Email 4904: Keith Briffa struggling with divergence problem in 1997 - "I think we have to ... not use recent data ... !"

ClimateGate FOIA grepper!

But to go to science - surely you have talked to Ed about the work I've been doing trying to document and understand the change in sensitivity of northern density (and ring widths) to temperature. I have a manuscript but the problem is a major one and is complicated by the issues of CO2 , temperature thresholds,standardisation regional coverage(i.e. spatial scale),appropriate climate signal etc. etc. Simply we see a loss of decadal scale sensitivity in much of our large spatial average data - and in our recent Russian calibrations. I know you talked about this in a couple of papers and have suggested a recent appearance of moisture sensitivity in your northern American trees. I do not know why this is happening but it seems clearer in the density data. We (at least Fritz) has explored the technological possibilities - i.e. that density is biased by an inability to record maximums correctly in very narrow rings but we do not believe this to be the case. I think some threshold may have been crossed that means the densities are limited in their ability to record high temperatures and of course it could be a drought type response in warm periods. These suggestions do not seem to be the answer - or at least all of the answer. Similar warmth before and less of an underprediction of temperatue then, plus the widespread (though not perfectly synchronous) manifestation of the phenomenon lead me to suspect synergistic influences. I really think that nitrates,CO2.tropospheric ozone - and ,certainly not least,
increased uv could each or all be playing some part. As for calibrating transfer functions , I think we have to somehow adjust recent tree growth records or not use recent data in the calibrations!

Via AJ