Factors controlling the magnitudes of, and short-term variations in, the potential temperatures of the snow surface and the air at the height of 2 m θS and θ2 m over Arctic sea ice in winter are analysed. The study addresses the winters of 1986–1987 and 1987–1988, and is based on the temperature, wind, and cloud observations made by Russian drifting ice stations. It also relies on the ERA40 re-analyses of the European Centre for Medium-Range Weather Forecasts, which were utilised to calculate the lateral heat advection at the sites of the ice stations. The cloud cover and wind speed were more important than the heat advection in controlling the magnitudes of θ2 m and θS, while on a time scale of 24 h, during steady forcing conditions, the heat advection was the most important factor affecting the changes in θS and θ2 m. During changing conditions, and considering individual factors separately, the monthly mean 24-h temperature changes were less than ± 5 °C: the effect of the cloud cover was the largest, and that of the heat advection was the smallest. When simultaneous changes in the three factors were analysed, the seasonal mean temperature changes were even of the order of ±15 °C, with the strongest warming events exceeding 35 K in a single day. The difference θS − θ2 m reached its lowest seasonal mean values during conditions of clear skies (−1.3 °C), light winds (−1.3 °C) and warm-air advection (−0.8 °C). θS and θ2 m followed each other closely, even during major synoptic-scale temperature variations.
Air temperature Arctic Cloud forcing ERA40 re-analyses Heat advection Sea ice