Large pressure exerting on soil due to dynamic response of a liquid storage tank may cause excessive settlement of the supporting soil. Analysis for the dynamic response of such a system should consider the sloshing of liquid in the tank, the associated hydrodynamic pressure, and the coupling effects of liquid with tank and soil foundation. In this study, a simple numerical model is proposed to model the effect of liquid sloshing in tank. The governing equations are derived for the response of a ground‐supported cylindrical tank filled with liquid. It is found that liquid sloshing is dominated by the first mode of motion. Based on this fact, a closed‐form solution is obtained to study the pressures exerted on soil due to harmonic excitation. The dynamic response of the system is examined for various tank sizes, height‐radius ratios, and soil properties. It is found that resonance of such a system occurs due to sloshing of liquid under slow excitation. Under higher frequency excitation, such as that of earthquakes, the effect of liquid also plays an important role in the response of the liquid‐tank‐soil system.

Dam‐canyon interaction effects on the response of earth and rockfill dams built in narrow canyons are considered. An analytical closed‐form solution is used for a dam in a semicylindrical canyon, subjected to obliquely incident harmonic SH waves. The dam is idealized as linearly hysteretic elastic body deforming only in shear, while the canyon is modeled as an elastic half‐space. The model accounts in a rigorous way for the wave reflection and diffraction phenomena caused by the presence of the dam‐canyon system. It is shown that the motion amplification is considerably less for a dam in flexible canyon than for a dam in a rigid canyon and that the effects of radiation damping and asynchronous base motion may be very important. For oblique SH waves, the presence of antisymmetric terms in the solution may lead to response substantially higher than that induced by vertically propagating waves. Finally, as the canyon narrowness increases, the amplification at higher frequencies increases significantly for a rigid base, but it may decrease for a flexible base.

A simplified method of analysis for determining the permanent displacement of a clay core in a heterogeneous earth dam subjected to dynamic excitation is discussed. Centrifuge studies on three identical models were used to determine the deformation characteristics of earth dams subjected to earthquakes large enough to produce flow slides. These studies suggest that the failure mode is considerably influenced by the reduction in shear strength caused by the void‐ratio increase and the subsequent decrease in effective stress at the sand‐clay interface. Conventional sliding block analysis does not consider this phenomenon. A simplified method using the concepts of critical‐state soil mechanics is presented that may be used to estimate both the reduced shear strength and the redistributed void‐ratio at the interface. Values predicted using this method were found to agree well with those observed in the centrifuge experiments. By integrating these quantities into Newmark's sliding block analysis, a better estimate of the permanent displacement of an earth dam may be obtained.

Consolidated‐isotropically undrained, triaxial compression (CIUC), unconsolidated‐undrained, triaxial compression (UU), and unconfined compression (UC) test results were used to develop interrelationships for the undrained shear strengths $\left(s_u\right)$ obtained from these test types. Stress path evaluations were used to provide a general framework for interpretation of these interrelationships. The results show that the normalized undrained strengths $\left[s_u\right(\mathrm{UU}\mathrm{or}\mathrm{UC})/s_u(\mathrm{CIUC}\left)\right]$ are dependent on the overconsolidation ratio (OCR) or undrained strength ratio $(s_u/\sigma {¯}_{\mathrm{vo}})$. Softer clays have lower normalized undrained strength, while harder clays, including fissured clays, result in higher normalized undrained strength. Suggested correlations for these normalized undrained strengths or undrained strength ratios, as well as the linear regression data and confidence intervals corresponding to one standard deviation, are presented in this paper and show a consistent data population. Direct comparison of $s_u$ from UU and UC tests shows that the UU results are closer to the CIUC results than the UC results, which is consistent with general supposition.

Current methods for retaining‐wall analysis are based on the classical theoretical formulations of Rankine or Coulomb, with the assumption that sufficient lateral yield will occur to mobilize fully active conditions behind the wall. This leads to relatively simple equations for active pressure distributions. The effects of the deformation parameters of the backfill and foundation subsoil, construction sequence, wall flexibility and other complex initial conditions are not taken into consideration. In this paper the finite element method is employed to investigate the behavior of concrete cantilever retaining walls. The analyses reveal that the wall and soil movements are of considerable importance in predicting the lateral earth pressures. The results indicate that the lateral pressures acting on the wall stem are generally close to the active condition only in the top two‐thirds of the wall stem. Based on this study, a simplified design procedure is proposed for estimating the lateral earth pressures in cantilever retaining walls with loosely placed backfill.

Results are presented from a series of one‐way cyclic undrained triaxial tests on Ariake clay, a highly plastic, normally consolidated marine clay from Kyushu in southwest Japan. Considering first pore water pressures developed during cyclic loading, it is shown that the buildup of pore water pressure is a time‐dependent phenomenon, independent of frequency. The rate of pore pressure accumulation decreases exponentially with elapsed time of cyclic loading and at any given time it is a function of the applied cyclic stress. A similar creep type of relationship also exists for the accumulation of plastic strain with time, and from this a relationship is developed between pore pressure and strain. Effective stress paths for samples taken to failure suggest that the projection of the critical state line may be taken as a failure criterion under cyclic loading. Combining the exponential relationship for pore pressure against time with this failure criterion, related stability criteria are developed for pore pressure and strains at different cyclic stress levels.

An improved two‐dimensional plane‐strain numerical procedure based on the incremental‐iterative nonlinear finite‐element method is developed to predict ground subsidence caused by underground mining. The procedure emphasizes the use of the following features: (1) An appropriate constitutive model that can accurately describe the nonlinear behavior of geological strata; and (2) an accurate algorithm for simulation of excavation sequences consistent with the actual underground mining process. The computer code is used to analyze a collapse that occurred in the Blue Goose Lease #1 Mine in northeastern Oklahoma. A parametric study is conducted to investigate the effects of some selected factors on the shape and extent of subsidence profiles. Analyses of the numerical results indicate that the nonlinear finite‐element technique can be employed to meaningfully predict and characterize the potential for ground subsidence due to underground mining.

The method for cyclic characterization of liquefiable sands, which employs the concept of dynamic backbone curve in conjunction with the Masing criteria for construction of unloading and reloading branches of cyclic loops, is refined and verified. The method also encompasses subsequent cyclic loading, when soil undergoes significant degradation due to pore‐water pressure buildup. To accurately describe the initial backbone curve and associated initial cyclic loop, and subsequent degraded backbone curves and associated subsequent cyclic stress‐strain loops, a well‐known hyperbolic model and a degradation model are modified. Although relatively simple, such modifications enable accurate description of the stress‐strain loops of the first and subsequent cycles, even in cases when soil rapidly degrades and liquefies in just a few cycles. Consequently, a much better agreement between the measured and calculated damping values can be also obtained. The investigation is based on test results obtained on five different sands, which either liquefied during past earthquakes or are located in potentially liquefiable zones. The results presented are important for improvement of the nonlinear seismic response analyses of liquefiable sites.

Three‐dimensional computer models of construction sites and subsurface stratigraphy are useful for visualization, analysis, and design in geotechnical engineering. A useful technique for creating such models is the geometric technique known as “solid modeling.” Solid modeling was originally developed in the fields of mechanical and aerospace engineering. In these fields, the shape and complexity of objects is relatively simple and well defined as compared to those of the earth's surface and subsurface. Accordingly, because of the complex nature of the surfaces involved, most available modeling systems are not well suited for modeling earth masses. To overcome this difficulty, a technique employing surfaces in the form of triangulated irregular networks (TINs) and set operations was developed. This technique facilitates the construction of complex three‐dimensional models of earth masses from data that are typically available to geotechnical engineers. These data usually consist of surface‐topographic surveys and borehole logs. Using the technique, a geotechnical engineer can create a three‐dimensional model of a complex site. Complicated excavations can be modeled and cross sections, or “fence” diagrams, can be constructed quickly and easily, anywhere on the model, to display the soil stratigraphy at the site. Volumes of excavations and fills also can be computed easily.

A new index test method, the notched constant tensile load (NCTL) test, for evaluating high‐density polyethylene (HDPE) geomembranes regarding stress‐cracking resistance is presented in this paper. Eighteen commercially virgin geomembranes and 7 field‐exhumed geomembranes were evaluated using the new test. Results show a wide variation in the onset of the transition time. For the 18 virgin geomembranes, the transition times range from 10 hours to 5,000 hours. In addition, all 7 field‐exhumed geomembranes, where failure actually occurred, show transition times less than 100 hours. An evaluation of the test results leads to a tentative recommendation that the transition time for an acceptable HDPE geomembrane should be 100 hours or longer. A comparison between the results of NCTL test and the currently used ASTM D 1693 bent‐strip test was conducted. It indicates that the bent‐strip test cannot adequately evaluate stress cracking resistance, and that the NCTL test is a more critical test in the assessment of such behavior.