John Hall wrote:
In article ,
Graham P Davis writes:

I've built a graph based on the 1975 GARP analysis of 700,000 years
data. This analysis used cycles of 100,000, 20,000, 2,500, 200, and 100
years. Adding these cycles as best I could - not having the original
data and having to rely on a small graph published 25 years ago - I get
a global maximum temperature occurring towards the end of the 1730s.
Other peaks from 1700 to 2100 are 1855, 1940, and 2050, with the latter
being the highest of the slowly increasing series.


Those cycles seem to have suspiciously "convenient" lengths. Also, apart
from 2,500 and 200 years, all the cycles exactly divide into one
another, which I imagine will tend to amplify the effects. Wouldn't the
results have been very different if the last two cycles were really,
say, 193 and 104 years? A very small error in the estimated length of
the shorter cycles could make a big difference if you attempt to use
them over such a long period as 700,000 years.

I agree that the lengths of the cycles are suspicious. The article in
which the graph appeared says "The panel on climatic variations of the
Global Atmospheric Research Programme (GARP) have examined paleoclimatic
records spanning the last 700,000 years. They have analysed the surface
temperature in terms of the supposition of five periodic functions,
quasi-periodic cycles of the chosen periods being evident in the
climatic record." The word "supposition" bothers me a bit! Other studies
suggest an 80-year cycle, possibly associated with changes in solar
activity, is more likely than a 100-year.

The cycles, although mostly dividing into each other, are out-of-sync
with each other. This leads to the slight variability in the frequency
of maxima and minima - about 85-115 years.

The article I've quoted from is /Man's Impact on Climate/ by A J Crane,
of the Central Electricity Research Laboratories. It is from a lecture
in 1971 published in /Food, Nutrition and Climate/ by Applied Science
Publishers in 1982.

The original data was published in 1975 - /Understanding Climatic
Change: A Program For Action/, US Committee for GARP, National Academy
for Science, Washington DC.

I would have hoped that this data would have been re-examined in the
past thirty years but I've seen no evidence of it - though I admit my
efforts have been limited to a bit of Googling.

Graham