Assessment of possible changes in air temperature and sea-ice extent in the Arctic Seas in the twenty-first century

Abstract

Attempts at super-long forecasts of ice extent for some regions of the Arctic Ocean have a long history. They have mainly focused on the sub-Atlantic part of the Arctic, because long series of ice observations are available for that area. Maksimov (1955) made the first forecast, based on space-geophysical factors and, primarily, a century-long cycle of solar activity. According to this forecast, maximum ice extent in the region was expected in 1990. This forecast, however, was not correct (nor were other forecasts based on the “100-year” cycle of solar activity). In updating Maksimov’s forecast, Latukhov and Sleptsov-Shevlevich (1995) based their work on the relationship of ice extent and the magnetic perturbation index K(inp), which also reflects the 100-year cycle; they predicted that maximum ice extent in 2000–2020 would be comparable to the conditions of the early twentieth century. As we can see now, this prediction was also incorrect. The results of reconstructing ice extent changes in the eighteenth and nineteenth centuries presented by these authors suggests some doubt about the validity of their methodology. According to their results, the second half of the nineteenth century was distinguished by decreased ice extent in the sub-Atlantic region of the Arctic. However, data collected by Norwegian scientists (Vinje, 2000) showed that during this period, increased ice extent was observed there.