Analysis of a link between fall Arctic sea ice concentration and atmospheric patterns in the following winter

The impact of anomalous fall Arctic sea ice concentrations (SICs) on atmospheric patterns in the following winter is revisited by analysing results for two time periods: the most recent, satellite-era period (1979–2010) and a longer time-period (1950–2010). On the basis of Sept...

Full description

Bibliographic Details
Published in:Tellus A: Dynamic Meteorology and Oceanography
Main Authors: Susanna Hopsch, Judah Cohen, Klaus Dethloff
Format: Article in Journal/Newspaper
Language:English
Published: Stockholm University Press 2012
Subjects:
Online Access:https://doi.org/10.3402/tellusa.v64i0.18624
https://doaj.org/article/26e418140f59414c843a945fc6e2978f
Description
Summary:The impact of anomalous fall Arctic sea ice concentrations (SICs) on atmospheric patterns in the following winter is revisited by analysing results for two time periods: the most recent, satellite-era period (1979–2010) and a longer time-period (1950–2010). On the basis of September SICs for each time-period, an index was constructed which was used to identify anomalous high/low SIC years for both the original, as well as for the linearly detrended sea ice index. Identified years were then used to derive composites for the following winter's monthly atmospheric variables. Mid-troposphere geopotential height composites for winter months are in general reminiscent of the North Atlantic Oscillation pattern with high latitude maximum shifted towards the Barents Sea. Also, lower troposphere temperatures indicate the presence of cooler conditions over the continents during low SIC years. However, differences in the composite patterns are significant only for areas with limited spatial extent. While suggested pathways in previously published studies seem reasonable, our results show that these findings are not yet robust enough from a statistical significance perspective. More data (e.g. provided by longer, climate-quality reanalysis datasets) are needed before conclusions of impacts and feedbacks can be drawn with certainty.