2004 Denver Annual Meeting (November 7–10, 2004)

Paper No. 5
Presentation Time: 2:45 PM

A STATISTICAL APPROACH FOR PREDICTING THE SHEAR STRENGTH OF MUDROCKS


HAJDARWISH, A.M.1, SHAKOOR, A.1 and DAHL, P.S.2, (1)Department of Geology, Kent State Univ, Kent, OH 44242, (2)Dept. of Geology, Kent State Univ, Kent, OH 44242, ahajdarw@kent.edu

Shear strength is one of the most important properties for design of engineering structures built on or within mudrocks (shales, claystones, mudstones, siltstones, etc.), and also the most difficult to evaluate. This is because it is usually difficult to obtain undisturbed samples of mudrocks, due to their weak nature, required for determination of shear strength parameters. The purpose of this research was to investigate geological and engineering properties that can be used statistically to predict the shear strength parameters of a broad range of mudrocks. Forty-five samples of various types of mudrock were collected from highway cuts throughout the United States. Clay content, clay mineralogy, water content, Atterberg limits, specific gravity, dry density, void ratio, absorption, adsorption, slake durability, and shear strength parameters (c and φ) were determined for each sample. Data were analyzed statistically, using bivariate and multiple regression techniques, to determine the correlations between shear strength parameters and geologic and engineering properties. Based on the statistical analyses, prediction models were developed for all mudrock types as a single group as well as separately for shales, claystones, mudstones, and siltstones. Preliminary results show that cohesion and friction angle parameters for all mudrocks, treated as one population, can be predicted from selected properties (amount and type of clay, Atterberg limits, % absorption, % adsorption, dry density, void ratio, and slake durability) with adjusted R2 values of 0.77 and 0.81, respectively. When mudrocks are subdivided into individual types and analyzed separately, the adjusted R2 values for predictive models are improved. The cohesion for shales, claystones, mudstones, and siltstones can be predicted with R2 values of 0.85, 0.98, 0.99, and 0.98, respectively, whereas the R2 values with respect to friction angle are found to be ~0.99 for all four types of mudrock.