On Dec 24, 12:55*pm, Dawlish wrote:
On Dec 24, 11:32*am, John Hall wrote:





In article
,

*Dawlish writes:
On Dec 23, 11:12*pm, Dave Cornwell wrote:

----------------------------------------
You may be right but surely this does not meet your normally stipulated
model consistency criteria. There is absolutely no agreement from
January 2nd onwards in the ensembles so your normal advice would be to
ignore it.

It's about consistency and agreement between the 2 main models, for me
Dave and both have been rock solid out to T+240. They still are this
morning. I use the ensembles only as a check at 10 days, because the
spaghetti that you find there is not useful, on its own, for
forecasting.

If there is spaghetti at 10 days, then surely that indicates that the
operational run cannot be relied on. Having said which, the most recent
ECM ensemble mean pressure chart is actually in excellent agreement with
the operational run and is showing an unusually small degree of standard
deviation, especially in the vicinity of the UK.
--
John Hall
* * * * * * *"The power of accurate observation is commonly called cynicism
* * * * * * * by those who have not got it."
* * * * * * * * * * * * * * * * * * * * * * * George Bernard Shaw

It is *highly* likely to change and the agreement you see now at 10
days is unlikely to be there at 9 days, or 8 days. It may be, of
course, but it is more likely to be a one-run wonder than it is
something to be relied on. That's why I require a day's operationals
and two models to judge consistency and agreement, John. It sounds
like the MetO forecasters do something similar with the time-lagged
ensembles........but then they give them to the tea boy to present to
the public, in the form of that dreadful 6-15 day forecast, which no-
one can trust at 10-days. There is no measure that tells you when an
ensemble can be trusted at 10 days and I'd love to see someone try to
forecast on a regular basis, and get it right with even reasonable
accuracy. I don't think it is possible. If it was, the MetO would use
it for 10-day forecasting and get the synoptic type correct, on a
basis, regular enough to judge accuracy and justify that accuracy to
their public. They don't. It's as simple as that.

I do wish the MetO would make their ensembles public - after all we do
pay for them - but whatever judgement they use is not accurate enough
to give the public any kind of consistent accuracy at 10 days. I also
wish that the MetO would give some indication of the confidence they
have in forecasts at that distance, but, sadly, they don't seem to set
store by them, for the public.- Hide quoted text -

- Show quoted text -

And nothing now really shows this cold zonal scenario.

http://www.jp2webdesign.co.uk/two/ensembles/