Alastair, you need to be a little cautious when trying to deduce trends or
cycles using running averages. It is a mathematical fact that any series of
unweighted means will show periodic fluctuations with a period two to three
times the averaging period. The effect has been discussed several times in
'Weather', (Burton June 1997, p195, Reynolds Weather 1978, pp 74-76).
Running averages of random numbers will produce some really convincing
cycles.

--
Bernard Burton
Wokingham, Berkshire, UK.

Satellite images at:
www.btinternet.com/~wokingham.weather/wwp.html


"Alastair McDonald" k wrote
in message ...
"Philip Eden" philipATweatherHYPHENukDOTcom wrote in message
...
Stats for here in Luton:
Year MeanMin MeanMax OverallMean
Snip

Philip Eden
Luton (obs site at 113m)

As you might expect, I was interested in that data for evidence of global
warming. Although this year's figures are greater than previous years,
they
are not a record even on a range stretching back only 25 years. The
figures in
1982 were higher than this year. I though that if I constructed a running
mean of 5 years then perhaps it would show a monotonic (a new word I have
just
learnt which roughly translated means 'steady' :-) increase in
temperature.
You will see below that this was not the case, however creating a 10 year
running average did show such an increase from 1992.

This raises the obvious question "Why?" The first answer that occurred to
me
was that there is an 11 year sunspot cycle, and a ten year running mean
was
long enough to remove its influence. So I constructed an eleven year
running
mean too. This shortened the montonicity to following year 1994. All are
shown below;

Year Mean 5 Yr 10 Yr 11 Yr
2004 17.7 15.1 15.0 14.8
2003 16.7 14.2 14.5 14.7
2002 14.1 13.8 14.6 14.6
2001 12.2 14.3 14.7 14.4
2000 14.9 15.2 14.6 14.4
1999 13.3 14.8 14.3 14.0
1998 14.4 14.8 14.0 13.8
1997 16.6 15.3 13.8 13.6
1996 16.9 15.1 13.4 13.2
1995 12.8 13.9 12.9 12.9
1994 13.3 13.9 12.9 12.9
1993 17.0 13.2 12.9 13.1
1992 15.3 12.2 12.8 13.4
1991 11.2 11.6 13.2 13.3
1990 12.5 11.8 13.5 13.7
1989 10.2 11.9 13.9
1988 12.0 12.5
1987 12.3 13.3
1986 12.2 14.7
1985 12.9 15.2
1984 13.1 15.8
1983 15.9
1982 19.6
1981 14.5
1980 15.9

It is interesting that there now seems to be a trend of +0.1C per year in
June
average temperatures. Since the sun spot cycle is actually 22 years it
would
be interesting to see what would happen with a 22 year running mean over
an
extended range of temperatures.

Cheers, Alastair.