In order to assess the impact of seasonal active layer thaw and thermokarst on river flow and turbidity, a gauging station was installed near the mouth of the Sheldrake River in the discontinuous permafrost zone of northern Quebec. The station provided 5 years of water level data and 3 years of turbidity data. The hydrological data for the river showed the usual high water stage occurring at spring snowmelt, with smaller peaks related to rain events in summer. Larger and longer turbidity peaks also occurred in summer in response to warm air temperature spells, suggesting that a large part of the annual suspension load was carried during midsummer turbidity peaks. Supported by geomorphological observations across the catchment area, the most plausible interpretation is that the rapid thawing of the active layer during warm conditions in July led to the activation of frostboils and triggered landslides throughout the river catchment, thus increasing soil erosion and raising sediment delivery into the hydrological network. These results indicate that maximum sediment discharge in a thermokarst-affected region may be predominantly driven by the rate of summer thawing and associated activation of erosion features in the catchment.

Not all places may in fact warm under global change scenarios. This paper presents an example of climatic cooling from northern Quebec, Canada. Ground temperature measurements along the southern shore of Hudson Strait, northern Quebec indicate a cooling trend over the last seven years (1987-93). Long-term air temperature records show that this area has actually experienced continuous cooling for more than 40 years. Related studies suggest that the cooling is likely to continue due to freshening of subpolar water in the North Atlantic and Labrador Sea area. A one-dimensional geothermal model was used to simulate the effect of continued cooling on permafrost thermal regime in Salluit, northern Quebec. The results show that, if the climatic trend continues for the next 50 years, the thickness of the active layer would decrease by 30 cm (from 2.3 to 2.0 m) in gneiss and by 20 cm (from 1.3 to 1.1 m) in till. Permafrost temperature at 20 m depth would decrease by 0.65 degrees C. Under the cooling scenario, rate of permafrost creep and slope activities would be reduced. Ice-wedge regrowth would continue, and the buried ice wedges may even become reactivated. The results also indicate that regional snow precipitation data cannot be used directly in predicting ground thermal regimes.