Heat Transfer and FrostThaw Penetration in Soil
Surrounding an Inclusion of Sand
Numerical Model Results Relevant to Electromagnetic
Sensor System Performance
LINDAMAE PECK AND KEVIN O'NEILL
INTRODUCTION
The silty soil is assigned a moisture content of
This study was motivated by field data taken
10, 17 or 25% (by weight); the sand, 3%. Because
at a Vermont site over three winters that indicat-
variation in the unfrozen moisture content of the
ed a slight temperature difference throughout
soil affects buried electromagnetic sensor sys-
each winter in the sandy loam soil surrounding
tems, the depth and the timing of freezing and
an inclusion of sand (16.5 cm wide, 15 cm deep,
thawing of the silty soil, with and without a
and 15 cm thick), compared to soil at the same
sand inclusion, are emphasized in the presenta-
depth in a location where there was no sand in-
tion of results.
clusion. The difference in soil temperature at the
two nearby locations (∼24-m separation) was not
explainable by either a different near-surface soil
BACKGROUND
condition or a different soil condition at depth
When a sensor cable is placed at shallow
(52.5 cm), since soil temperatures at these depths
depth in soil subjected to freezing, a precaution
at both locations were similar during the winters.
often taken is to surround the cable with sand.
To investigate the differences in wintertime
The low moisture content of the sand reduces
soil temperature histories that are attributable to
the possibility of ice lens formation and associat-
the presence of a body of sand within the soil,
ed frost heave adjacent to the cable. This is im-
numerical simulations of heat flow in silty soil
portant when it is necessary that the depth of the
with and without a sand inclusion have been
cable remain constant along its length. A related
done. The sand inclusion is 15 cm thick, its top is
consideration is that, should there be any differ-
15 cm deep, and its width is variable. The mois-
ential motion of the cable, the potential for abra-
ture content of the soil and the temperature his-
sive or puncture damage to the cable is less
tory of the soil surface are varied to explicitly
when a rock-free material such as sand sur-
examine their influences on soil temperature pro-
rounds the cable.
files. The surface boundary conditions are a
If the cable is intentionally constructed to leak
sequence of freeze and thaw episodes from Octo-
radio frequency energy as part of an electromag-
ber to April (BC1); a freezing episode lasting
netic sensor system, then sand is an advanta-
from December into April, with a minimum soil
surface temperature of 4C (BC-Warm); a freez-
geous burial medium because of the low electri-
cal conductivity associated with its low moisture
ing episode from December to April, with a mini-
mum soil surface temperature of 9C, that is
content. The higher the electrical conductivity of
the medium, the greater the loss of the electro-
interrupted by a January thaw during which the
soil surface warms to 0C (BC-Cold); and a freez-
magnetic signal with distance from the cable,
and the less effective the sensor system is.
ing episode from January into March, with mini-
mum surface temperature of 1C (BC-Warmest).
Because the unfrozen moisture content of soil
Previous Page
