January 22nd 10, 07:02 PM
posted to uk.sci.weather
|
|
external usenet poster
|
|
First recorded activity by Weather-Banter: Dec 2009
Posts: 31
|
|
I've gone right off gfs ;-)
non existant forecast #293982377424724
chicken
Dawlish wrote:
On Jan 22, 6:07 pm, "Will Hand" wrote:
"Keith(Southend)" wrote in message
...
Gfs seems to be consistently not keen on cold weather in it's runs lately.
I know Will's mentioned it's gone off the rails a little, but I just
wonder whether it *may* turn out to be right?
I shall have to turn to another model to get my fix!
--
Keith (Southend)
http://www.southendweather.net
e-mail: kreh at southendweather dot net
Keith, and others. The runs just don't look realistic as well as being out
on a limb. I mentioned several weeks ago that I felt it's handling of lows
on occasion was weird, some of the isobars in that 12Z run look shaky and
too wobbly it's almost as if there are numerical instabilities present in
the calculations. Sometimes its OK, but recently it has been unrealistic.
The 12Z GFS OP run is once again a warm outlier.
UKMO has been pretty good of late!
Will
--
I'd counsel not to dismiss the gfs, Keith or Will, or you might just
miss the time when it is likely to be right and UKMO is no use at 10
days. The models show what they show Keith. I don't think they have a
bias towards mild, or cold, but if you like either of those two in
winter and a model doesn't show what you are hoping for, I appreciate
that could get frustrating. In forecasting at 10 days I can't afford
to be looking for a particular weather type, or it would introduce a
bias that would have wrecked my success percentages! That's why the
winter-cold-forecasters never actually have long-term track record of
success, but, ironically, garner all sorts of praise as forecasters
(from other coldies, almost always) when a cold winter, like this
colder half-to-two-thirds-of-a-winter (so far), happens along!
3 runs with high pressure close to, or over us at T+240. Just enough
to perk up my interest!
|