Effect of shear strength parameters on slope stability analyses in the Winnipeg area

Abstract

Several previous investigators including SUTHERLAND (1966), MISHTAK (1964) AND BARACOS (1961) have shown that slope stability analyses in the Winnipeg area, using laboratory test results for the Winnipeg clays, will substantially overestimate the factor of safety. The majority of the slope failures occur in the stratum of highly plastic lacustrine clay of glacial Lake Agassiz which varies in thickness from 35.0 to 55.0 feet and overlies glacial till or limestone bedrock. The clay is noticeably varved. Consideration is given to various methods of both total and effective stress analysis and different sets of shear strength parameters as determined by triaxial and direct shear tests. All analyses are applied to the best documented case history which is the P.F.R.A. test pit constructed for the Great Winnipeg Red River Floodway. Using conventional shear strength parameters as determined by conventional triaxial tests, all methods of total and effective stress analysis substantially overestimate the factor of safety. Bishop's method of analysis used with various sets of peak shear strength parameters determined by triaxial tests and direct shear tests with the failure plane perpendicular to the varves also substantially overestimates the factor of safety. Residental shear strength parameters substantially underestimate the factor of safety. Peak shear strength parameters determined by direct shear tests with the failure plane parallel to the varves used in conjunction with Bishop's Method of Analysis produced safety factors of 1.09 and 1.20.