Thursday, April 12, 2012


date: Tue Nov 2 16:10:25 1999
from: Tim Osborn <>
subject: Re: dendro data
to: Julie Jones <>

At 02:51 PM 9/22/99 +0200, you wrote:
>How are you? I hope things are going well. I'm OK, a bit tired still
>after Japan, with a brief stopover in Brentwood on the way back.
>I was speaking to Phil about the dendro data of Schweingruebers that you
>have. He said to mail you about it, and said that I could have some
>chronologies, which I've forgotten the name of, but they sounded Swiss.
>Do you have any idea which ones they may be? (I've just remembered
>Schweingruber is Swiss, so that's probably not much help whatsoever!).

Hi Julie

Sorry to be so very sssssslllllloooooowwwwwww in replying - busy as usual! I hear that you're coming to the CRU Christmas Dinner in December - it'll be good to see you.

The tree-ring data Phil was referring to are probably a network of tree-ring density chronologies. They cover much of the Northern Hemi (high latitudes or high altitudes, to make them more sensitive to growing season temperature), not just Switzerland. The year-by-year densities have been measured by Schweingruber's lab in Switzerland, so that's probably where that connection comes in (although he may have mentioned that they have been Hugershoff standardised - which also sounds a bit Swiss! - which is a method of removing the bias due to changing tree age, although it can also remove some low frequency variability).

Anyway, Keith Briffa, Phil and I have been working on these for just over 2 years now. We used them in those two Nature papers that we had last year, but Keith wanted to wait until we'd submitted more general papers describing the data set, it's climate sensitivity and some calibrate reconstructions before releasing it. These papers are now near to being ready (not sure what near means) and I've persuaded Keith to set up a web-site to disseminate them.

What we've got are:

(1) the raw chronologies (dimensionless time series between 100 and 600 years long from about 390 locations).
(2) a list of those chronologies that correlate significantly with growing season temperature.
(3) nine regional averages of the chronologies, that have been calibrated to produce reconstructions of regional April-September mean temperatures.
(4) one hemispheric wide calibrated reconstruction.

and what we've almost got are:

(5) gridded, calibrated reconstructions of growing season (April-September) temperature on the Jones (that's Phil not Julie!) 5 by 5 grid.

I should tell you that there is a fairly strong temperature signal in the tree-ring density series, but that a non-temperature trend is also apparent post-1950 that gets bigger and deteriorates the temperature relationship. This makes calibration somewhat harder (hence I've been working on them for 2 years...!), but you also have to make the assumption that this non-temperature signal is something anthropogenic and didn't occur in the past.

The web-page and dissemination will be done by the end of the year, but some (e.g., 3 & 4 above) might be available sooner. I assume that the individual tree cores (ring widths and ring densities) and the chronologies will be submitted to the International Tree-Ring Database at about the same time. You might like to think about whether you want raw or calibrated series, and regional means or gridded data (the latter apparently give more spatial detail, but uncertainty ranges are bigger).

Hope it was worth the wait!



No comments:

Post a Comment