led to shrinkage of nearby veins, thus increasing the
gain more water outflow and to use a pipette with finer
spread of vein width sizes. Blockage of water flow by
entrapped air is another likely cause of reduced ice per-
tingly accomplished for Sample A, would also increase
meability. Both of the causes presume that air enters
the water volume. Growing a larger sample and dis-
the veins. Redistribution of impurities would lead to
carding the contaminated ice would ensure greater ho-
vein shrinkage and growth, respectively, in contami-
mogeneity.
nated and purer regions of the samples.
LITERATURE CITED
CONCLUSIONS
Black, P.B. (1986) RIGIDICE model of frost heave and
From the relative success of the MC experiment and
its input functions. Ph.D. thesis, Cornell University,
from physical evidence of a large, bubbly region within
Ithaca, New York.
the rotted ice, we conclude that air did permeate Sample
Black, P.B. (1990) Three functions that model empiri-
A. Although water volumes were somewhat affected
cally measured unfrozen water content data and pre-
by temperature fluctuations, our data suggest that ice
desaturates and rewets under pressure in a fashion simi-
Regions Research and Engineering Laboratory, CRREL
lar to unsaturated soil. Air entry and water re-entry pres-
Report 90-5.
sures generally supported our theoretical estimates (eq
Black, P.B., and A.R. Tice (1989) Comparison of soil
5a and 5d) and predictions from the ClausiusClapeyron
freezing and soil water curve data for Windsor sandy
equation (eq 6). The observed MC curve for this one
loam. Water Resources Research, 25(10): 22052210.
experiment was highly hysteretic, in line with our ide-
Colbeck, S.C. (1973) Theory of metamorphism of wet
alized flow path model of alternating vein segments
snow. U.S. Army Cold Regions Research and Engineer-
and nodes. Because of small water volumes and inter-
ing Laboratory, Research Report 313.
vening melt episodes, MC results were much less con-
Colbeck, S.C. (1976) Water flow through veins in ice.
clusive for Sample B. The few usable data points, how-
U.S. Army Cold Regions Research and Engineering
ever, are again in line with theoretical estimates. Ancil-
Laboratory, CRREL Report 76-6.
lary information on flow rates through Sample B cor-
Colbeck, S.C. (1979) Grain clusters in wet snow. Jour-
roborates the MC runs and supports the hypothesis that
nal of Colloidal and Interface Science, 72: 371384.
air entered the sample. For two runs, we speculate that
Dullien, F.A.L. (1992) Porous Media. Fluid Transport
water re-entry into the side air gap was delayed by three
and Pore Structure. San Diego: Academic Press, sec-
and seven days until the sample was resaturated. For
ond edition, p. 1574.
another, water levels continued to drop when pressure
Harrison, W.D., and C.F. Raymond (1976) Impuri-
was reversed upwards, thus suggesting the continued
ties and their distribution in temperate glacier ice. Jour-
flow of water under tension into unfilled airways.
nal of Glaciology, 15(74): 173181.
We propose that water inclusions reside in the verti-
Jordan, R. (1991) A one-dimensional temperature
ces of air-filled veins and take the form of concave tri-
model for a snow cover: Technical documentation for
SNTHERM.89. U.S. Army Cold Regions Research and
Engineering Laboratory, Special Report 91-16.
a reversal in curvature of the vein wall through melt-
Jordan, R.E., E.LAndreas, and A.P. Makshtas (1999)
ing. The formation of these near-cylindrical conduits
Heat budget of snow-covered sea ice flow at North Pole
has important implications for the permeability of ice
4. Journal of Geophysical ResearchOceans, 104(C4):
in non-laboratory settings. Finer-grained soils under-
77857806.
lying basal ice could draw water from the veins and, in
Lliboutry, L. (1996) Temperate ice permeability, sta-
warm spring conditions, began this process of vein
bility of water veins and percolation of internal melt-
water. Journal of Glaciology, 42(141): 201211.
enlargement. Our laboratory results show that
Mader, H.M. (1992a) Observations of the water-vein
resaturated ice under hydrostatic tension is nearly imper-
meable, with permeabilities on the order of 1 1018
system in polycrystalline ice. Journal of Glaciology,
m2. Low permeabilities may result from air entrapment.
38(130): 333347.
We conclude that moisture-tension experiments on
Mader, H.M. (1992b) The thermal behavior of the
ice are doable and offer an alternative tool for studying
water-vein system in polycrystalline ice. Journal of
ice thermodynamics in the temperate regime. Tempera-
Glaciology, 38(130): 359374.
ture fluctuations were the major drawback in this experi-
Morris, E.M. (1993) A theoretical determination of the
ment and could be stabilized by adding an encapsulat-
characteristic equation of snow in the pendular regime.
ing brine bath around the Tempe cell. We suggest in-
Journal of Glaciology, 36(1123): 179187.
creasing the diameter of the sample and Tempe cell to
Muskat, M. (1937) The Flow of Homogeneous Fluids
13