Date of Award

Spring 1-1-2011

Document Type


Degree Name

Doctor of Philosophy (PhD)

First Advisor

Dobroslav Znidarcic

Second Advisor

Hon-Yim Ko

Third Advisor

John McCartney


Geotechnical engineers are interested in flow phenomena of unsaturated soils, as they play an important role in interpreting the stability and deformation of slopes, retaining walls, and pavements under unsaturated conditions. Fundamental material characteristics of flow phenomena such as soil-water retention curve (SWRC), which is required to define the driving potential for the flow process, and hydraulic conductivity function (HCF), which is required to define the resistance to the flow process, are still primary concerns for characterizing the flow phenomena in unsaturated soils. Characterizing the SWRC and the HCF in flow phenomena of unsaturated soils is a challenge for many researchers even though the several measurement techniques have been proposed.

A flow pump system (FPS) was developed in early 1990 at the University of Colorado in order to characterize unsaturated soils. FPS has been improved over the past 20 years for characterizing the flow phenomena of unsaturated soils, and the new FPS technique is developed for measuring the suction response and SWRC induced by the flow pump during drying and wetting cycles. And, HCF is evaluated by using an inverse problem solution approach.

Three materials such as a uniform sand and well graded soils are used in this study for investigating flow phenomena in unsaturated soils. The following conclusions are drawn from the research result. The prediction models commonly used for HCF cannot properly represent the actual hydraulic conductivity of unsaturated soils. The results also show that the liquid-phase flow of pore water stops at relatively high degree of saturation and low suction. The threshold saturation and corresponding hydraulic conductivity values are difficult to determine for each material, and this confirms the hypothesis that hydraulic conductivity for unsaturated soils must be determined through careful experimentation and that no prediction model is valid for all soils. This research developed an adequate methodology to conduct such experiments straightforward, and paved a rational way to predict suction changes in soils due to environmental conditions.