Hydro-Mechanical Properties of an Unsaturated Frictional Material

2.3. HYDRAULIC FUNCTIONS 31
draulic properties of porous media, where water content and pressure were measured.
The pressure and water content measurements were obtained in different depths and
thus the suction-water content curve could be not obtained directly. Experiments in a
column apparatus were also performed by Yang, Rahardjo, Leong & Fredlund (2004b)
and Yang, Rahardjo, Wibawa & Leong (2004), where the pore water pressure and also
the water content was measured pairwise in several depths. Their study considered infil-
tration tests on coarse materials. However, the effect of hysteresis was not investigated
by applying several drainage and imbibition cycles to the soil. If water content and
suction can be measured as a function of time, soil-water characteristic curve can be
determined by pairing the water content value with the suction value at a given time un-
der transient state condition. Measurements of this type of experiment were conducted
by Watson (1965), Topp et al. (1967), Wana-Etyem (1982), Perroux et al. (1982). In-
filtration tests were performed by Ahuja & El-Swaify (1976) and instantaneous profile
method by Watson (1965), Rogers & Klute (1971), Vachaud et al. (1972).
As already mentioned, it is traditionally assumed that the soil-water characteristic curve could
be measured either under steady state condition (equilibrium condition) or transient state
condition (non-equilibrium condition). Comparison of suction-water content measurements
obtained under conventional method (steady state condition) and using transient state method
indicates the presence of the so called dynamic effect for some type of soils. Investigation of
several authors showed that the hydraulic properties could also depend on the dynamic of
water flow (e.g. rate of suction changes).
Davidson et al. (1966) performed drainage as well as imbibition experiments. They found,
that more water was removed from the sample when one large suction increment was ap-
plied rather than many small suction increments during drainage process. In contrast more
water was absorbed when many small suction increments were applied during imbibition pro-
cess. Topp et al. (1967) conducted tests on a sand column under static state condition and
transient state condition. At a given potential the water content measured under transient
state condition was significantly higher than the water content measured under static state
condition during drainage process. Topp et al. (1967) found that the difference between static
soil-water characteristic curve and transient soil-water characteristic curve depends on the
size of the imposed pressure step and on the time, which is required to reach equilibrium
condition in the tested soil. The dynamic flow in a sand column was investigated by Stauffer
(1977), who also measured higher water content at a given suction during drainage process
compared to static state tests. One-step and multi-step outflow experiments were performed
by Wildenschild & Hopmans (1997). The soil-water characteristic curve was determined from
tensiometer measurements and the average water content. For large suction increments the
water content was higher at given suction value. Dependence of flow rate on unsaturated
hydraulic properties has been investigated by few authors. Experiments were carried out