The national figures, up to August 31, are surprisingly uninteresting
in this respect:
England&Wales 1356h so far (annual norm 1530)
Scotland 1102h so far (norm 1276)
N.Ireland 1145h so far (norm 1403)
Irish Republic 1173h so far (norm 1387)
Philip Eden

Which prompts the question: How do you actually arrive at a 'national'
figure? For example, if Sheffield gets on average 1230 hours, London 1494
and Bognor Regis gets 1750. how do you get to the 'norm' of 1530 for
England? And if Glasgow gets 1400 and Edinburgh 1351, where does the 'norm'
of 1276 for Scotland come from?

Just curious

Anne

Which prompts the question: How do you actually arrive at a 'national'
figure? For example, if Sheffield gets on average 1230 hours, London
1494
and Bognor Regis gets 1750. how do you get to the 'norm' of 1530 for
England? And if Glasgow gets 1400 and Edinburgh 1351, where does the
'norm'
of 1276 for Scotland come from?

Fair question, and the sort of thing I usually omit to explain, basically
because I'm playing about with these numbers on pretty much a daily
basis and I forget that other people aren't as familiar with them.

For England and Wales I use six regions, each of which has several
longstanding reporting stations (on average six in each region). The
total sunshine is converted into a percentage of the 30-year mean
which means that the effect of the gradual loss and replacement of
stations over a very long period of time is minimised, and you can get
a reasonably homogeneous series extending back to the dawn of
history. The mean of these stations' annual (or monthly) percentages
provides a regional mean which is then converted back to a total
number of sunshine hours. These are then combined using a formula
which weights each region according to the fraction it comprises of the
geographical area of the whole of England and Wales.

Same for Scotland (divided into three regions), for the Irish Republic
(two regions), and Northern Ireland (one region).

And rainfall is done in the same way.

Philip Eden

Thanks Philip, that is very interesting. Are these 'regions' chosen for
having a comparable climate or on some other basis? How do you make sure
that the stations are 'representative'? There isn't a lot of similarity
between the 30-year averages for Kinlochewe and Kinloss, for instance - the
driest month at Kinlochewe being on average almost twice as wet as the
wettest month at Kinloss, to quote but one figure. (Guess why, having lived
near Kinlochewe, I now prefer to live nearer Kinloss!)

Anne

"Anne Burgess" wrote in message
...

Thanks Philip, that is very interesting. Are these 'regions' chosen for
having a comparable climate or on some other basis? How do you make sure
that the stations are 'representative'? There isn't a lot of similarity
between the 30-year averages for Kinlochewe and Kinloss, for instance -
the
driest month at Kinlochewe being on average almost twice as wet as the
wettest month at Kinloss, to quote but one figure. (Guess why, having
lived
near Kinlochewe, I now prefer to live nearer Kinloss!)

The regions are pretty much the long-standing regions used by the Met Office
since the flood. The actual regional and national sunshine values are, as
you
have probably inferred, fairly meaningless(1) ... the value of producing
them
is to make comparisons and detect trends over a long period of years.

(1) The Met Office have produced, quite recently, series of temperature,
rainfall and sunshine starting 1961 which, they claim, provide real areal
means of each particular element by interpolating values on a regular grid.
Can't remember what the grid resolution is ... 10km sounds sensible ...
can anyone confirm?

Philip Eden