Here's why I'm
really intrigued by this bout of -NAO we're seeing the next few weeks:
Does early winter NAO behavior tell us anything about its behavior later in winter (Jan-Mar) during El Niños?! ?
The data says yes! ?
The relationship is positive (they reinforce each other) & it's significant-very significant.
This means that a more negative/or less positive NAO in December begets an even more negative NAO in January-March when an El Niño is present. This is objectively a very good early sign for cold/snowy winter in the SE US & mid-Atlantic states.
Below is the December vs January-March averaged 500mb correlation (R) for El Niño winters only, since 1855 in NOAA's 20th Century Reanalysis Version 3 (NOAA 20CRv3). I applied a 30-year sliding base period that lagged by 20 years to . Notice that areas contoured and stippled in white are where the correlations are significant at the 99% level. This basically says that there's a < 1 % probability that the correlation between December NAO & January-March NAO in El Niño winters occurs simply by chance.
Dec-Mar 500mb correlation w/ the -NAO (for a sanity check & comparison purposes).
I've also looked at MSLP data in NOAA 20CRv3, because it's directly assimilated into the reanalysis model & thus tends to be the most reliable variable to look at. Not surprisingly, it supports the 500mb data w/ very significant correlation between Greenland & Scandinavia. This corresponds pretty well w/ the SLP loadings for the -NAO too (like the 500mb data).
Dec-Mar MSLP correlation w/ the -NAO.