Research engineer, Shell Development Co., Houston, Tex.
Professor of civil engineeringMember of ASTM, University of Texas, Austin, Tex.
Vice-president, Haley and Aldrich, Inc., Cambridge, Mass.
Analyses to obtain theoretical time-settlement curves for embankments typically make use of soil properties measured in the laboratory on small disturbed samples subject to strain rates and hydraulic gradients that are several orders of magnitude greater than those encountered in the field. Only analyses of well-documented case histories can demonstrate the validity of such extrapolations. Refined analyses based on detailed field observations of settlements and pore water pressures at various depths beneath the main I-295 highway embankment adjacent to the Fore River in Portland, Me., have demonstrated that: (1) the field e versus log gs curves are typically above the laboratory curves and have a sharper break at ¯σmax; (2) cv is higher in the field than in the laboratory; (3) at this site cr/cv ranged from 1 to 2; and (4) there was a generally good correlation between field properties and reconstructed laboratory properties. At this site the large differences between strain rates and hydraulic gradients in the laboratory and the field apparently did not introduce important errors.
Paper ID: GTJ10586J