Interannual variations in the opening date of the Prudhoe Bay shipping season: links to atmospheric and surface conditions

• Published in: International Journal of Climatology
• Main Authors: Drobot, S. D., Maslanik, J. A., Anderson, M. R.
• Format: Article in Journal/Newspaper
• Language: English
• Published: Wiley 2008
• Subjects: Bering Sea; Chukchi Sea; Beaufort Sea; Chukchi; Prudhoe Bay; Sea ice
• Online Access: http://dx.doi.org/10.1002/joc.1725; https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1002%2Fjoc.1725; https://rmets.onlinelibrary.wiley.com/doi/pdf/10.1002/joc.1725

Abstract This paper examines interannual variability in the opening date for the Prudhoe Bay shipping season (1953–2005), considers how variations in antecedent sea‐ice and atmospheric conditions influence the opening date, and then develops a forecasting technique to predict whether the opening date will be early, normal, or late. Analysis of antecedent sea ice and atmospheric conditions indicates that there are significant differences in the Bering Sea ice cover as early as February in years preceding early versus late opening dates. In particular, prior to early opening years, the sea‐ice cover in the southern Bering Sea is reduced in February, and as the season progresses, sea‐ice concentrations in the central and northern Bering Sea remain lower in years preceding early opening dates. Analysis of accumulated freezing degree days (FDDs) also suggests that temperatures are warmer over a broad area, ranging from the Bering Sea through the Chukchi Sea and the Beaufort Sea, in winter and spring months preceding early opening dates. Although the warmer temperatures may be related to reductions in sea ice in the Bering Sea, our results suggest that 500 hPa flow patterns and sea level pressure (SLP) patterns play are a larger role in controlling accumulations of FDD in the Chukchi and Beaufort seas. Finally, an ordinal regression model is presented to forecast whether the opening date will be early, normal, or late. On the basis of April sea‐ice concentrations and accumulated FDD, the model correctly forecasts the opening date class in 32 of the 53 years. Copyright © 2008 Royal Meteorological Society