Theory of Water Flow through Unsaturated Soils - Unsaturated Soil Mechanics in Engineering Practice - page 374

coefficients of permeability for the CL sand ranged from
3

permeability values calculated from the inferred SWCC. The
permeability functions show a loglinear relationship in the
suction range between 10 and 10,000 kPa. The slope of
the loglinear portion of the permeability function appears to
increase with an increase in the proportion of fines con-
tent in the soil. The permeability functions obtained are
located in a relatively narrow loglinear permeability band
with the upper permeability values about 5-10 times greater
than the lower permeability values at the same suction.
A column length of 400-600mm appears to be appropri-
ate when using the wetting-front advancing column test.
Each wetting-front column test can be performed in 1 or
2 weeks.

10 − 7 m/s. The slope in the loglinear
part was 2.18. As the soil suction increased from 10 to
10,000 kPa, the permeability decreased from 9 . 0

10 − 8

× to 3 . 0 ×

10 − 8

× to

10 − 13 m/s.
The permeability functions for sandy ML and SC with
gravel are shown in Figs. 7.50 and 7.51, respectively. The
permeability functions calculated from the monitored suc-
tions agree well with the permeability function obtained
when suctions are inferred from the SWCC. The perme-
ability functions interpreted based on the measurements at
all four monitoring sections appear to be quite similar. The
calculated permeability functions is located along a loglin-
ear permeability band, where the upper value was about
3-8 times greater than the lower values at the same suction.
The Gardner (1958a) equation also best fit the calculated
data when using a saturated coefficient of permeability of
2

1 . 0 ×

7.6.5 From Measurement to Estimation
of PermeabilityFunction
The theory of water flow through unsaturated soils has been
verified through experimental testing. The hydraulic con-
ductivity or coefficient of permeability can be presented as
a function of soil suction. While it is possible to experi-
mentally measure the permeability function for a soil, the
time and costs are generally considered to be too prohibitive
in geotechnical engineering practice. As a result, consider-
able attention has been given to the use of “estimation”
techniques for obtaining the permeability function.
A significant portion of the next chapter is devoted to
methodologies for the estimation of the permeability func-
tion and its usage for solving practical geotechnical engi-
neering problems.

10 − 6 m/s for the ML soil and a saturated coefficient of
permeability of 6

×

10 − 7 m/s for the SC soil.
The results of the wetting-front advancing column tests
appear to provide an acceptable procedure for the measure-
ment of the unsaturated permeability function for a range
of coarse-grained materials. The permeability values inter-
preted based on measurements at different sections along
the column appear to follow a consistent trend. It appears
that the interpreted permeability function is independent of
the location at which the suction or water content mea-
surements are made. The calculated permeability function
based on measured suctions appears to agree well with the

×