In message , Norman Lynagh
writes
The medium range models are a continuing source of light entertainment. The 00z
operational runs this morning show the following predictions for the British
Isles for Tue/Wed 17th/18th

ECMWF: Bitterly cold easterlies setting in.
GFS: Very mild SW'lies
GEM: Very strong, cold westerlies with polar maritime air

All that can reasonably deduced from that lot is that, on the basis of current
knowledge, the evolution of the weather pattern that far ahead is
unforecastable to any useful level of reliability at present. Long may that
state of affairs continue :-)


I think that's a bit sweeping. All one can say based on the above is
that, from this particular set of initial conditions, the weather 9-10
days ahead is not forecastable by at least two of the three models. I
suspect that your statement is actually correct, but one would need to
examine a lot more instance to say it with total confidence.

Of course one shouldn't be looking at just the operational runs but at
the ensembles, which might allow a better picture to emerge, but that
involves more work than most of us want to put in.
--
John Hall
"One can certainly imagine the myriad of uses
for a hand-held iguana maker"
Hobbes (the tiger, not the philosopher!)