Northwest Territories Geological Survey

Ice and all-weather roads are critical lifelines for northern communities and resource development activities. Icings are sheet-like masses of layered ice that form on the surface during the winter by freezing of successive overflows of water. Icing development over ice and all-weather roads negatively impacts their performance, and there can be adverse effects on bridges and culverts. Consequently icings are an ongoing maintenance and safety problem in northern regions. The mechanism controlling overflow timing and behaviour is poorly understood. Overflow timing and stream bed water pressures have been linked to air temperature fluctuations in several regions, but the relation between these two signals is not well defined and the mechanism is unknown. The Shield region contains the highest density of resource development infrastructure in the Northwest Territories, much of which is supported by the Tibbitt to Contwoyto Winter Road (TCWR). The main operational difficulty faced by operators on the overland portions of the TCWR is overflow and icing. The objective of this work is to provide much needed geoscientific insight on the icing process, in order to reduce risks for land-based transportation infrastructure.

Overflows in the subarctic Canadian Shield may be controlled by periodic increases in hydraulic head during open-system freezing. Local observations at the TCWR Portage 23 (P23) overland section have shown that overflow occurs during the period of active-layer freezeback when it is likely hydrologically connected to the upstream lake. Based on our previous work at P23, we installed vertical series of pressure transducers and temperature data loggers to determine the relations between barometric pressure, air temperature, water/ice temperatures, and water/ice pressures during the freezing season. To examine variation in hydrological setting, replicate apparatus were installed in a shallow lake, the lake outlet, and in the riparian zone downstream of the outlet where icing has been observed.

Initial results suggest that during the freezing-season, high frequency fluctuation of absolute pressure in the lake is driven entirely by barometric pressure, with no detectable increase with snow loading. At the outlet, hydraulic pressures behaved similarly to lake pressures. Absolute pressures at the downstream sensors also fluctuated at high frequency with barometric pressure. However, in contrast to the lake and outlet sites, hydraulic pressure increased by up to 2.2 kPa over a two-week period that followed freezing onset. Hydraulic pressures gradually tapered off from peak values until freezing occurred at depth.
The static water level observed at the lake and outlet suggests that the lake is either an entirely open system with water entering and exiting the lake at the same rate over the winter, or alternatively, that the lake is entirely closed with respect to water flow. The latter is supported by our observation of decreasing hydraulic pressures at the downstream site. The initial increase in hydraulic pressures observed at the downstream site indicate the semi-confined nature of the freezing riparian area. Further signal analysis is being undertaken to investigate the effects of freezing front progression on hydraulic pressure at the downstream site.