On Nov 5, 9:52*am, Graham P Davis wrote:
Dawlish wrote:
You mean there's a statistically significant relationship between
Atlantic SSTs in October and the nature of the coming winter?

Sorry, got carried away and missed your reference to "seasonal". It's true
that there are statistically significant relationships between Atlantic SSTs
and the following month's pressure anomalies but they'd weaken after that as
they'd be less chance of the SST anomaly persisting.

--
Graham P Davis, Bracknell, Berks., UK. *E-mail: newsman not newsboy
"I wear the cheese. It does not wear me."

Yes, I agree, Graham. There's also the relationship between May SSTs
and the sign of the NAO, which, although the correlation is fairly
weak at about 45%, hindcast forecasting produces a much better
accuracy of about 66% and both values are significant at the 95%
level. That's not 66% correct hindcast of our UK winter weather, of
course, as the sign of the NAO is by no means perfectly linked to that
winter weather, but I think that what the MetO are doing in this area
is very interesting. One would think that SSTs would be coupled
closely to the state of the atmosphere above, but the effects of SSTs
may also be thought to not persist. The MetO NAO research on May SSTs
points otherwise with the latter, however and other reseach on, if if
I remember rightly, particular SST Atlantic tripoles (seas off
Newfoundland , seas off Norway and off our SW approaches) may also
give clues as to the weather in the coming season, but there's nothig
that can really aid seasonal forecasting AFAIK, at present.