Found a possibly interesting correlation with "strength" of eQBO in the Autumn, and the following Winter weather patterns.
I decided to run a composite for three types of QBO. In each category I used years that began with a west-based QBO as we've had this year.
1) Years that trended to a weak eQBO (0 to -10) into the Autumn (Years 1981, 1976, 1967, 1986)
2) Years that trended to a moderate eQBO (-10 to -20) (Years 2009, 2000, 1991, 1962)
3) Years that trended to a strong eQBO (below -20) (Years 2014, 1979, 1972, 2007)
So I found that for the weak to moderate Autumn eQBO's, the composite for the Winter pattern shows a significant High Pressure anomaly over Greenland and to the north of the UK, with Low Pressure more a feature to the south, resulting in a cold/snowy Winter set up. 1991/92 is the only Winter that fails the test, but even this Winter produced some severe cold in the early part of Winter.
But surprisingly, the years that trended to a strong Autumn eQBO led to strongly zonal and mild Winters, in fact the complete opposite to the effect of a weak/moderate eQBO!
I also noticed that in years that started with a wQBO, and progressed to a strong Autumn eQBO, there was a rapid transition during the following Winter back to neutral/weak eQBO and into wQBO territory by soon after Winter. The only example I can find of a strong Autumn eQBO lasting into the Winter was in 2014/15, and that was a strongly zonal Winter!
Now interestingly, 2017 so far seems to be going down the moderate eQBO route, and normally by now if it was going to be strong, we would have already seen very low figures being returned on the QBO index (sub -20). Years that develop a moderate eQBO in the Autumn tend to stick with that through the Winter, so it does look like that the Winter 2017/18 will have a much better chance than normal of developing a colder snowier regime. Not a guarantee of course, as 1991/92 has shown, but clearly is a good sign for cold.
Originally Posted by: Andy J