The negative bias of Anthony Watts, Hockey Sticks and the Dunning Kruger Effect

Another funny from Anthony Watts.  I won't dwell on the subject matter. Despite the fact that its undoubtedly a fascinating field, dendrochronology is a very specialised area in which I haven't the slightest bit of expertise.  But neither does Anthony Watts. (That's me fawning :))

He's come across a paper in his daily trawls of denialist websites (or via email or smoke signals or whatever).  It's a new paper by Cecile et al in the journal Climate of the Past: A likelihood perspective on tree-ring standardization: eliminating modern sample bias. The paper is discussing a new method to determine what is apparently widely known in dendrochronology circles as "modern sample bias".  I've seen the term in this 2008 chapter/paper by Keith R. Briffa and Thomas M. Melvin, for example.

Anyway Anthony is jumping up and down clapping his hands with excitement.  Why?  I think it's because the authors used the word "negative" and "bias" together.  He reckons it means the Hockey Stick is finally broken!  The five millionth nail he's tried to drive in the coffin - still the Hockey Stick lives on.

Here is his jubilant headline and the first line of his article:

And here are the tags - so he's not trying to be funny :)

Anthony highlights the bits of his copy and paste that he thinks are important - I've italicised Anthony's bolded bits:

Dendrochronologists observed that the older a tree was, the slower it tended to grow, even after controlling for age- and time-driven effects. The result is an artificial downward signal in the regional curve (as the older ages are only represented by the slower growing trees) and a similar artificial positive signal in the final chronology (as earlier years are only represented by the slow growing trees), an effect termed modern sample bias. When this biased chronology is used in climate reconstruction it then implies a relatively unsuitable historic climate. Obviously, the detection of long term 15 trends in tree growth, as might be caused by a changing climate or carbon fertilization, is also seriously compromised (Brienen et al., 2012b). More generally, modern sample bias can be viewed as a form of “differing-contemporaneous-growth-rate bias”, where changes in the magnitude of growth of the tree ring series included in the chronology over time (or age, in the case of the regional curve) skew the final curve, especially 20 near the ends of the chronology where series are rapidly added and removed (Briffa and Melvin, 2011)....

...Furthermore, modern sample bias produced a significant negative bias in estimated tree growth by time in 70.5% of chronologies and a significant positive bias in 29.5% of chronologies. This effect is largely concentrated in the last 300 yr of growth data, posing serious questions about the homogeneity of modern and ancient chronologies using traditional standardization techniques.

Looks as if Anthony's put two and two together and come up with two and two fifths.   Now I won't pretend to understand the details of the issue of "modern sample bias", but it seems clear enough that it's to do with deriving signals from aged trees and the application of statistical analysis.

In keeping with the tree theme and hockey themes, I'll stick my neck out and say that Anthony's got the wrong end of the stick.  If, as is stated in the excerpt Anthony quotes, the majority of chronologies (70.5%) have a "significant negative bias", that would mean they underestimate whatever the parameter is that is being measured (eg temperature) in the last "300 yr of growth data".  So if one was to apply that to the Hockey Stick chart as a stand-alone - it would surely mean that the hockey stick is even steeper in modern times!  In fact, had Anthony read the paper, he might have noticed this:

D’Arrigo et al. (2008) suggest that modern sample bias may be responsible for the “divergence problem” in dendroclimatology, the widespread reduction in temperature 10 sensitivity of tree-ring chronologies in recent decades

Not that I'm suggesting that the various published Hockey Stick charts are negatively biased in the recent end of the data.  These dendro scientists know their trees too well for that.  I also know that there are numerous temperature reconstructions that all show the Hockey Stick shape and use many more proxies than tree rings - as well as the obvious fact of having modern thermometers these days that show the sudden rise in temperature in recent decades. And all these different studies match each other fairly closely and, in the periods of overlap, are a close match with the instrumental data sets.

I'll stop here before I go all DuKE myself!  Professor Mann and all the other Hockey Stick producers who followed, would know a zillion times more about "modern sample bias" and dendrochronology than I (or Anthony Watts) will ever know.


It seems a nice little example of Dunning Kruger Effect in action at WUWT - don't you think?