On Fri, 23 Apr 2004 00:11:20 +0100, "Philip Eden"
philipATweatherHYPHENukDOTcom wrote:

I'll bite my tongue and not comment on the content. But, Julian,
what
do you make of the boundaries used between different colours?
These are, presumably, rounded figures which approximate to quint
boundaries ... is this cartographically acceptable these days, is
it a
whim of the compilers who perhaps don't give a damn what is and
is not normal practice, or is it, ahem, best described otherwise
in
private? Or am I plain out of date?

I'm not sure about the science of it but I much preferred the old
contoured mapped averages. It's very hard for me to see the boundaries
between similar colours on these new maps (shades or red or blue),
particularly where are several colour changes in a small area.

Also, the boundary values used on some of th maps are frankly weird
and make it hard or impossible to discern the differences in climate
in, say, lowland Southern England. This was easy to see with contoured
maps.

Example - Colour key for Days with Ground Frost in January:
White: 2 to 16 days
Light blue: 17 days
Greeny blue: 18 days
Dark blue: 19 days.

What on earth?????

What's wrong with old fashioned 2 to 5, 6 to 10, 11 to 20? Come on,
Met Office - does anyone really care if somewhere averages 17, 18 or
19 days? NO! Because in practical terms, the difference is
insignificant. On the other hand, the difference between a monthly
average of say 3 days and 15 is HUGE and could be important. Not to
mention interesting.

So, I don't like these new maps one little bit - all they had to do
was lightly colour the contoured maps and everyone would have been
happy. Change for no good reason. To the statisticians and graphics
designers involved, I say "Pah!"

--
Dave