On Feb 1, 6:54*pm, John Hall wrote:
In article
,
*Dawlish writes:
On Feb 1, 4:51*pm, Dawlish wrote:
Presently we've got the 00z ECM showing easterlies at T240, the 12z
gfs showing easterlies at T240 and the GEM showing westerlies at T240,
with the UKMO showing easterlies at T144. All these 4 are just about
the *exact opposite* of what they were showing at this time yesterday.
This is Olympic class *tumbling, never mind somesaults! I don't think
I can ever remember this happening and two up the models (gfs and ECM)
have had upgrades during January. *O
Fascinating, fun and pretty useless for getting a 10 day forecast
correct!- Hide quoted text -
- Show quoted text -
.............and thinking about that over a rather nice tea, I think
that yesterday, each model was showing almost the exact opposite of
what it was the day before that, which actually means that each model
is now showing, to a greater, or lesser extent, what it was 2 days
ago!
I think you are on to something Len, but instead of daily, maybe we
should restrict ourselves to dipping into the models once every 2 days
(or restrict model runs to once every 2 days, thus saving a fortune,
as Ronald wished). If we'd done that, we'd never have had any of this
silly and unnecessary excitement.
*))
At last we seem to have some agreement between the main models, with
today's 12Z operational runs from GFS, ECMWF and UKMO all suggesting
that an easterly will set in as early as this weekend though, at least
initially, it doesn't look like being particularly cold.
A new feature is that from about a week out the 12Z GFS has a
substantial number of very cold ensemble members, with 850mb temps for
London at or below -10C. They're still in the minority, though, with the
ensemble mean no more than a degree or two belong the long-term mean.
--
John Hall
* * * * * *"Acting is merely the art of keeping a large group of people
* * * * * * from coughing."
* * * * * * * * * * * * * * * Sir Ralph Richardson (1902-83)- Hide quoted text -
- Show quoted text -
There's some agreement this morning John; both the gfs and the ECM now
agree on us *not* having easterlies at 10 days!! shakes head
The gfs 00z ensemble mean suddenly looks mild right out to the edge of
reality - with a fairly tight plume out past 10 days and only one
member below the 30-day mean for a 4-day period from 7 Feb to 11 Feb.
Staggeringly mild compared to last night's ens.
http://www.wzkarten3.de/pics/MT8_London_ens.png
This was the ECM ensembles from the 12z yesterday, with the gfs 12z
operational close to the mean right the way through.
http://www.meteogroup.co.uk/uk/home/..._forecast.html
In years of Internet model watching and a dreadful case of OCD in
monitoring them ( *)) ), I think I can confidently say that I've
never seen such a level of inconsistency, from all the major models,
over a 4-day period. That's why 10-day forecasting is *so* difficult
and is simply not possible, with reasonable accuracy, most of the
time.