Journal of Cold Regions Engineering

This paper presents a design approach to analyze the effects of lateral loads on piles embedded in a layered frozen soil profile. This design approach makes use of a commercially available computer program that uses $p\text{-}y$ curves to model lateral pile loading and soil interaction. Shear strength and strain criteria for both fine-grained and coarse-grained frozen soil are presented for use in constructing $p\text{-}y$ curves. These strength criteria are parameterized based on frozen soil temperature and load duration. Inuvik lateral pile load test measurements and deflections calculated using the strength criteria are compared. This comparison indicates that reducing shear strengths by a safety factor of 1.22 will ameliorate most unaccounted-for variations in soil profile. Deflections and moments are also calculated using the strength criteria for two Prudhoe Bay, Alaska, lateral pile load cases.

Consider that the Mable Creek silt in central Alaska liquefied during the November 2002 earthquake. Subsequently, frozen samples of these fine-grained soils were retrieved for laboratory study. This paper presents the influence on dynamic properties by temperature changes from below freezing to near freezing, and by freeze-thaw cycles. Sample temperature change was simulated by gradually applying the increasing temperatures of $–0.2$, 0.5, 1, 5, and 24°C. Sample seasonal climate change was simulated by applying 1, 2, and 4 freeze-thaw cycles. Tests on specimens conditioned at 0.5 and $–0.2°\mathrm{C}$ through different thermal conditioning paths were also performed. Determination of the soil’s dynamic properties was investigated using triaxial strain-controlled cyclic tests. The shear modulus was found to decrease when the temperature increased from near freezing to above freezing; however, the damping ratio reached a maximum value when temperature was at or near freezing. Applying 1, 2, and 4 freeze-thaw cycles resulted in an increase in both the dynamic shear modulus and the damping ratio. Thermal conditioning paths with the same target near-freezing temperature were found to impact dynamic properties.

Snow falling on roads is rapidly compacted into a hard crust which is difficult to remove, and which may become slippery at temperatures near melting. To avoid this, chemicals are often spread in a measure called anticompaction to keep the snow plowable until it can be removed by highway maintenance personnel. The use of chemicals is, however, harmful both to the environment and to structures and vehicles, meaning that more efficient anticompaction measures are needed. Testing the efficiency of an anticompaction measure in the field is an extensive job, and as an alternative different lab tests have been developed. Unfortunately these correlate poorly with field tests and one of the causes for this has been hypothesised to be the type of snow which is used. In this paper the effect that different degrees of metamorphism had on the hardness of snow immediately after compaction was studied. Three different snow types were used, as follows: (1) dendritic snow, (2) granular snow, and (3) snow consisting of large round grains. The snow types were characterized by specific surface area, snow type classification, and micrographs. The dendritic snow was allowed to coarsen between 0 and 168 h to increase the number of different snow morphologies. The different snow types were compressed at $-6°\mathrm{C}$ to a density of $500\mathrm{kg}/{\mathrm{m}}^3$, after which their hardness was measured with a micropenetration test. The results show that the compact hardness decreases rapidly with the degree of metamorphism; after 24 h the hardness was 80% of that of fresh snow and snow left to coarsen for 1 week before compacting only had 40% of the fresh snow hardness. The granular snow was still weaker and the snow consisting of large rounded grains did not consolidate at all, but remained as a collection of loose particles. This shows that when comparing the effect of different chemicals on anticompaction it is important that the experiments are performed on a snow similar to what can be expected to fall on a road. Old natural snow or granular artificial snow are likely to give misleading results.