2010-10-07 22:28:33Tropospheric hot spot argument in peer-reviewed form
John Cook


A who's who of skeptic scientists - Christy, Pielke Sr, Spencer, Douglass, have published a new paper:

What Do Observational Datasets Say about Modeled Tropospheric Temperature Trends since 1979?

Expect to hear about this paper a bit.

2010-10-07 22:59:16


Skeptics already running about, waving it in the air like they'd just won "capture the flag." Do they actually care about the "sciencey" part? No. 


2010-10-07 23:59:41
Ari Jokimäki


Umm... this doesn't seem to be about the hot spot, but lower troposphere temperature trends.

2010-10-08 00:32:04hot spot
John Cook

Just based on a quick skim, they seem to be talking about the lower tropospheric trend compared to surface trend - saying the tropospheric trend should be larger according to models. I assumed this amplification of the surface warming was talking about the tropospheric hot spot - could be wrong.
2010-10-08 01:47:30
Ari Jokimäki


The "hot spot" issue is about upper troposphere temperature trends in tropics. This new paper is about another issue - lower tropospheric temperature trends where there has been dispute for a quite long time. It basically got solved around 2004-2005 but Christy et al. keep publishing papers every now and then suggesting that it isn't.

The upper troposphere hot spot in tropics started to be an issue only about 2006-2007. It is very easy to confuse these two issues - I know that from my own experience... ;)

It amazes me that they discuss how RSS could be wrong in this and that way but at the same time they try to make it look like UAH is spot on. I guess they haven't learned from their mistakes. It would be nice to see NOAA STAR here too. That is globally more close to RSS than UAH but I don't know the situation for tropics.

2010-10-08 03:30:31tropical troposphere
Dana Nuccitelli

Yes, the paper is entirely about the lower tropical troposphere.  I read it a few weeks ago and it confused the hell out of me for a while, because often they just refer to the troposphere temperature - like in the freaking title!  But then in the paper, they claim the UAH trend is 0.067°C per decade, which isn't even close (for the lower troposphere it's actually 0.14°C per decade).

Like I said, this confused the hell out of me until I realized all their references to the 'troposphere' were actually talking about the tropical troposphere, where temperature trends are lower.  Anyway, ultimately by combining various different methods, they estimate a tropical troposphere trend of 0.09°C per decade, which is lower than the surface troposphere trend.

Unless I'm still confused, because they do talk about the lower troposphere as a whole quite a bit.  But all the data (or at least all the data I saw) pertained to the tropical troposphere.

2010-10-08 04:35:04


Not a very convincing paper. They desperatly tried to lower the uncertainties neglecting some dataset (thermal winds) and even RSS in the probability assesment. RSS is an outlier on the high side, but UAH is on the low side; this differnce alone should tell you what the uncertainties are.

They lowered model error by using the standard error of the mean instead of the standard deviation of the distribution. This error has been highlighted before.

The error in SR looks unlikely small. My guess is that they did not consider the error in the surface trend in the denominator.

Finally, I can't think of any mechanism by which you can get an SR<1 while warming. It would imply less convection and an increase temperature gradient.

These are my very first thoughts after a quick read.