabric anisotropy) characteristics is presented. Correlations between mechanical properties relevant to the analysis of soil liquefaction and electrical indices defining the structure of a soil are given, and the influence of fabric anisotropy on mechanical behavior of soils is demonstrated. The nondestructive nature of the proposed method provides a means to predict in situ properties and thus overcome the difficulty of retrieving undisturbed samples. Level ground soil‐liquefaction analysis, which considers the pore pressure generation and dissipation during and after an earthquake, utilizing the method presented herein to obtain the necessary input properties, is presented in a companion paper (Arulanandan and Muraleetharan 1988).

Current methods for level ground soil‐liquefaction analysis involve the use of a stress ratio required to cause liquefaction obtained from laboratory testing of small element samples. Limited field evidence and considerable centrifuge model studies have shown that the soil liquefaction is a boundary value problem. There are numerous computer programs available that treat soil liquefaction as a boundary value problem. However, these analysis procedures do not take into consideration the properties of soil at low effective stresses, and most of the input parameters for these procedures have to be determined from test results of small “undisturbed” samples. These procedures have not been verified extensively. A method is presented for level ground soil‐liquefaction analysis, which is treated as a boundary value problem. This method uses a semiempirical, one‐dimensional, elastoplastic constitutive model and finite difference method to solve the governing differential equation for the prediction of pore pressure generation, dissipation, and settlement characteristics during and after dynamic excitation. A compressibility function incorporating effects at low effective stresses is included in this method, which is verified by centrifuge model tests. This procedure utilizes input properties representative of field conditions determined by a nondestructive electrical method. The proposed method for analyzing the generation and redistribution of pore pressure, therefore, provides a means for predicting liquefaction of level ground sites—a boundary value problem—based on the results of in situ testing.

A conceptual framework for understanding the effects of overconsolidation in reducing the compressibility of all types of soil is presented. A generally applicable method for estimating the settlement of footings on granular soils is proposed. The procedure uses a combination of dilatometer and cone‐penetration test results to identify the preconsolidation pressure, while soil moduli—either Young's modulus or constrained modulus, depending on the boundary conditions—are obtained from the dilatometer test results. Calibration chamber test results are used to adjust the dilatometer moduli for the effects of stress path and for disturbance due to insertion of the instrument. Detailed examples are given to illustrate the use of the method and to compare the results obtained with those calculated using currently accepted methods for estimating settlements.

A calculation model for estimating static behavior when a double sheet‐pile wall structure is subjected to horizontal force is proposed, and a comparison is made between the results of calculations and those of model tests to examine the validity of the calculation model. The calculation model is based on the theoretical solution obtained by assuming that the portion of the structure above ground level is a composite of sheet piles and sand fill and that the portion below ground level consists of sheet piles receiving horizontal resistance from the ground, taking into consideration the elasto‐plastic characteristics of the sand fill and the subsurface soils. The calculation results show good agreement with the test results and indicate that the calculation model can be effectively used for predicting the deformation of the wall, the bending moment of the sheet pile, and the tension of the tie‐rod, which are the most important items for structural design.

Jointing can have significant effects on the permeability, shear strength, compressibility, and slope stability of fine‐grained soils. Field measurement of jointing characteristics is an important step in the development of models that simulate fracture flow and directional strength. Preferred joint orientation, which can introduce anisotropy in permeability and strength, can be detected through independent and random measurement of joint orientation combined with graphical or statistical techniques. PATCH, a computer program based on the Poisson statistical test (Mahtab et al.. 1972), was used to evaluate the preferred orientation of joints measured in Wisconsin till units. The program was tested for its ability to reject spurious concentrations of joint poles and calculate mean joint orientation and dispersion. From an analysis of till joint data, we conclude that the use of the statistical computer program alone provided spurious results and required supplementation with other methods to properly determine joint set orientations.

A large testing program, consisting of triaxial compression, triaxial extension, and oedometer tests on 12 artificially prepared soils was performed. The soils exhibited a wide range in plasticity index, from 10–55. These tests were performed to study whether it is possible to determine parameters specifying a constitutive soil model simply by using the plasticity index. Linear correlations between the soil parameters for constitutive equations and the plasticity index were obtained with high values noted in the coefficients of correlation. Numerical illustrations to check the overall accuracy of the correlations were also developed. Direct application of these correlations to undisturbed natural clays will require further research.