Interannual variations of the thermal regime of the active layer and near-surface permafrost in northern Alaska

Air, ground surface, active layer and permafrost temperature measurements were made every four hours from 1986 through 1992 near Prudhoe Bay West Dock (WD), Deadhorse Airport (DH) and Franklin Bluffs (FB). Mean annual air temperatures (MAAT) were nearly identical (-12.6 °C at WD, -12.9°C at DH, -12.5°C at FB) while the amplitudes increased from the coast inland (17.2°C at WD, 18.4°C at DH, 20.8°C at FB). Interannual variations were similar (3.2°C at WD, 2.9°C at DH and FB). Mean annual ground surface temperatures (MAGST) were similar at FB (-5.8°C) and DH (-6.2°C) and much colder at WD (-8.8°C) with amplitudes of 14.5°C at WD, 13.5°C at DH and 15.2°C at FB. Interannual variations were about the same at WD (3.8°C) and DH (3.9°C) and much smaller at FB (1.4°C). Differences between MAGST and MAAT were about the same at FB (6.8°C) and DH (6.7°C) and much smaller at WD (3.9°C). The results suggest that the snow cover was thinner at WD and/or its properties were significantly different compared with the inland sites. Mean annual permafrost surface temperatures (MAPST) were the same at FB and DH (-6.8°C) and much colder at WD (-9.1°C) with interannual variations of 2.7°C at WD, 3.0°C at DH and 1.8°C at FB. Thermal offset (MAPST - MAGST) was about the same at WD (-0.7°C) and DH (-0.67°C) and larger at FB (-1.03°C). Generally, thawing at the ground surface began earlier at FB than at DH and WD. Average duration of thawing (99 days at WD, 104 days at DH, 107 days at FB) and dates of the start of ground freezing (16 September at WD, 18 September at DH, 17 September at FB) were similar at all three sites. Average durations for freezing the active layer were similar at FB (62.5 days) and DH (68 days) and much less at WD (42 days). Active layer thicknesses increased from the coast inland (range of 0.21 m to 0.46 m at WD, 0.42 m to 0.69 m at DH, 0.57 m to 0.72 m at FB) and showed systematic temporal variations. These variations in active layer thicknesses may have significant temporal and spatial effects on the carbon balance of the tundra.