Dr. Roger Pielke Sr. on two recent "game changing" climate papers

Pielke Pere, he knows where.
===============

We shouldn’t forget that the results are corroborated by other evidence, such as
1) The continual upward revisions of recent warming on GISSTEMP (accompanied by downward revisions of the pre-1940 data)
2) The Steirou and Koutsoyiannis paper on weather station data homogenization, which showed up to half of global warming is artificial.
3) The Steig paper on Antarctic warming in generating greater warming than is justified by the raw data.
4) Issues of bias within the Australian and New Zealand temperature records.

Given the timely coincidence with Watts et al 2012, Mr Muller is just making a clown of himself.
Running kind of a family business is not easy nowadays, but this is a desparate business case and bad timing too.

An analysis team led by Anthony Watts has shown that 70% of the USHCN temperature
stations are ranked in NOAA classification 4 or 5, indicating a temperature uncertainties greater than 2C or 5C, respectively. First sentence from Muller 2011?
I have seen some other statements from NOAA that equates station classification to degrees of uncertainty. Implicit in these discussions seems to be that as a station’s classification moves from 1 to 5, the standard deviation of the temperatures and their anomalies is expected to increase, but the mean bias is expected (assumed?) to be and remain zero.
It is true that BEST, by removing absolute temperature from the analysis and only looking trends within and between stations makes an effort to make the assumption moot.[1] However, the assumption of mean bias = 0 for all classification might be at the heart of NOAA adjustments that Watts-2012 addresses.
IF one can argue that if the mean bias of class 1 stations is zero and so is the mean bias of class 5’s, then these can be homogenized to reduce overall uncertainty.
Look at the physics of the thermometers related to the classification scheme to assume contamination that increases the uncertainty. It strains credulity to believe that the contamination is equally likely to be cooler contamination as it is hotter. My hypothesis is that the mean bias from the class of cat 5 stations is > (hotter) than cat 1 as sources for heating contamination are more common than cooling. Mean Bias of cat 5 stations cannot be assumed to be zero and cannot be assumed to be equal to cat 1 stations. When you remove the assumption that mean bias is not zero, then the homogenization become much less tenable. Mixing cat 5 with cat 1 will raise temperatures.
So what? Trends are all that matter, right. The problem is that a cat 5 station today was probably not installed as a cat 5. Another hypothesis: the vast majority of stations start out as cat M and become today as cat N, where M <= N and mean_bias(M) <= mean_bias(N). The bias grows with time — UHI (Macro and micro) growth over time.
Two hypothesies to be tested using Leroy-2010 methodologies.
[1] In my opinion, BEST does finesse the mean bias well, but their scalpel and suture commits an even larger bias by removing low frequency data content from the data processing and is therefore no more trustworthy than the NOAA adjustments.

Well I hope you know this means you and Dr. Pielke are off Mikey Mann’s Christmas card list.

Some people are claiming that time of observation changes in rural stations may have messed up Anthony’s results, but if TOBS was a problem in Anthony’s current paper wouldn’t it have messed up his last one as well?

From Pielke Sr’s post toward the end:
My Comment: There statement that there is “the tendency to cancel random errors” is shown in the Watts et al 2012 and McNider et al 2012 papers to be incorrect. This means their claim that “the large number of stations also greatly facilitates temporal homogenization since a given station may have several “near-neighbors” for “buddy-checks.” is erroneously averaging together sites with a warm bias.
Hasn’t this been a major point that the temperature records show warming with certainty?

In my humble opinion, climate science has no better scientist than Pielke, Sr. When genuine science replaces activist/clownish science, climate science will look just like Pielke Sr’s work.
Pielke Sr’s endorsement of Watts’ work (and of other papers) is the gold standard at this time.

In the Fall, Watts, Nielsen‐Gammon, Jones, Niyogi, Christy, and Pielke Sr paper (2011), in Figure 4, the trend in the raw data is about 0.2°C per decade. The trend in the data corrected for differences in the time of observation is 0.3°C per decade. (The rest of the homogenization does not change the mean temperature much.)
Thus the difference between the trends in the raw data and the one in the homogenized (adjusted) the manuscript Watts et al. (2012) found are most likely due to forgetting to correct for the time of observation bias (TOB). This is an important issue, which is why McIntre is having some doubts about the manuscript.
As http://rabett.blogspot.de/2012/07/bunny-bait.html points out: “There is practically no time of observation bias in urban-based stations which have taken their measurements punctually always at the same time, while in the rural stations the times of observation have changed.”
Thus the differences in the trends for the different station quality classes are likely due to forgetting to correct for the TOB. Another likely problem with the analysis of Watts et al. (2012) is that the classification of the stations was performed at the end of the study period. Stations that were poor at the end, but average in the beginning will show an artificially stronger trend. Similarly stations that were good at the end, but average at the beginning will slow a weaker trend (or even negative trend). This selection bias may well explain the differences found in the trends for the various quality classes. The very least, these are issues, which a rigorous scientific paper would discus.
For now, I will not study this manuscript any further, expecting that it will never be submitted. If it is, I am happy to review it more closely, knowing how much Anthony Watts likes blog review.

I wonder whether Muller is loathing the day he decided to get into climate, and to engage with the critical community 🙂 Everything could be nice and cushy… now he has to read blog posts and deal with the whole enchilada that trails it.

@Bill Parsons , July 31, 2012 at 1:18 pm
Yep, them there thunderstorms is knowed as a negative feedback.
🙂