Title

Investigating the reliability of H/V spectral ratio and image entropy for quantifying the degree of disintegration of weak rocks

DOI

https://doi.org/10.1016/j.enggeo.2016.04.020

Document Type

Article

Publication Date

6-3-2016

Publication Title

Engineering Geology

Abstract

Characterization and quantification of disintegration behavior of weak rocks, such as shales, claystones, mudstones, and siltstones, are usually a difficult task because of their extremely diverse nature. These rocks present numerous stability problems along cut slopes around the world because they can change rapidly from rock-like to soil-like materials upon exposure to natural climatic conditions. Engineers often overestimate the design parameters for cut slopes in these rocks when laboratory tests are performed on fresh samples. Adjusting the design parameters for the anticipated amount of disintegration can be problematic because of the complex and time-dependent nature of disintegration behavior. In this study, we investigate the reliability of using microtremor H/V spectral ratio, derived from microtremor signals, and image entropy, obtained from image analysis techniques, for quantifying the varying degrees of disintegration. During 2011-2014, both techniques were used to study disintegration of slopes cut through claystones, mudstones, and limestones along the 80-km stretch of a highway between Bogotá and Tunja in central Colombia. For image analysis, a new algorithm was developed to discriminate between varying degrees of disintegration across the complete color structure of the image, avoiding the gray-scale transformation that is usually time consuming. The results indicate significant correlations between the degree of disintegration of the cut face, the dominant frequency estimated from the microtremor H/V ratio, and the image entropy values, suggesting that H/V ratio and image entropy can be reliable techniques for quantifying the degree of disintegration.

Volume

207

First Page

115

Last Page

128

ISSN

00137952

Identifier

SCOPUS_ID:84964875011

Documento original

Compartir

COinS