View Single Post
  #20   Report Post  
Old 20-07-2017, 08:44 PM posted to uk.rec.gardening
James Wilkinson Sword[_2_] James Wilkinson Sword[_2_] is offline
external usenet poster
 
First recorded activity by GardenBanter: Jan 2017
Posts: 37
Default "% Chance of rain"

On Thu, 20 Jul 2017 13:20:28 +0100, Asha Santon wrote:

On 2017-07-20 11:58:05 +0000, James Wilkinson Sword said:

On Thu, 20 Jul 2017 12:44:54 +0100, Asha Santon wrote:

On 2017-07-20 11:13:43 +0000, James Wilkinson Sword said:

What do you believe weather forecasters mean when they say "40% chance
of rain"?

1) It will rain for 40% of the day.
2) There's a 40% chance it will rain at some point.

I googled it and apparently it depends on the forecaster? 1) and 2)
are completely different meanings.

It means there is a 40% chance that it will rain in the area to which
the forecast applies and during the period (often a day) specified.

It never means (1) and does not depend on the forecaster and is not a
matter of belief. Nor, happily, does it depend on google.


So it's utterly useless then? 40% chance of rain could mean there's a
40% chance I'll get a light shower for half an hour at some point, or
it could mean there's a 40% chance it will bucket down all day. They
should estimate the AMOUNT of rain that is going to fall, eg so many mm.


No, not useless. Some forecast services offer the probability for each
hour of the day which is more than sufficient.
It has happened on occasion that I have looked out of my front window
and it has been raining while out of the rear window it has been sunny
and dry. There is no way for a forecast to be that precise and 50%
chance of rain would be a perfect forecast for that.
Unless it is an unusual event, there is no way to predict the amount of
rain which will fall on any given occasion. That is why they don't do
it.

If the risk of rain is say 90% then a forecaster may well say it 'will'
rain but with a island weather, it is seldom so easy.

All forecasts are verbalised accounts of probabilities (or risks) of
events happening and that is all there is. The science does not exist,
and probably never will, to accurately predict wat are virtually random
events.


Surely they can make a reasonable guess as to how much rain will fall throughout the day. That would be far more useful. Like:
"Thursday will be spitting on and off for about 3 hours throughout the day" or
"Thursday will probably have a couple of heavy showers lasting 30 minutes each."

--
More people in the UK are injured by standing on upturned mains plugs than by electric shocks.