"Keith Dancey" wrote in message
...
In article ,
writes:

If I understand correctly, the Met Office are suggesting that there is
a 2 in 3 chance that the coming winter will have temperatures lower
than the mean for the past ten years. There is also a 2 in 3 chance
that the rainfall will be less than the mean for the past ten years.

To be pedantic, I suspect the Met Office is actually comparing the
expected figures with the median figures for the past ten years, but
that isn't clear in the press release.


(Indeed... it could have been more explicitly worded.)


In support of Jon's posting, I think the Met Office were referring to the
the *long-term" average (ie. a rolling 30-year average) when they talked
about "averages" (because those were the figures they produced in the
forecast).


So, a ~66% chance of being colder than the current 30-year average, but
not
as cold as 1995/96. For Southern England that would indicate a mean
winter
temperature of between 3.5 and 4.5 C.

For other Regions:

Scotland - between 1.8 and 2.7
Northern England - between 2.4 and 3.5
Wales - between 2.8 and 4.2
Northern Ireland - between 3.7 and 4.3



The forecast in
http://www.metoffice.gov.uk/research.../headline.html
says "The balance of probability is for a winter colder than those
experienced since 1995/6.". As far as I can see, it doesn't mention "colder
than the mean" or "colder than the 30 year average". What it does do is
compare 2 recent winters against the long term average and the winter of
1995/96. I would have thought that there should be no confusion as regards
the Met Office forecast.
As of the 24th October (when they issued it) they were expecting a fairly
high probability that this winter would be the coldest since 1995/96.
Whether that turns out to be correct or not we'll have to wait and see.

Victor