What was done
The authors utilized a statistical procedure known as wavelet transform to analyze the relationship between the long-term variability of winter European sea level and large-scale atmospheric circulation as represented by the North Atlantic Oscillation (NAO) and Arctic Oscillation (AO) winter indices over the past 150 years.  Specifically, they sought to determine the strength of the relationship between sea level and NAO/AO and how stable that relationship was over time.

What was learned
Results indicated that the NAO and AO explained between 10% and 35% of the variance in winter mean sea level among the dozen or so stations examined.  However, the results also showed that the relationship "is not stable and varies considerably over time."  Correlations were weaker between the variables during the 19th century as opposed to the 20th century, for example, with the highest correlation coefficients experienced during the last 30 years.

What it means
Sea level is a valued indicator of climate change, and model scenarios suggest that the rate of sea level rise will accelerate in the future due to direct (thermal expansion of the ocean) and indirect (melting of glaciers) responses to CO2-induced global warming.  In light of what they learned from their analyses of winter European sea level and NAO and AO winter indices, however, Jevrejeva et al. conclude that "relationships detected for relatively short time periods should be carefully used in climate reconstruction or prediction," as things other than thermal expansion of the ocean and melting of glaciers - of which large-scale atmospheric circulation reorganization is but one - may also have a significant impact on sea level.