The San Francisco waterfront was constructed between 1850 to 1920 by placing fill along the San Francisco Bay. The fills, reaching thicknesses of up to 40 ft (12 m), consist primarily of dune sand, rock fragments, bay mud, and construction debris. They were placed largely by dumping into the waters of the Bay. Not surprisingly, certain fill areas underwent large displacements in the 1906 San Francisco earthquake. Yet, at the same time, other areas showed little displacement. This behavior, plus that recorded in other seismic events, provides a unique record of fill behavior under earthquake loading. The study of the waterfront fills involved an historical review of waterfront construction, field testing and sampling, laboratory tests, and analytical studies. The field work was done both in areas where seismically induced movements were large and where they were small. The results of the investigation serve to explain the observed behavior of the fills; loose zones of the dune sand were found in the large movement areas which undoubtedly liquefied in the 1906 event. These were not present where only small movements occurred.

TWO probabilistic models are proposed to study the development of pore pressure in saturated sands under random loading. The first of these is based on laboratory data, while the second uses an analytically based effective stress technique. Both incorporate uncertainties in soil parameters, laboratory data, and earthquake loading parameters, and compute the cumulative distribution function of pore pressure at the end of any cycle of loading. The two models are cast into a methodology which evaluates the probability of developing a certain level of pore pressure over a given time frame under a specified probability of seismic loading. They allow probabilities to be defined as to the generation of any pore pressure ratio from 0 to 1. They are suited to assess the relative effects of soil resistance, seismic environment, and geotechnical uncertainty. Limitations in application of the methods lie primarily in definition of soil and seismological parameters; however, future basic research should help to resolve this problem.

This paper describes the results of two full scale rock socket load tests conducted in conjunction with the foundation design of a sulfur dioxide removal plant located at the Four Corners Generating Station near Farmington, New Mexico. The tests were conducted in a soft shaley sandstone with an average unconfined compressive strength less than 1,500 psi $(10.3\times {103}^3\mathrm{kPa})\text{.}$ The strength parameters of the sandstone were investigated by means of laboratory triaxial shear tests as well as Menard pressure testing in the field. The sockets were loaded to a maximum of 2,000 kip $(9.2\times 105\mathrm{kg})$ and were monitored by sets of Carlson strain gages along the length of the socket and along the bottom diameter. The results of these tests indicate the skin resistance mobilized to be nonuniform and that socket length greater than 2 diam or 3 diam do not significantly affect socket capacity.

The phenomenon of transient water flow in earth embankments is examined. State‐of‐the‐art solution techniques and their shortcomings are outlined. It is hypothesized that better solutions are obtainable through centrifugal modeling of the process. An experimental program is described. In order to interpret the experimental results, theoretical scaling relations for quantities pertinent to transient water flow are developed through consideration of the basic independent variables governing the flow. Testing was conducted on a geotechnical centrifuge in a container specially constructed for the test series. Acceleration levels of 25 g, 37.5 g, and 50 g were used. Pore pressures were measured and recorded during the transient rise and fall of the headwaters by four miniature pressure transducers. From these data, histories of total head as a function of time and equipotentials during headwater rise and drawdown are plotted. Steady state flow rates are also measured. Analysis of the data leads to the conclusion that theoretical scaling relations are valid and that this method does indeed give a better understanding of the phenomenon of transient flow and may be useful in design practice.

Methods for estimating the stability of the ocean floor under wave loading and the determination of wave‐induced pore‐water pressures are critically reviewed. Two new computer programs for analyzing the effects of waves on the seafloor are described: (1) STAB‐MAX for computing transient pore‐water pressures and effective stresses; and (2) STAB‐W for computing residual porewater pressures and estimating liquefaction potential. Predictions by STAB‐MAX are verified by field measurements of transient wave‐induced pore‐water pressures. Ranges of validity of simple procedures for predicting transient porewater pressures are established. Parametric studies are described which provide data on the pore‐water pressure response of dense and loose sands over a wide range of anisotropic permeabilities.

Cyclic shear stresses developed below the ocean bottom as a result of a passing wave train may progressively build up pore pressure in certain soils. Such built‐up pore pressure may eventually be large enough to result in significant deformation or develop cyclic liquefaction of these soils. A typical ocean wave‐induced liquefaction analysis requires data on ocean wave characteristics, geotechnical details of the subbottom soil profiles, cyclic shear stress distribution below the sea bottom resulting from passage of waves, and laboratory test results defining the cyclic shear strength of the liquefiable soils. Analytical and testing methods currently available for performing such analyses are complex, time consuming, generally too expensive, and not correlated with field experience. A simplified procedure has been developed which can be used as a first step to check if a more detailed analysis is warranted. The procedure is examined in the light of four case histories.

Waves propagating over a permeable seabed will alter the stresses and fluid pressures within the bed and affect the erodibility and stability of bottom sediments. This paper examines the cyclic stresses which occur within a rigid bed subjected to wave loadings by means of a rational pseudo‐static approach. The results of this study suggest that wave loadings on a saturated sand bed can alter the effective stress state throughout the bed. Cyclic stress ratios are maximum at the sediment surface and decrease with sub‐bottom depth. Laboratory and field experimentation are needed to verify existing theories and to better understand the stressing mechanisms involved in this complicated problem.

Measured wave induced pore pressures in a model sand layer are presented. The measured pressures vary with the wave height and period, sand layer thickness, and sand density. For short‐period waves, measured pressures compare well with theoretical (2nd order wave theory) pressures on the surface of the sand layer. Measured pressures are greater than theoretical when the model wave period is greater than 1.7 seconds. Pressures within a sand layer decrease more rapidly with depth than predicted by published theories. A time lag between sinusoidal pressure variations within the sand layer was measured during the laboratory study. This time lag, if it occurs in the prototype, may cause bottom instability during the long‐period (storm) waves.

Failure of a major offshore platform caused by a seafloor slide and hurricane wave forces is described. Data on the soils and waves are combined with limit equilibrium, finite element and layered continuum analytical models to determine soil movements and forces. Estimates are made of the ultimate strength of the platform foundation. Comparison of force and resistance characterizations indicates good agreement. Ultimate strength design methods are suggested to supplement present elastic methods in design of slide‐resistant platforms.