"Col" wrote in message
...
Dartmoor Will wrote:
"Col" wrote in message

Agreed.
If you don't know, then say so. If the odds of it being wet in two
weeks' time are 50:50 then don't commit to anything, it's just
guesswork. This may lead to claims of 'They don't know what they're
bloody
doing' but better in the long run to take this cautious approach
than nailing your
colours to the mast and end up being spectacularly wrong.


If it really was 50-50 then the MetO would say something more
definite as 50-50 does not mean "I don't know" but there is a 50%
chance of a situation happening which is useful information. I
suspect that the ECMWF ensemble products are giving lots of possible
Lamb types as possibilities therefore there is no useful signal. I.e.
10 lots of 10% is worth nothing.

A 50% risk of a shower for the day ahead is a useful forecast
because it allows people to make judgements at a very basic level
like whether to take a brolly to work.
But a 50% chance of a week as a whole being seen as 'wet' or 'dry'
doesn't to me provide much in the way of useful information beyond
'we don't really know what is going to happen'.

A week ahead it would be useful Col as it would mean you would have to
prepare for rainy weather, a 95% forecast of dry would mean you could relax
and, depending on your cost/loss ratio plan accordingly e.g. just pack a
brolly or take nothing if you are prepared to get wet. My point is that the
MetO use the whole range of probabilities from 0 to 100% and saying 50%
chance of rain does not mean they don't know it is saying that there is a
50% chance of rain and, depending on your circumstances, you can make use of
that. If I was planning a walk in a benign area a forecast of 50% chance of
rain means that I pack waterproofs (the cost) a forecast of 10% chance of
rain means that I wouldn't as it would not be serious (the loss). A walk in
a more challenging area would mean me packing waterproofs whatever.

Will
--