Soil-Water Characteristic Curves for Unsaturated Soils - Unsaturated Soil Mechanics in Engineering Practice - page 188

surface, the water is acted upon solely by the force of gravity
and behaves according to the laws of groundwater flow.”
Soil is capable of permanently retaining considerable
quantities of water within its voids above the phreatic
surface. The permanent presence of water within this
zone supports the existence of forces that counteract the
mechanical effect of the force of gravity. These forces are
referred to as capillary forces.
Terzaghi (1942) used the complexities of the SWCC to
explain the relationship between capillarity and negative
pore-water pressure or soil suction. Terzaghi's writings made
it clear that water in the pores of a soil above the phreatic
line could participate in groundwater flow. Water was shown
to rise through capillary action to heights greater than 9 m.
The First International Conference on Soil Mechanics and
Foundation Engineering (1936) contained research papers
that discussed a wide range of unsaturated soil mechanics
problems. Vreedenburgh and Stevens (1936) studied unsat-
urated soil seepage problems and explained that the coeffi-
cient of permeability was a function of the degree of satura-
tion of the soil. Van Mourik Broekman and Buisman (1936)
noted that negative pore-water pressures played an important
role in the stability of slopes. The early developments in soil
mechanics appear to have included consideration of soils
in the unsaturated zone. Unfortunately, many of the early
findings seemed to become lost during subsequent decades.
Terzaghi (1943) devoted significant portions of two chapters
of his topic to unsaturated soil behavior. Later textbooks on
soil mechanics gave less attention to the behavior of unsat-
urated soils.
Terzaghi was well aware of the research work undertaken
in soil physics and often quoted from their research findings.
Terzaghi recognized the nature of the SWCC and its value
in understanding the permeability function for an unsatu-
rated soil. His perception was that the uppermost part of the
moistened zone is in a state of partial capillary saturation.
The rate of flow of the water appeared to be determined by
capillary potential which was a function of the degree of
saturation of the sand (Baver, 1940). However, the methods
of computation which are based on the concept of the cap-
illary potential had not yet reached a stage such that they
could be applied to the solution of engineering problems.
Simplifications were made for solving seepage problems.
For example, the saturated coefficient of permeability was
used as long as the soils were within the capillary fringe
zone. In other cases, the phreatic surface was assumed to
form the uppermost boundary of flow, as was the case for
flownet solutions proposed by Casagrande (1937).
Lambe (1958b) attempted to find a single soil property
for flow in the negative pore-water pressure zone. “Capillary
head” was used to describe “draining” and “wetting” condi-
tions. Geotechnical engineers attempted to further understand
and apply water flow theories to unsaturated soils during
the 1950s and 1960s. A common misconception appears to
have prevailed which assumed that water could only flow in

saturated soils within the “capillary fringe” and in the positive
pore-water pressure range (Barbour, 1998).
The 1960 Conference on Pore Pressure and Suction in
Soils witnessed a difference of opinion with respect to for-
mulating mathematical equations for solving unsaturated soil
problems within the engineering profession. British engi-
neers attempted to explain unsaturated soil behavior through
use of a single-stress-state variable. Australian engineers
emphasized an approach that advocated the use of a con-
tinuous mathematical function to represent the SWCC. The
Australian approach found its basis in earlier research in
soil physics (Aitchison, 1961). The Australians felt that the
historical research in soil physics had much to contribute
toward establishing an acceptable fundamental, contextual
framework for solving unsaturated soil problems in geotech-
nical engineering.
Gardner (1961) combined the concepts of water potential
and its relationship to the coefficient of permeability for an
unsaturated soil. This gave rise to the application of a con-
tinuous function for describing seepage through unsaturated
soils. The result was a nonlinear analysis that would need
to wait for the development of the computer in order to
solve highly nonlinear engineering problems. The finite ele-
ment numerical methodology would later utilize continuous
permeability functions to solve steady-state seepage formu-
lations (Freeze, 1971; Papagiannakis and Fredlund, 1984).
Water storage functions were added shortly thereafter for
solving transient seepage analyses (Lam et al., 1987). The
computational power of the computer provided a powerful
tool for the solution of geotechnical engineering problems
that contained nonlinear soil property functions.
In 1977, Fredlund and Morgenstern began to establish
a common theoretical stress state basis for describing and
formulating geotechnical engineering problems involving
unsaturated soils. Amacroscopic phenomenological approach
based on principles of multiphase continuum mechanics
was advocated and used to provide a theoretical basis for
defining the stress state variables for an unsaturated soil. The
stress state variables formed the basis for the formulation
of volume-mass, shear strength, and water flow constitutive
relations. In the 1980s and 1990s, the SWCC would become
the basis for the estimation of nonlinear unsaturated soil
property functions for all types of geotechnical engineering
problems.

5.1.6 Need for Unsaturated Soil Property Functions
The SWCC has had an important role to play in the imple-
mentation of unsaturated soil mechanics (D.G. Fredlund,
2002a). The SWCC was initially viewed as a means of esti-
mating in situ soil suctions by measuring the natural water
content and using the SWCC as a fixed relationship between
suction and water content. Laboratory data showed that the
drying and wetting SWCCs were significantly different (i.e.,
hysteretic), giving rise to a wide range of possible soil suc-
tion values. The SWCC later proved to be of significant