Comments on: A Georgia Tech EAS8001 Assignment

By: fFreddy

fFreddy — Fri, 23 Nov 2007 22:41:03 +0000

But, hey, this is climate science.

Maybe, when it grows up. Until then, it is climatology.

By: Sam Urbinto

Sam Urbinto — Fri, 23 Nov 2007 22:09:06 +0000

But, hey, this is climate science.

By: Steve McIntyre

Steve McIntyre — Fri, 23 Nov 2007 00:16:25 +0000

There are two black series on the left – they are the English temperature history and the Pfister Europe documentary history which has a lot of temperature data (which I grouped with temperature). The yellow are the tree inrg T reconss.

By: MarkR

MarkR — Fri, 23 Nov 2007 00:07:27 +0000

Steve M. The Tree Ring Recon now looks Black to me, same as Instrumental T?

By: Andy

Andy — Thu, 22 Nov 2007 21:42:43 +0000

And people wonder why we at UGA referred to GT as the "North Avenue Trade School"? They take up a whole semester with this, when all they need to do is flash up Mosher's PennState Mann in the first 5 minutes of the first day of class. 'Nuff said.

By: steven mosher

steven mosher — Thu, 22 Nov 2007 19:58:29 +0000

As I suggested to JEG. A guest speaker on dec. 3 would be a very instructive
class. Cheese, crackers ( the kind you eat) and fruity wine in a bag will be served.

By: Steve McIntyre

Steve McIntyre — Thu, 22 Nov 2007 19:30:46 +0000

The online SI says that:

Surface temperature gridpoint data from 1856-1980 (from Rutherford et al, 2005 as cited in paper)

It turns out that Rutherford et al 2005 used HadCRU v1 rather than the then available HadCRUT2.

We use the HadCRUT dataset (Jones et al. 2001b) rather than the more recently published HadCRUT2 (Jones and Moberg 2003) version.

These is now two generations old. There are some puzzling differences between HadCRUT1 and HadCRUT2, which are a large job to analyze, but which I noticed a few years ago. For example, a Greenland gridcell which has a lot of measurements in HadCRU1 (thus qualifying for their dense network) has no measurements in HadCRU2. No explanation was given for these mysteries. HadCRU3 is different again. Gridcell versions are conveniently available at ftp://ftp.cru.uea.ac.uk/data/ . I have collation scripts in R if required by any EAS8100 students. In addition, the SI says:

Correlations were calculated between indicators and both (1) local temperatures (average over the 4 nearest instrumental temperature gridboxes)

It’s hard to see that either of these differences would account for the differences, but perhaps they do.

By: Steve McIntyre

Steve McIntyre — Thu, 22 Nov 2007 19:00:58 +0000

Yellow are tree ring temperature recons calibrated in deg C. – sorry about that.

By: Etienne

Etienne — Thu, 22 Nov 2007 18:54:52 +0000

same here on the yellow bars

By: Dan Evens

Dan Evens — Thu, 22 Nov 2007 18:51:33 +0000

Maybe it’s just my browser, but in every graph I see yellow bars and no yellow on the legend.