Key points are not available for this paper at this time.
A monthly storm-track climatology is derived from monthly maps of cyclone tracks for the winter season, October through March, averaged over 23 years, 1957/58–1979/80, for a 2° latitude×4° longitude grid bounded by 51°N, 65°N, 157°W and 171°E. There is a decrease in the number of cyclones with latitude in all months and division into two storm tracks, one propagating north-northeast along the Siberian peninsula and one entering the southern Bering Sea on a northeasterly course and either curving northward into the central Bering Sea or continuing parallel to the Aleutian Island chain. Monthly average ice extents are established for February and March 1958–80 along a line from Norton Sound southwest toward the ice edge, perpendicular to the average maximum extent. Comparison of composite cyclone charts summed over the winter season and over the five heaviest and five lightest ice years shows a shift in cyclone centers toward the west in light ice years. The correlation between maximum seasonal ice extent and the difference between the number of cyclone centers in the eastern minus the western part of the basin over each winter season is 0.71. The relation of sea ice extent and the location of cyclone tracks is consistent with previous observations that advance of the ice edge in the Bering Sea is dominated by wind-driven advection and that southerly winds associated with cyclone tracks to the west inhibit this advance. These results indicate that the interannual variability in seasonal sea-ice extent in the Bering Sea is controlled by an externally determined variation in storm-track position related to large-scale differences in the general circulation. A skewed distribution of ice extents toward heavy ice years, however, suggests the possibility of an oceanographic constraint on the magnitude of extreme seasonal ice extents, such as the inability of melting ice to cool the mixed layer beyond the continental shelf to the freezing point or the increased influence of the northwestward flowing, continental slope current.
Overland et al. (Fri,) studied this question.