Submitted to: Journal of Hydrology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/3/1997
Publication Date: N/A
Interpretive Summary: Infiltration rates in frozen soil may be extremely low due to pore block- age by ice, leading to accelerated runoff and erosion. It is difficult to test our understanding of the processes that take place during infil-tratio o frozen soil because incoming water may freeze while resident ice may melt. We used air flow measurements instead of water to determine the permeability of frozen soil. The measurements were then related to water using physical properties such as density and viscosity. Working first in unfrozen soils, we showed that air flow can be described with the same equations as water. Then we used equations for soil water move- ment to predict air flow. This worked if we considered that some air becomes trapped when soil is wetted. When those equations were applied to frozen soils we assumed that the expansion of ice would cause a reduction in permeability. The measured reduction in permeability was much greater than our predictions. We think that the reason for this is that when the soil water freezes, the largest pores freeze first, which creates a gradient for water movement from the smaller pores to the larger pores. This is called internal redistribution and can result in a large reduc-tion ermeability. We concluded that use air to measure permeability works well in frozen and unfrozen soil but that we cannot predict the effect of freezing on permeability based on the expansion of ice alone.
Technical Abstract: Infiltration rates in frozen soil may be extremely low due to pore blockage by ice, leading to accelerated runoff and erosion. It is difficult to test models of ice blockage effects during infiltration of water because both infiltrating water and resident ice undergo phase changes. We used air permeability, which can be measured without phase changes, to test the applicability of currently used soil hydraulic models to the simulation of ice blockage effects. We confirmed that Darcy's equation described air flow for both dry and moist soil con-ditions. It was then shown that a combination of the van Genuchten soil-water retention equation and Mualem's hydraulic conductivity model, modified to account for the presence of trapped air, described the air permeability of unfrozen soils over a range of soil-water contents. However, when this model was extended to frozen soils, the measured reduction in air permeability in frozen soil was much greater than the model predicted. It appears that hydraulic models can be used to describe the conductivity of the air-filled porosity but that processes other than expansion of water when frozen, possibly internal redistribution of water, affect the conductivity of the air-filled porosity. Until freezing processes are better understood, calculations of infiltration based on soil water content alone are subject to large errors.