IPCC Fourth Assessment Report: Climate Change 2007
Climate Change 2007: Working Group I: The Physical Science Basis Landfast Ice Changes

Interannual variations in landfast ice thickness for selected stations in northern Canada were analysed by Brown and Coté (1992). At each of the four sites studied, where ice typically thickens to about 2 m at the end of winter, they detected both positive and negative trends in ice thickness, but no spatially coherent pattern. Interannual variation in ice thickness at the end of the season was determined principally by variation in the amount and timing of snow accumulation, not variation in air temperature. An analysis of several half-century records in Siberian seas has provided evidence that trends in landfast ice thickness over the past century in this area have been small, diverse and generally not statistically significant. Some of the variability is correlated with multi-decadal atmospheric variability (Polyakov et al., 2003).

For the Antarctic, a combined record of the seasonal duration of fast ice in the South Orkney Islands (60.6°S, 45.6°W) has been compiled for observations from two correlated sites for the period 1903 to 1992 (Murphy et al., 1995). The ice duration in these coastal locations is linked to the cycle of pack ice extent in the Weddell Sea, and the duration shows a likely decrease of 7.3 days per decade. This decrease is not linear over the 90-year period and occurs within a strong 7- to 9-year cyclical component of variability over the latter 30 to 40 years of the record. Fast ice thickness measurements have been intermittently made at the coastal sites of Mawson (67.6°S, 62.9°E) and Davis (68.6°S, 78.0°E) for about the last 50 years. Although there is no long-term trend in maximum ice thickness, at both sites there is a trend for the date of maximum thickness to become later at a rate of about four days per decade (Heil and Allison, 2002).