In my previous post I commented on the various responses of sceptics now that both that satellite and surface record show global warming is happening. (The map below shows the global warming trend for the troposphere from the satellite record.) Scott Church left a comment giving a detailed explanation of the satellite data. He concluded:
“Thus, the real lower to middle troposphere trend appears to be around +0.10 to +0.13 deg K/decade uncorrected for stratospheric cooling, and +0.18 to 0.21 deg K/decade with this accounted for—exactly as predicted by theory and the latest models. Due to the problems I’ve described above, few people today believe UAH[Spencer and Christy]’s lower and middle troposphere trends. The only exceptions to this are global warming skeptics, who prefer them because they yield the desired low warming that their belief system requires. It is not surprising that they typically avoid discussing the work of RSS and other groups, and the few times they do, they are dismissive of it without being particularly clear as to why. Outside of ultra-conservative and industry special interests, the claim of satellite measured atmospheric cooling is a colossal red herring. It is sad to see so many cling to a sinking ship for reasons that are entirely ideological rather than scientific.”
So, to all who worry about global warming, to all who think that people threatening to blow up millions to get their political way is no big deal by comparison, chill out. The science is settled. The “skeptics”—the strange name applied to those whose work shows the planet isn’t coming to an end—have won.
And how, you are perhaps wondering, did the sceptics win? MS&D write:
Three bombshell papers have just hit the refereed literature that knock the stuffing out of Blix’s position and that the United Nations and its Intergovernmental Panel on Climate Change (IPCC).
And who wrote these “bombshell” papers? Why Michaels, Singer and Douglass, of course. According to MS&D these bombshell papers settle the issue: global warming just isn’t happening.
“Bombshell” 1 (abstract preprint) temperature trends are “mostly negative”. How come? Well, they used Spencer and Christy’s version of the satellite data rather than the other versions that show more warming, but more importantly, they chose the time interval 1979 to 1996 for their analysis. The Wikipedia page on satellite temperatures has a table showing how the S&C trend changes if you use different ending years:
Notice how 1996 is the last year for which the trend is negative? If they had used any more of the data they would have found warming trends. They don’t explain why they truncated the data at 1996. They do claim that extending the analysis to 2002 does not substantially alter their conclusions, but they don’t give any details, just a reference to “bombshell” 2.
From the reference in “bombshell” 1 you might think that “bombshell” 2 has an analysis with data from 1979–2002. You would be wrong. That paper (abstract preprint) also truncates the data in 1996, and, surprise, surprise, finds that the S&C satellite data shows no warming trend. They offer two justifications for the truncation. First, that the 1997 El Niño affects the trend. However, the trend is also affected by the four previous El Niño events between 1979 and 1996, so this justification doesn’t make sense. Second, that there is a break in 1998 in another series they examine (the R2 series). But there is no break in the S&C satellite data so this does not justify truncating that series, especially since that truncation removes the warming trend from the satellite data.
The audacity of their next move is breathtaking. In their Tech Central Station article they write about the discrepancy between the surface and satellite records:
When this was noted in the first satellite paper published in 1990, some scientists objected that the record, which began in 1979, was too short. Now we have a quarter-century of concurrent balloon and satellite data, both screaming that the UN’s climate models have failed, as well as indicating that its surface record is simply too hot.
Now that we have a quarter-century of satellite data we know that the scientists saying that the record was too short were right. If you use all of the data it shows significant warming, just as the surface data does. The only way they were able to make it “indicate” that the surface was too hot was by deliberately leaving out the last one-third of the record. MS&D have turned the truth on its head.
Next MS&D attack the accuracy of the surface record, suggesting that it is inaccurate because of the urban heat island effect. They agree that the surface temperature record shows warming of 0.17°C per decade, but they suggest that might be the urban heat island effect rather than global warming:
The IPCC claims to have carefully corrected the temperature records for the well-known problem of local (“urban,” as opposed to global) warming. But this has always troubled serious scientists, because the way the U.N. checks for artificial warming makes it virtually impossible to detect in recent decades—the same period in which our cities have undergone the most growth and sprawl.
Virtually impossible to detect? Here’s a thought—why not not look at the data from rural stations? Which, surprise, surprise, is what they actually do. How on earth could MS&D write something so plainly untrue?
After four years of one of the most rigorous peer reviews ever, Canadian Ross McKitrick and another of us (Michaels) published a paper searching for “economic” signals in the temperature record. McKitrick, an economist, was initially piqued by what several climatologists had noted as a curiosity in both the U.N. and satellite records: statistically speaking, the greater the GDP of a nation, the more it warms. The research showed that somewhere around one-half of the warming in the U.N. surface record was explained by economic factors, which can be changes in land use, quality of instrumentation, or upkeep of records.
This is straight from the John Lott school of econometrics. They found a correlation between warming and GDP but correlation does not imply causation. If you want to mount a serious argument that more GDP is causing warming in the surface record you must provide a plausible mechanism and control for other factors that could have produced a correlation. They don’t do either of these.
The mechanism they propose is that the recorded warming was caused by land use changes, instrumentation quality or upkeep of records but they don’t have any measurements of these things. Nor do some of them even make sense. If richer countries can afford better instruments and records that would mean that the surface record understates warming because the lack of warming in the poorer country could be the product of poor records.
It is more likely that the correlation was found because both GDP and warming are correlated with a third factor that they did not control for—latitude. Richer countries are at higher latitudes and there is more warming at higher latitudes (see map at the top of this post). Robert Grumbine wrote a scathing criticism of an earlier version of this paper. He made an offer to help McKitrick with it but, as Grumbine relates,
Unfortunate that it was refused. Things like understanding averaging and the expected (and reasons for expecting that) climate change patterns would have gone some distance to winding up with a meaningful result. Instead, he went with Michaels, which was an easy route for ensuring the predetermined conclusion.
Update: The McKitrick and Michaels paper was even worse than I thought. It turns out that they confused degrees and radians in their model.
I’m afraid that their three bombshells turned out to be duds. They haven’t proved that there is no such thing as global warming. All they’ve proved is that they can cherry-pick to remove the warming trend from the satellite data and don’t know how to control for relevant variables in econometrics. And declaring victory with such a thin justification smacks of desperation.
Appendix: There were too many things wrong with MS&D’s article for me to fit in this post so I had to put them in an appendix.
Scott Church sent some comments which give more details about how selective Singer et al were in choosing their data.
I read over one of the paper too the other night. First, not only did Singer et al use only UAH, they used UAH Version D (Christy et al, 2000) even though Version 5.0 is available and contains updates to non-linear correction factors and an extended dataset (Christy et al, 2003). What’s more, they said in one of the papers that only the UAH analysis (Ver. D that is, NOT 5.0) is “confirmed” by the radiosonde data. Their basis for this? One (count ’em, one) analysis – the LKS sonde dataset (Lanzante et al, 2003). While this is a good dataset, for what it’s worth, it is but one of several that are equally valid, and none stands out as exceptionally well characterized. They ignored HadRT, RIHMI, and Angell 54, all of which have their strong points, and which differ in their conclusions from LKS. LKS was picked because it gives the best fit to the UAH dataset, and Version D was picked because LKS only goes to 1997, so only UAH Version D, and NOT the more up-to-date Version 5.0, is an apples to apples comparison – THAT’S why they stop in 1996!
Also, if you look at the coverage maps for most of these analyses, the best coverage (and therefore the best “vindication”) is in the northern hemisphere where UAH and RSS largely agree. The difference between UAH and RSS global trends boils down to differences in the southern hemisphere, northern Africa, and a few other places where sonde coverage is scant or non-existent. So the sonde data can’t truly discriminate between them. So much for the big vindication!
MS&D start their article with:
How many times have we heard from Al Gore and assorted European politicians that “the science is settled” on global warming?
The answer seems to be “zero“.
This is a detail from figure 1 of “bombshell” 2. It purports to show how the warming trend from the surface record varies throughout the world. Notice how there seems to be almost as much cooling as warming? Actually there isn’t. They just used the same colour (dark blue) to indicate missing data as they did to indicate maximal cooling. (Thanks to William Connolley for pointing this out.)
Later in the MS&D article when they are hyping their papers:
Three bombshell papers have just hit the refereed literature that knock the stuffing out of Blix’s position and that of the United Nations and its Intergovernmental Panel on Climate Change (IPCC).
Obviously ignorance of basic science is a job requirement for a Tech Central Station editor, but it is disappointing to see a mixed metaphor like this slip through. Surely “Three bombshell papers … that blow Blix’s position sky high” would be better? Or alternatively “Three piñata stick papers … that knock the stuffing out”. I like the second one better, because we already have a hockey stick paper and having a piñata stick paper or three would be way cool.