Comments on: McShane and Wyner Discussion

By: Richard Drake

Richard Drake — Thu, 07 Jul 2011 16:51:38 +0000

Yep, your poetry’s gone downhill, my friend – though the multilingual aspect is creditable.

By: kim

kim — Thu, 07 Jul 2011 15:35:06 +0000

kim, we never knew ye!
=======

By: Donald Rapp

Donald Rapp — Mon, 03 Jan 2011 00:19:08 +0000

It seems to me that buried in the basis of all these manipulations by Mann et al. are the assumptions that each proxy measures a local or regional temperature, that each such local or regional temperature is the sum of a global average temperature influence term and a regional bias term, the regional bias terms are either positive or negative and when averaged over many proxies, sum up to zero. Hence, adding up many proxies produces a measure of the global average temperature according to this theory. Unfortunately, these assumptions do not seem to have been validated. Even a casual examination of the actual proxies show that they vary wildly from one to another. What seems to happen is that even if these assumptions were true (a consummation devoutly to be wished) the regional bias terms are orders of magnitude greater than the global average term, and the global average term is buried in an ocean of noise generated by the regional bias terms. Mann and company have attempted to use complex methods to extract the signal from the noise but as M&W and many commentators have shown, the process suffers from many ills, not the least of which is the imperfection of the proxies themselves. I would quote Carl Wunsch:
“Sometimes there is no alternative to uncertainty except to await the arrival of more and better data” (Wunsch, 1999).

By: Luis Dias

Luis Dias — Fri, 17 Dec 2010 03:03:20 +0000

That site is “skeptical” of skepticism….. their words. I guess that says it all really…

By: klee12

klee12 — Thu, 16 Dec 2010 22:53:47 +0000

When using statistics there should be a sharp distinction between the field in question and statistics. When we apply statistics to a particular set of data, all the meaning of the data has been abstracted away; the statistics should, I think, treat a time series the same way regardless whether it came from economics or paleoclimatology. Outliers should be determined by objective metrics. If a data set contains questionable data from the specific field, the data set should be not be used. I believe that MW deals only abstract statistics on the data which could represent economic data.

In SMR the statement (page 2, lines 30-33)

In the frozen 1000 AD network of 95 proxy records were used by MW, 36 tree-ring records were not used by M08 due to their failure to met objective standards of reliability

If the data was unreliable, why was it included in the data set? If the data is considered unreliable due to mathematical problems (i.e. outliers) then what is the objective measure to determine that they are unreliable? The issue is not whether or not 36
tree-ring records should be included or not, but whether SMR conflated paleoclimatolgy with mathematics (in the form of statistics).

McShane and Wyner wrote in their rejoinder states page 4

Consequently, the application of ad hoc methods to screen and exclude data increases model uncertainty in ways that are unmeasurable and uncorrectable

I think MW are saying the same thing that I was trying to get across. Introducing paleoclimatic reasons to exclude records is conflates paleoclimatology with mathematics.

Does anything I wrote make any sense. I’m not a statistician.

klee12

By: Jason

Jason — Thu, 16 Dec 2010 18:08:14 +0000

M&W Rejoinder == Best reconstruction smackdown ever

By: bender

bender — Wed, 15 Dec 2010 20:31:35 +0000

They are also forced to make a hand-wavy argument about it being a TEMP proxy not a PRECIP proxy. Not very satisfying.

By: Steve McIntyre

Steve McIntyre — Wed, 15 Dec 2010 20:28:25 +0000

One also has to guard against direct anthropogenic increases in ammonia concentration.

And then one has to allow for things like Bona-Churchill delO18 going the “wrong” way – but which hasn’t been published or archived by Lonnie Thompson – something that distorts the proxy literature.

By: bender

bender — Wed, 15 Dec 2010 20:08:47 +0000

Using these NH4+ concentrations as a temperature proxy, we reconstructed tropical South American temperatures over the last ∼1600 years. Relatively warm temperatures during the first centuries of the past millennium and subsequent cold conditions from the 15th to the 18th century suggest that the MWP and the LIA are not confined to high northern latitudes and also have a tropical signature.

Disagrees with Romm:
http://climateprogress.org/2010/09/21/hockey-stick-global-warming/

It sort of disagrees with Mann et al. (2009) – except that there are a couple of anomalous warm brown MWP spots in Peru and Iran near the middle of their blue Pacific and Pacific pools in (Fig 1). Curious that these spots are not smudges.

By: bender

bender — Wed, 15 Dec 2010 19:51:24 +0000

Ammonium proxy from S. America worth a look too (Kellerhals et al 2010). It’s calibrated against CRUTEMP, so there will be a need to re-do that calibration after UHI properly removed.