On Tuesday, July 12, 2016 at 3:42:00 PM UTC+1, Norman Lynagh wrote:
Freddie wrote:

Graham Easterling Wrote in message:

Haha i guessed this post would be about point probability
forecasts before I read it!! :-)

At 08:00 this morning we had a shower which dropped 3.4mm in 10mins.. We
are surrounded by showers, with little gap inbetween. The radar
demonstrates this. The MetO Fax chart for 06:00 shows a trough overhead,
there is clearly deep convection but . . .

Chance of rain for Penzance @ 08:00 & 09:00 still shown as 5% (it's now
09:20 by the way - so it's happened). If anyone involved could explain
that, I'd be interested.
It's an automated forecast with no manual input. I would hazard a
guess that the model wasn't forecasting showers at the grid point
nearest you at that particular time. Why it wasn't forecasting
them is the question that needs an answer...

Because it ain't good enough to do reliable forecasting that has a high level
of precision in both time and space simultaneously. At least, that is my
experience for my part of the country. By far the best short-range forecasting
tool is the high-resolution (time and space) radar imagery.

--
Norman Lynagh
Tideswell, Derbyshire
303m a.s.l.
http://peakdistrictweather.org

Absolutely agree. But the 5% chance of rain was at variance with all other aspects of the forecast. I fully realise that this is an automated generalised model forecast, but if the model shows a showery trough bang overhead, how come the rainfall % chance doesn't reflect this? After all the air was seriously unstable, so very impressive Cb, over the sea as well as the land. (SST now 16C)

As it happens, currently yet another heavy shower, the 4th or 5th today - precipitation chance still 5%.

Hopefully like this again on Thursday http://www.cornwallcam.co.uk/

Graham
Penzance