Graham P Davis wrote:

What might have appeared obvious is not completely true. In the late sixties
and early seventies, I used to brief the Met Office long-range-forecast
team each month as to ice conditions in the Arctic. To be fair, the amount
of actual data was variable. We were only able to start using satellite
data in about '65 and that wasn't a great deal of help in the depths of
winter. Even summer was a problem due to the large amounts of low cloud
over the Arctic. Infrared pictures came in later but they didn't show up
thin ice and also low cloud was again a problem.

We had ship reports and some aircraft recces but coverage wasn't great.
Commercial airlines had just started flying over the Arctic and we
persuaded them to mark the ice edge on a postcard as they over-flew it and
to post it to us after landing. Where we had no reports we relied on
degree-day data and pressure charts (mainly 10-day means).

The normals in ice atlases were generally quite reliable but with a few
glaring exceptions. For instance, until the first satellite data arrived,
it was believed that the centre of Hudson Bay remained ice-free throughout
the year as opposed to what we soon found out, which was that it was
covered for half the year.

That's all drifting off the subject a bit however. I can't remember what
data N E Davis used - this was in a talk he gave at the Met Office at the
end of 1969 and my memory isn't that good - but I suspect that tree rings
came into it somewhere. As far as I recall, these cycles applied to the UK
only. In 1975, the Global Atmospheric Research Program produced an analysis
of global temperature in terms of five cycles, ranging in length from 100
to 10,000 years. This was based on 700,000 years of paleoclimatic data.
There has probably been more research on these lines since then but I'm a
bit out of touch.

Thanks for the reply. I haven't anything positive to say against
dendochronology per se but I have a deep suspicion of it.

And of course it is the only chronologically logical model we can rely
on, there being massive problems with every other dating method. The
best of which (radio active carbon) is extremely iffy.

Imagine putting great pedestals of modern thought and the abandonment
of thousands of years of ancient ideals, on ideas less reliable than
our knowledge of the surface conditions of the Hudson Bay.

Crackers isn't it? I feel like going out and voting for a chimpanzee.