The Trieste PR Challenge conference has an interesting “White paper on tree rings submitted by Keith Briffa and Ed Cook, entitled “What are the Sources of Uncertainty in the Tree-Ring Data: How can They be Quantified and Represented?
Good questions. I urge readers to read this candid paper in full. I detect a lot of Cook’s influence in this paper – among the senior authors in this field, I find myself placing considerably more weight on Cook’s articles, which are usually pretty thoughtful.
Among their observations:
Low-frequency tree-ring variance [centuries to millennia] is virtually unresolved in all but a few chronologies worldwide. In many multi-century-length series it is undefined or random!
Curve fitting methods suffer from ‘segment length curse’. They can also suffer from “end-effect bias” when recent growth forcing signal is increasing (Melvin and Briffa paper in press). Some methods for processing tree-ring data (Regional Curve Standardisation and Age-Band Decomposition) do preserve more medium and long-timescale evidence of growth forcing changes, but they very prone to bias associated with non-homogenous samples and potential end-effect bias (Briffa and Melvin paper in press).
I like this one:
A fundamental problem is that tree-ring data from a site/region can produce very different chronologies according to specific sampling and processing – this is confusing for secondary users and other non-dendroclimatologists.
Here’s a point about Hansen and similar adjustments, originally raised in connection with the adjustments to the Dawson series used in tree ring calibration – which was actually a departure point for some of my interest in surface station records:
The way in which the climate data [i.e. station temperatures] have been pre-processed (i.e, homogenized) is also an issue that can profoundly affect interpretations of tree-ring data. This makes it doubly hard to identify and assess the signal(s) in the tree rings because it all may not be the tree rings fault!
Now for a real dagger:
There exists very large potential for over-calibration in multiple regressions and in spatial reconstructions, due to numerous chronology predictors (lag variables or networks of chronologies – even when using PC regression techniques). Frequently, the much vaunted ‘verification’ of tree-ring regression equations is of limited rigour, and tells us virtually nothing about the validity of long-timescale climate estimates or those that represent extrapolations beyond the range of calibrated variability.
Using smoothed data from multiple source regions, it is all too easy to calibrate large scale (NH) temperature trends, perhaps by chance alone.
Many chronologies need updating, but existing data sets need additional sampling (especially in ‘proven’ areas) to improve replication and allow improved standardisation methods to be used and enable longer calibration/verification and to explore responses of tree growth to recent climate trends in many areas of the world (i.e. just as stressed by IPCC AR4 Chapter 6).
The ITRDB is a great resource. It needs to be continually improved to allow easy storage of other than “usual” tree-ring width data. Improved meta data should be sought for all submissions, including tree dimensions and architecture and information on context of measurements (routinely including estimates of missing rings to pith). When standardised indices are archived, precise details of standardisation options should always accompany them.
An issue not mentioned at CA but for which I have complete sympathy:
Major crisis looming here are the physical samples. We are loosing the trees. Steady can tell you about his efforts in SE-Asia. In NZ, we have 40,000 year old ancient kauri being mined. I reckon it will be exhausted within 10 years. The holocene sites in 5 years. Saw-millers are already starting to buy farms so that they can secure some future supply. We have set-up an archive at a local museum for biscuits of kauri for future research programs. In other words I have adopted a fire-fighting approach – save as many samples as I can and hope there might be funding to work on them later. Steady has funded me over the last 5 years to collect silver pine (Halocarpus biformis) from the West Cost. We have multi-millennial chronos thanks to that investment – but some sources have been completely destroyed by the land being converted to dairy pastures. The other area is now a kiwi habitat sanctuary so the permit process for further sampling has become much harder. So, data archiving is vital, but I’m first trying to save samples!
Another issue that Pete and I ran into with Graybill in terms of poor archiving of samples, an issue that is under the control of the dendro community and all the more critical if the forests themselves are being lost through reasons out of the control of the dendros:
as old dendrochronologists whither away, their sample collections often disappear with them!
All in all, the Briffa-Cook White Paper reads a bit like a manifesto from Climate Audit. I wonder whether they even intended that this paper circulate to the public – it’s presently online only at the conference center facilities rather than at any of the author’s websites and I wonder whether they realized this.
I’m not even sure what, if anything, I disagree with in this paper. It’s hard to reconcile Briffa’s comments to a private community with his comments to the public in his capacity as the author of the corresponding section of AR4. I guess the problem that they have with Climate Audit is not so much with anything that we say, but that it’s said in public.