Field Observations of Thawing in Soils: Edward C. Mcroberts

CAN.
GEOTECH. J.
VOL.
12, 1975
Field Observations of Thawing in Soils

EDWARD
C. MCROBERTS
R. M . Hardy and Associates Limited, 219-18 Street, S.E., Calgary, Alberta T2P OT5
Received August 20, 1974
Accepted September 3, 1974
Case histories are considered which indicate that under some field conditions a thaw interface
proceeds as predicted by a simple model. Some reasons for possible deviations are investigated
and comparisons are made between calculated and measured rates of thaw.
L'Btude de cas pratiques indique que dans certaines conditions in situ un front de degel se
dtveloppe comme prkdit par un modkle simple. Quelques causes de deviations possibles sont
Ctudites et des comparaisons sont faites entre les vitesses de deael calcul6es et mesurkes.
[Traduit par la Revue]
Introduction
Morgenstern and Nixon (1 97 1) have focused
attention on the importance of thaw-consolidation theory and its role in controlling various
aspects of the mechanical response of a thawing
soil. They derive an analytic solution for the
excess pore pressure set up at a thaw interface,
assuming that the Neumam solution describes
the movement of the thaw plane as
where X denotes the depth of the thaw plane
t is time, and
a is a constant with metric units (cm/s")
Thaw-consolidation theory has been applied
to a one-dimensional stability analysis of thawing slopes by McRoberts (1972) and McRoberts
and Morgenstern (1974), who assumed that thaw
progressed in accordance with Eq. [I]. Since
this analysis is a useful method with which to
evaluate the stability of thawing slopes, it is of
considerable interest to inspect the applicability
of Eq. [I.] in practice and to consider possible
departures from it. It should be pointed out
that this study is more concerned that the model
used for thaw, i.e. Eq. [I], is applicable to field
situations and is less concerned with the ability
to accurately calculate measured values. However, we shall also investigate the power of a
simple analytic solution to provide reasonable
estimates of observed thaw rates.
Measured and Calculated Rates of Thaw

Case histories of thaw have been analyzed by
plotting the depth of thaw against the square
root of time. A summary of 15 case records
Can. Geotech. J., 12,126(1975)
available in the literature is given in Table 1

and plots of the progression of the depth of
thaw with time are shown on Fig. 1.
As noted in Table I the position of the thaw
interface can be established either from temperature profiles or by probing. Generally,
probing is accomplished by pushing a sharpened
steel rod down to the base of the active layer,
and the record for the entire thaw season is
usually based on probes at different locations
under the same surface covers. The beginning
of the thaw season has been evaluated using a
variety of techniques depending upon the information presented in the original case record.
In some case records, the first day of thaw was
assumed to occur on a date intermediate between the last day of known frost in the ground
and the first observation of thaw. In other cases
various meteorological observations were considered in order to obtain an estimate of the
first day of thaw. For example, in some cases
the date of the final disappearance of snow
cover and of the rapid rise in air temperature
was used to obtain an estimate of the first day
of thaw. In other cases the progression of thaw
over the entire season was given and the initial
day was therefore defined. The best fit line,
beginning at the origin can then be placed
through the data points. A value for can then
be caiculated from the slope of this line. These
values have been recorded for each case in
Table 1.
For most of the case records there is a
marked linear relationship in accordance with
Eq. [I], although a close inspection of the data
presented on Fig. 1 indicates that there are
certain deviations from a truly linear relation-
NOTES
FIG. 1. Case records for depth of thaw.
ship. We shall consider some reasons for this

departure in a subsequent section.
It is also of interest to attempt to calculate
the observed thaw history at each site. Equation
[ l ] states that there is a linear relationship between the depth of thaw and the square root
of time when a step change in temperature is
applied to the surface of a frozen, homogenous,
semi-infinite medium. Nixon and McRoberts
( 1973) have considered the several assumptions embodied in Eq. [I] and have assessed
the relative importance of the various paramThey have also
eters necessary to compute LY.
reviewed some practical simplifications of the
Neumann problem, the most basic of which is
the Stefan solution, written as:
where T, denotes the step temperature applied,

k, denotes the conductivity of the thawed
soil above the thaw front, and
L denotes the volumetric latent heat of
the frozen soil
and where
Equation [2] is derived based on the assumption that the frozen ground is at a temperature
T , = 0.0 O C (see Nixon and McRoberts 1973).
While it is evident that surface temperatures are
rarely constant and that soil properties are often

variable with depth, contrary to the assumptions
implicit in Eq. [2], we shall see that this solution yields reasonable results.
In order to determine LY using Eq. [3] the
values of (k,,, L, T,) must be estimated for each
case record. The thermal parameters k,, and L
can be related to the amount of water in the
soil. The thawed conductivity can be computed
using formulae given by Kersten ( 1949) for
silt-clay or sand-silt mixtures once the thawed
water content is known. In this study it is
assumed that the soil is saturated. The presence
of unfrozen water in a frozen fine-grained soil
will influence the magnitude of L. Nixon and
McRoberts ( 1973) have presented a method
by which the effects of unfrozen water content
may be incorporated into a calculation for LY,
but this approach has not been followed here
because there is a lack of information for most
case records. In fact, in most of the case records
field measurements of the ice contents are not
available and L values have been estimated
based upon measured thawed water contents in
the active layer.
The third variable that must be determined
is T,. As the case records considered herein
have been assembled from a variety of sources
it is evident that various methods will have to
be used and different assumptions made for
each individual record. In cases 7, 8, 14, and
15 as noted on Table 1, the magnitude of the
CAN. GEOTECH. J.
VOL.
12, 1975
TABLE1. Case histories for rate of thaw

Measured
Code
(Fig. 1)
Calculated
cm/s*
Code
Soil
temperature
At 3-in. depth
mean 4 "C
estimated
surface 5.0 "C
Mean, measured
surface temp.
6 "C
Thaw index -+
thaw season =
5.8 "C from air
temp.
Thaw index +
thaw season =
10.5 "C from
air temp.
Avg. soil
surface temp.
7.5 OC.7
Avg. soil
surface temp.
11 OCt
Avg. soil
surface temp.
15 "Ct
Monthly mean
surface: 10 "C
Monthly mean
surface: 11 "C
Soil profile
Avg. w/c 35%
Avg. w/c 35%
Avg. w/c 23%
Avg. w/c 20%
Organics and
silty clay
Avg. w/c 35%
Organics and
silty clay
Avg. w/c 35%
Tundra soil
Avg. w/c 60%
cm/si
Comments and
surface cover
Salix alnus (Gill 1971)
Salix alnus (Gill 1971)
Salix equisetum (Gill 1971)
Equisetunz (Gill 1971)
Picea (Gill 1971)
Picea (Gill 1971)
Tundra site
(Drew et al. 1958)
Estimated
w/c 35%
Kelley and Weaver (1969)
Avg. w/c 35%

in silt and
clayey sand
18 cm organic
soil at surface
(Aitken 1965)
6 cm peat w/c
56% over silty
clay, avg. w/c
overall 30%
Diamiction: sand,
silt, clay. Avg.
w/c 20%
Diamiction. Avg.
w/c 17%
Sparse grass cover

(Aitken 1964)
Site ES6, TCS 6B.

No organic cover
(Washburn 1967)
TCS 7B. No organic cover
(Washburn 1967)
Diamiction: Avg.
ice contents 22%
TCS 7A. No organic cover

(Washburn 1967)
Silty Clay.
Avg. w/c 30%
Sand with clay
lenses avg. w/c 20%
Sparse vegetation
(Powell 1961)
Sparse vegetation. Data
are the average for 2
stations (Powell 1961)
*T, deduced from average of thaw index, mean of maximum and minimum measured surface temperature, and surface temperature extrapolated from temperature grad~ents.
fAverage soil surface temperature from extrapolation of measured gradients.
Code T = depth of thaw from temperature profiles.
Code P = depth of thaw from probing.
temperature at the surface can be obtained

directly from measured soil surface temperatures or near surface temperatures. In cases 11,
12, and 13 the soil surface temperature was
obtained by extrapolating measured temperature gradients within the active layer to the soil
surface. For these cases the available soil surface temperature data was then plotted against
time and an average value picked for T,.
In other less completely documented records
the surface temperature must be deduced from
air temperatures. In this study, it was simply
assumed that the surface temperature is equal

to the ambient air temperature. It is emphasized
that this need not be the case. In cases 9 and
10 the average step temperature was calculated
by dividing the quoted thaw index by the length
of the thaw season. In cases 1, 2, 3, 4, 5, and 6
a combination of methods were used, as noted
on Table 1.
The magnitude of a for the 15 case records
can then be calculated using Eq. [3] and the
above mentioned assumptions. These calculated
values are listed in Table 1 and compared with
the measured values on Fig. 2. Keeping in mind
the relatively crude assumptions that have been
made, it is evident that the agreement between
measured and calculated values of the thaw
parameter, a, is quite good.
Discussion
The main aim of this study was to investigate
whether or not Eq. [ I ] was a reasonable model
for thaw in field situations. As has been shown
in Fig. 1 this is indeed the case.
It may be argued that a time-dependent surface-temperature boundary condition governs
the in situ progression of thaw and that a
sinusoidal surface temperature may be more
appropriate. Nixon and McRoberts (1973, Fig.
11) have presented an example of the history
of a thaw interface under both a sinusoidal

surface temperature and an equivalent step
temperature and they observed a reasonable
correspondence between the two interfaces. If
case records 2, 5, 7, and in particular 12, Fig.
1, are closely inspected it can be seen that there
is some departure from linearity. This may be
interpreted as being due to a sinusoidal surface
temperature. If a natural active layer is made
up of a thick layer of ice-rich organic soil overlying mineral soil, as is often the case, then
Eq. [ I ] can only be used in the overlying organic soil and various extensions to thaw
penetration theory must be considered (see
Nixon and McRoberts 1973).
It is also of importance to observe that there
is not a large range in the magnitude of the
measured a values, Fig. 2. The reason for this
can be clearly seen if we solve Eq. [3] for a
reasonable range of step temperatures, say 2 "C
and 20 "C and for water (ice) contents of 5 %
to 70% (Fig. 3 ) . The range of computed
values is from 0.015 to 0.095 cm/slh and encompasses all measured values (Table 1 ).
It has been shown that in situ thawing of
naturally occurring active layers can proceed as
modelled by Eq. [I]. On the other hand more
accurate calculations of a values are difficult
and are handicapped by a lack of complete case
records. For the purposes being considered in
this study a well documented case history would
include the following information: a record of
air, surface, and sub-surface temperatures; the
ice content of frozen soil; the water content arid
STEP TEMPERANRE
OC
002
004
006
20
40
60
WATER (ICE) CONTENT
CALCULATED o ( c r n / ~ ' ' )~
FIG. 2. Comparison of measured a with

lated by a Stefan solution.
cu
calcu-
FIG. 3. Thaw parameter

Stefan solution.
(%)
calculated by the
130
CAN.
GEOTECH. J.
bulk density of the thawed soil; a detailed soils

profile and record of the surface cover and
condition; and finally, a plot of the depth of
thaw with time.
Conclusions
This study has assessed the adequacy of a
particular thaw model often used in thaw-consolidation theory and shown it to be a useful
model for a variety of field situations, although
in others, extensions to the theory are required.
We have also seen that the magnitude of the
range of the rate of thaw is not large. This
conclusion is of itself important as it substantiates arguments made elsewhere (McRoberts
1972; McRoberts and Morgenstern 1974;
Nixon and McRoberts 1973) concerning the
relative importance of thermal and geotechnical
properties in calculating deformation and stability of thawing soils.
AITKEN,G. W. 1964. Ground temperature observations,
Gulkana, Alaska. U.S. Cold Reg. Res. Eng. Lab.,
Hanover, New Hampshire, TR106.
AITKEN,G. W. 1965. Ground temperature observations,
Barrow, Alaska. U.S. Cold Reg. Res. Eng. Lab.,
Hanover, New Hampshire, TR105.
DREW, J. V., TEDROW,J. C. F., SHANKS,R. E., and
VOL. 12, 1975
J. J. 1958. Rate and depth of thaw in arctic

KORANDA,
soils. Trans. Am. Geophys. Un. 39, pp. 697-701.
GILL,D. 1971. Vegetation and environment in the Mackenzie River Delta. Unpub. Ph.D. thesis, Dep. Geol.,
Univ. British Columbia, Vancouver, B.C.
KELLY,J. J., and WEAVER,D. F. 1969. Physical processes
at the surface of the arctic tundra. Arctic, 22, pp.
425437.
KERSTEN,M. S. 1949. Laboratory research for the determination of the thermal properties of soils. Final Report, Eng. Experimental Stn., Univ. Minnesota, Minneapolis, Minn.
MCROBERTS,E. C. 1972. Discussion. Proc. Can. North.
Pipeline Res. Conf., Ottawa, N.R.C.C. Tech. Rep.
104, pp. 291-295.
N. R. 1974. A
MCROBERTS,E. C., and MORGENSTERN,
study of landslides in the vicinity of the Mackenzie
River, Mile 205 to 660. Environmental-Social Comm.,
Northern Pipelines, Task Force on Northern Oil Development Rep. No. 73-35.
N. R., and NIXON, J. F. 1971. OneMORGENSTERN,
dimensional consolidation of thawing soils. Can.
Geotech. J. 8, pp. 558-565.
NIXON,J. F. 1973. Thaw-consolidation of some layered
systems. Can. Geotech. J. 10, pp. 617-631.
NIXON,J. F., and MCROBERTS,E. C. 1973. A study of
some factors affecting the thawing of frozen soils.
Can. Geotech. J. 10, pp. 439452.
POWELL,J. M. 1961. Operation Hazen. McGill Univ. Arctic Met. Res. Group Publ. in Meteorology, No. 38.
WASHBURN,A. L. 1967. Instrumental observations of
mass-wasting in the Mesters Vig District, Greenland.
Medd. om Grpnland. 166(4), pp. 1-297.
Pore Water Expulsion during Freezing

EDWARDC. MCROBERTS'
R. M. Hardy and Associates, Calgary, Alberta T2E 6J5
AND
NORBERTR. MORGENSTERN
Department of Civil Engineering, University of Alberta, Edmonton, Alberta T6C 2C7
Received September 18, 1974
Accepted October 7, 1974
When a freezing front advances through a saturated soil water may either be expelled or
attracted to the freezing front depending upon soil type, stress level, and rate of freezing.
Experimental evidence is considered which shows that coarse-grained sandy soils expel water
under most conditions while fine grained soils can be made to expel water only at higher
overburden pressures. A solution for the excess pore pressures that can be generated due to
impeded drainage by pore water expulsion in an open system is presented.
Lorsqu'un front de gel progresse B travers un sol saturk, l'eau peut i t r e expulsee ou au
conIraire attirke vers ce front, selon le type de sol, le niveau de contrainte et la vitesse de
'Formerly: Graduate Student, Department of Civil Engineering, University of Alberta, Edmonton, Alberta.
Can. Geotech. J., 12,130(1975)

Field Observations of Thawing in Soils: Edward C. Mcroberts

Uploaded by

Copyright:

Available Formats

Field Observations of Thawing in Soils: Edward C. Mcroberts

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Field Observations of Thawing in Soils: Edward C. Mcroberts

Uploaded by

Copyright:

Available Formats

CAN.

Field Observations of Thawing in Soils

Measured and Calculated Rates of Thaw

available in the literature is given in Table 1

FIG. 1. Case records for depth of thaw.

ship. We shall consider some reasons for this

where T, denotes the step temperature applied,

rarely constant and that soil properties are often

TABLE1. Case histories for rate of thaw

Picea (Gill 1971)

Kelley and Weaver (1969)

Avg. w/c 35%

Sparse grass cover

Site ES6, TCS 6B.

TCS 7A. No organic cover

temperature at the surface can be obtained

assumed that the surface temperature is equal

of a thaw interface under both a sinusoidal

WATER (ICE) CONTENT

CALCULATED o ( c r n / ~ ' ' )~

FIG. 2. Comparison of measured a with

FIG. 3. Thaw parameter

bulk density of the thawed soil; a detailed soils

VOL. 12, 1975

J. J. 1958. Rate and depth of thaw in arctic

Pore Water Expulsion during Freezing

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.