There's been a lot of nonsense posted at WUWT in the last few days as usual. One thing that I noticed was an article (an ahem "guest essay") by David Dohbro (archived here). This is a good example of Anthony Watts not bothering to read what he posts. Or am I wrong? Perhaps Anthony is still having his own troubles with anomalous anomalies :)
David Dohbro is an "ice age comether". He has previously recommended selling global temperature stocks because he reckoned they were about to fall through the floor. He doesn't say, this time, whether he reckons they are still a sell or whether you should hold or buy them. (Archived here.)
What David's done today, or at least what he claims to have done, is compare three global surface temperature data sets with two satellite lower troposphere data sets. He's differenced some numbers, and it's not clear to me if he even knows what his different data series are each measuring.
What David says he found was that "Three land-based data sets consistently report monthly higher values". I don't think he means "land only". I think when he says "land-based" he's referring to the combined land-ocean surface temperature. Whatever he means, what he claims sounds a bit odd because these are in anomalies rather than "values", and if they are "consistently" reporting higher anomalies, then by now one would expect them to have a markedly steeper slope than the satellite lower troposphere data. As you'll see below, there are differences in the slopes but the differences aren't all that great. (I haven't tested for significance.)
The other odd thing is that he refers to all the data sets as GSTA's. That is, as global surface temperature anomalies. Yet the two satellite sets of data that he uses for comparison are for the lower troposphere, which is an average of the air temperature over a vertical distance above the land and sea surface.
David put up a pretty pointillistic chart - not as pretty as a Seurat but pretty just the same. I've copied it below. Click to enlarge as usual, if you must:
|Source and all credit to David Dohbro at WUWT|
|Bathers in Asnières by Georges-Pierre Seurat (1859 – 1891)|
David describes the chart as follows:
I then simply subtracted the UAH monthly GSTA from the corresponding monthly GSTA of the other data sets (in this case, GISS, Hadcrut 4, RSS and NCDC; Data from January 1979 through March 2014). I then plotted these differences for each corresponding month and performed linear regression through each set of differences (Figure 1). A value of 0 means that the UAH data and the other dataset are similar, a value >0 means the other dataset reports a higher monthly GSTA compared to UAH and vice versa.
As I said earlier, UAH (and RSS) are not global surface temperatures. They aren't even global lower troposphere temperatures because they don't quite cover the entire globe. But let's put that to one side. It looks to me as if David should have done a bit more work and thought things through before writing his article. (As for Anthony Watts, goodness knows what he was thinking when he posted it.) For example, David makes the not so startling find that:
As you can see, the three land-based data sets consistently report monthly GSTA higher than that of UAH.
Of course they do. UAH is an anomaly from the 1981 to 2010 mean. GISTemp is an anomaly from the 1951 to 1980 mean and HadCRUT is an anomaly from the 1961 to 1990 mean. So of course the anomaly is going to be greater for data sets with an earlier baseline.
David also comments about the difference between UAH and RSS, writing:
The other satellite based data-set, RSS, reports values rather similar to UAH (average difference of 0.058°C).It should be different. UAH anomaly is from the 1981 to 2010 mean whereas RSS anomaly is from the 1979 to 1998 mean, so the RSS anomalies should have been higher than the UAH anomalies.
David concludes by writing:
In summary, all five GSTA datasets analyzed here show an average GSTA over the past 35 years of between 0.01 to 0.42°C above their respective baseline period that varies between each data set. The land-based data sets report in all most all cases monthly GSTA that are higher than the satellite based GSTAs. In addition, there is a general trend towards larger differences between the former and later data-sets over time (since 1979). The GISS data-set has the strongest trend in difference over time and will soon report the largest difference with UAH if this trend continuous, as well as diverge more from the other land-based data-sets. The continuing divergence to the point where the difference is larger than the long term averages between satellite-based and land-based reported GSTAs warrants more in-depth analyses and attention.
Interestingly, he calls for more research, more in-depth analysis and attention - which is unusual for someone at WUWT :)
Since we're on the subject, I've plotted all except NCDC (four is enough) having adjusted all the sets to the same base - 1981 to 2010. Here is the result:
|Data sources: UAH, RSS, Hadley Centre, GISS NASA|
As you probably know, there's not a great deal of difference between the two surface data series. GISTemp has a slope of 0.0159 and HadCRUT a slope of 0.155. UAH and RSS are pretty close except for the last couple of years, where RSS is lower than everything else. UAH on the other hand starts off higher than anything else. Also, RSS only goes from 70s to 82.5N, whereas UAH supposedly covers more area. (I can't remember where UAH starts and stops - can someone remind me?) However I'd expect that the UAH team have to deal with the problem that prompts RSS to leave out Antarctica - that the mountains get in the way and muck up the readings.
From the WUWT comments
Streetcred, despite the name, is a fake sceptic and says:
May 19, 2014 at 12:28 am
GISS not all that Kosher? Tell Steven Goddard something that he didn’t already know. There’s clearly a fiddlin’ going on about there. ;)
Mike Jonas is one of many who gently points out the obvious and says:
May 19, 2014 at 12:44 am
Because of the different base periods for anomalies, maybe the comparisons are not as useful as they might be. Can you get hold of the base period data for all series, then re-construct the absolute temperature series for all except UAH (say) and re-base them on UAH’s base period. The results could be a bit different.
If you can’t get hold of the base period data, then you can rebase them all to a common period (eg. 1979-1989), provided you then report annual averages not monthly data. Given that your main findings are expressed in deg p.a., the results will I think be equally valid. The graphs might also be easier to interpret.
Nick Stokes goes one further and points to an interactive chart he's prepared and says:
May 19, 2014 at 1:10 am
“As you can see, the three land-based data sets consistently report monthly GSTA higher than that of UAH. With NCDC > GISS > Hadcrut4 > RSS. NCDC’s data set reports on average a monthly GSTA 0.41°C higher than that of UAH”
As Mike Jonas says, this is meaningless unless you put them on the same anomaly base. The trend differences are meaningful, but the only one that stands out is the difference between UAH and RSS. UAH and the surface measures are relatively close. It is Lord M’s favourite, RSS, that is the outlier.
Theer is an interactive graph here of those five indices, plotted monthly on a common base. It is interactive – you can rescale etc. Scroll up for details.
Is thegriss being sarcastic or serious?
May 19, 2014 at 1:12 am
Its good to know that 1 out of 4 is close to reality.
Hmm, looks like thegriss wasn't being sarcastic, but is a resident WUWT conspiracy theorist, and says:
May 19, 2014 at 1:15 am
With Gavin and Phil still in charge of Had and Giss, the real temperatures will continue to show divergence,
Gavin and Phil will continue to try in vain to CREATE a positive trend as the temperatures start to drop slightly over the next several years.