... the GFES output certainly shows the ensemble members clustering
(tightly) around 5C above the long-term average at 850T. This for our
location:-

http://www.wzkarten.de/pics/MS_-251_ens.png

Martin.

ronaldbutton wrote:
I certainly understand the red line on that graph Martin,what as to the rest
of the stuff ...bloody spaghetti !.

.... indeed, though these aren't 'spaghetti', they're different animals!

.... essentially, each of the *thin* lines is an individual output based
on slightly perturbed initial conditions from the Control ('Kontroll')
run, which it also shown - the control is a lower resolution version of
the operational or main run. The deterministic/operational run (thick
green line / 'Hauptlauf') is also shown. If they *all* cluster tightly
together, as they do for at least the first 5 or 6 days, then you can
have a high confidence that the output (in this case the 850 hPa
temperature) is going to behave as shown. The lines at the bottom are
for precipitation BTW.

These output are amongst the first I look at each day - until you know
whether the model atmosphere is potentially 'focused' & knows what it is
doing (close spacing of plumes) or 'noisy' (wide/widening spacing), then
it's not much good slavishly watching the operational stuff - it could
be 'pants' :-)

Martin.


--
Martin Rowley