Monday, April 30, 2012


cc: Popa Ionel <>, Phil Jones <>, Keith Briffa <>, Simon Tett <>, Gabi Hegerl <>, "Gerald R. North" <>, Claudia Timmreck <>, Stephan Lorenz <>, Sandy Tudhope <>, "raymond s. bradley" <>
date: Tue, 11 Nov 2008 16:44:51 +0000
from: Thomas Crowley <>
subject: Medieval9
to: Bo Vinther <>


Hi Bo,

just to whet your appetite of our new product, here is an updated
reconstruction of 30-90N temperatures (land) for 994-2007

(I wanted to focus on annual data to validate new volcano simulations
being run by some modelling groups...)

this simulation incorporates several features not previously included

new data from

Alberta (Canada) - (Luckman extended record)
Alaska (D'Arrigo-Wilson)
Carpathian region (Popa, CD this year)
Mongolia (some of the finally released Jacoby data)
Alps (Jan Espers work)

the method combines long reconstructions from nearby sites of Yamal and
Polar Urals in order not to overweight one region

I only use sites that have records extending continuously from 994-1960
- calibrated with instrumental data over interval 1880-1960 (r=0.64,
error = 0.25 C)

the nine sites have very nice spacing - White Mtns (Nevada), Alberta,
Alaska, Scandinavia, Alps (SudTirol), Carpathians (region we never had
before - big hole), West Siberia, East Siberia (Taimyr), and Mongolia

would be nice to have an annualized time series from China, but so far
cannot track one down

used 30-90N (land) because that is where the best paleo data - that is
where we can best validate volcano simulations, and, in general, most
people still live on land - somemore more useful metric than global temp.

note approximate 2.5 C range in temperature from depth of Little Ice Age
to present (also have extended instrumental series to 2007 - thank you
Phil) - pretty big

zero line represents Phil's calibration interval for instrumental data
(Phil - 1930-1960?)

note only ONE year rises above Phil's zero reference level -- AD 1031 -
beginning about 1920 values consistently rise above that, therefore
supporting Gabi's interpretation of detectable global warming signal by
mid-20th century

sending this out to others for any comments/questions - when we get the
annualized Greenland O18 we will be done, unless someone knows of a
reliable annual time series from China (one published last week in
Science was unfortunately biannual)

with regards, Tom

ps 1258 cooling only about 0.5C, supporting conclusions from work I am
doing with the Hamburg group that large flux at that time was associated
with increased particle size, which led to increased absorption of
longwave radiation and damping of cooling signal (which should have been
10X Pinatubo)

The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.


No comments:

Post a Comment