… for questioning the science. Ironic, don’t you think? All this thinking about Flood got me contemplating global warming, so I wandered through some global warming-related sites and ended up at the infamous Watts Up With That, where I found two egregious examples of poor science, made particularly annoying by their patronizing tone and vicious ill humour. I’m not an expert on global warming science, but I know a little bit about research methodology, the validity of classification methods, and time series analysis, so I decided, perhaps foolishly, to comment on the two offending posts. Within two days I had been banned from the site, though not before I’d been roundly insulted by its resident flying monkeys and not with an official declaration – just quietly shuffled off to the spam queue.

Me, spam!!! Oh, the humanity!

Poorly documented and subjective classification methods

The first, and most egregious example of bodgy science that I commented on was an example of the ludicrous “surface station audit” that Watts and his cronies are engaging in. Recently Watts reposted one of the surface stations that his crew have audited, with a glowing commentary that reproduces the basic facts of the original audit. This post also includes a healthy dose of green baiting, because some governor or other has established a pro-anthropogenic global warming (AGW) website and this is just the worst thing ever so should be sneered at. Unfortunately, if you follow the link to the original “audit” (here) you find in comments that a major piece of the audit is completely wrong. Basically, at issue in this audit is the construction of a tennis court close to the temperature station, and the presence of  a trash burning can (some weird American invention) very close to the weather station. Watts claims that the tennis courts were built in “the early 1980s” and that this corresponds with a sudden increase in temperature readings – obviously this could be linked to the tennis courts. But in the original audit someone who lives in the area has posted to say that the tennis courts were built in 1973 and gives a link. In reposting the “audit,” Watts ignores this information and directly links the tennis courts with the 1980s temperature jump – which is false. He also consistently refuses to answer questions about the facts related to the tennis court.

The big problem here, that Watts won’t admit to, is that his classification system in the surface station audit is completely bogus. It suffers from three problems, which anyone who knows anything about research methodology would immediately recognize as “junk science” (to use a Watts-ism).

  1. It is not objective: he has given no criteria for distance of a tennis court, no standards for waste heat sources, no standard methods for obtaining data, no system for establishing potential changes in data collection. In trying to establish when the tennis court (a potential heat source) was installed, he just chatted to a neighbour who admitted she wasn’t there when the courts were built. No reference to land and environment courts, no attempt to contact the owners. Worse still, when he is given data about the actual time when the court was built, he ignores it and reblogs the original (erroneous) date. This is, to use a scientific term, bullshit. There’s no objective assessment standard here at all.
  2. It is inaccurate: It is clear that Watts doesn’t care about little details like time, frequency or distance. He didn’t care to establish the exact date of the tennis court’s installation – was wrong by 10 years! – and has consistently failed to answer the questions I asked him about the trash burning can. How long was it there for? When were temperatures recorded – before or after the commencement of use of the burning can? Why was it removed? There’s no way a trash burning can can induce observations consistent with global warming if it was only there for 6 months. Watts can’t answer these questions because he doesn’t care about a 10 year error in estimation of when a heat source was placed near the station, and he doesn’t care about how long a waste heat source was functioning right next to a surface station.
  3. It is not blinded to the observed data: this is the worst of all possible worlds. In his post Watts makes it clear that his assessment of the validity of the station siting is directly related to his assessment of the temperature trend at the station. This is precisely the wrong way to categorize stations. Furthermore, when challenged about his bullshit assessment criteria, he defends them by saying nearby stations don’t show the same trend. This is both arrant shite (nearby stations could be underwater for all I know), and a further example of the same bias. You never, ever classify experimental groupings by the outcome! This is rule number one of good science! Imagine if a pharmaceutical company excluded from its research any patients who did not respond positively to its drug because they were “biased.” Would you trust that drug? That’s what Watts’s station audit does.

This is absolutely the stoney end of the scientific method but it is common practice at Watts Up With That. When he posted his latest “paper” (haha, see below) for “peer review” (actually, editing), he was accused of the same thing – using the result he was looking for to classify stations. It’s a consistent problem in his work, and in other work he references. Which leads me in to the second thread I commented on …

Tautological reasoning with erroneous statistics

Watts recently posted a scan of a 1987 conference presentation by one Jim Goodridge, entitled Population and Temperature Trends in California, which predates Mann’s famous “hockey-stick” presentation and claims to find major flaws in the California temperature record. Unfortunately, this paper is an extremely poor research article, poorly presented with extremely bad statistical methods, and it shows nothing. Attempting to point this out to Watts is probably what got me banned, because criticizing mainstream climate science is fine but criticizing a paper Watts likes – even a crappy one – is unacceptable.

This paper basically divides all of California’s temperature stations into “urban” and “rural” categories and shows that the urban stations are warming and the rural ones are not. Unfortunately, this isn’t much of a gotcha moment because the method of defining “urban” and “rural” is to divide the stations into two groups according to their temperature trend. Those that have a temperature trend greater than 0.0125 per year are “urban” and those that don’t are “rural.” That is, Goodridge concludes that a set of stations he defined by their warming trend show warming, and a group of stations that he defined by their cooling trend don’t show warming. Alert the press!

Goodridge then goes on to show that the rural stations’ temperature trend correlates with sea surface temperatures, while the urban stations correlate with temperature. This is treated as a revelation by Watts’s flying monkeys, but is completely irrelevant because it’s purely statistical artifact. Under standard global warming theory, sea surface temperatures warm more slowly than the land – they should be expected to correlate with data from a group of weather stations defined purely through their low warming trend. Similarly, population is increasing over time, so should be expected to correlate with any other time series that is warming over time – e.g. data from a group of weather stations identified by having a warming trend greater than 0.0125 per year.

This is exactly the same kind of subjectivity being used in the surface stations audit. If this was done in a trial of a new drug it would never get published. It’s reprehensible! The flying monkeys try to justify it by pointing out that Goodridge (who was a “state climatologist”) observed that

In general the classification of records as urban or rural is fairly close to reality as the writer knows it from viewing most of the sites

as if this is an objective criterion. It’s not, and if it were Goodridge would have divided his sites up according to his classification from the very beginning. On this thread I repeatedly pointed out that there are established standards for defining “rural” and “urban” and he could have just used them, but funnily enough no one was interested. They’re much more satisfied with Goodridge’s post hoc justification than with objective classification. I pointed out that if a climate scientist did the same thing they would be up in arms. Imagine if a climate scientist said this:

stations were defined as ‘accurate’ if the regression slope of temperature was greater than 0.01. Accurate stations showed a significant warming trend, while inaccurate ones showed cooling. This is clear evidence that the world is becoming warmer.

I don’t think the Watts crew would let that fly. But if someone who agrees with them does it, his subsequent clarification that the station division seems about right is good enough for them.

That’s not science. Furthermore, Goodridge’s analyses don’t include any time series adjustments, and you absolutely do not calculate correlations for time series using the standard Pearson correlation coefficient (as he did). This is weak. But if you say so you draw a huge amount of flak and ultimately get banned – with no warning or announcement.

The rank hypocrisy of denialism

Watts up with that is a classic example of the hypocrisy of denialism. In the very post that I commented on, Watts posts up two emails between climate scientists, in which one refers to Goodridge as a “nitpicky jerk.” In response, Watts says these scientists “are some piece of work, aren’t they?” After 50 or so comments on this thread, I was forced to post this:

Since I’ve posted on this thread I’ve been called a Nazi, accused of personal attacks, called a hater (by you), stupid, accused of getting “uppity” and “worked up” (by you), accused of “nit-picking” (ironic in light of the emails you’ve published), accused of being incapable of understanding what others write, and you’ve had to censor one comment for saying bad things about me. Also, even though it’s clear that I don’t want my identity revealed, you’ve tried to do so and have even published my place of work, in a very hostile forum.

Watts’s response was to deny the Nazi part, and after I pointed out to him that he himself had censored the comment containing the phrase, he subsequently implied I had accused him of calling me a Nazi (which I clearly didn’t). So when climate scientists use the word “jerk” they are “a piece of work,” but when Watts’s flying monkeys lay into me en masse it’s just robust debate. When it comes to robust debate they’re complete naifs, however – nothing I said there is anything like the challenge I’ve been subjected to by commenters like Noism, Paul and various drive-by commenters in my Tolkien and Fascism or Allies World War 2 Race Trap posts, and I didn’t call anyone in those threads a nazi, hater or idiot. I didn’t call Noisms a coward, either, for not using his real name – which Watts did in his very next reply to me. Charming people over there, Watt?

But the hypocrisy doesn’t stop there. Just a couple of posts prior to this post – which contains two stolen emails – Watts reposts an opinion piece from the Heartland Institute in which they decry the theft of emails from the Heartland Institute. Apparently when denialists do it it is a quest for justice, but when a climate scientist does it it is a crime. After just two days of challenge they banned me from their threads – but they complain if climate scientists get defensive and evasive after 20 years of the same treatment. But compare: currently Tamino has a thread in which he critiques a paper by Hansen, who the denialists hate, and that thread has 104 comments of reasonable discussion about whether the paper is flawed. Yet I’m supposed to believe it’s the climate scientists who are prickly about criticism?

In case you thought it stops there though, Watts’s approach to the ethics of publication is founded entirely in hypocrisy. He and his flying monkeys are constantly making jokes about climate scientists rushing to press with their results, and they dedicated a whole thread to bitching and moaning about Muller releasing his results to the press before peer review was complete. But Watts rushed out a press release about his own latest “paper” before he had even submitted it to a journal. When Watts does it it is an urgent effort to present pressing and important information, but when Muller did it it was an arrogant scientist protecting their reputation and drumming up funding. Ultimately of course, Watts’s paper is unpublishable – it contains no measures of uncertainty and doesn’t adjust for Time of Observation Bias, so it’s fundamentally flawed – but that didn’t stop him getting it into the press. He and his flying monkeys would be singularly angry at a climate scientist doing that. Furthermore, Watts won’t release the site classification details for his latest paper – having been deeply involved in a campaign to force mainstream climate scientists to release every scrap of their data and code. Of course, the climate scientists, being government funded, are covered by FOI rules and obligations to funders and courts. Not so the idiots at WUWT, who can make any claims they want and keep their data entirely private. The whole site is like an object lesson in projection – every single criticism they lay on climate scientists, they themselves are doing in spades. Even the omnipresent “alarmism” accusation is rife across the site. In the same breath that they talk about “alarmist” scientists exaggerating AGW for a political agenda, they will talk about the risk of impending fascism due to the political machinations of environmentalists and elites. That’s alarmism! Watts depends on this alarmism for donations, but saying so is oh so incredibly crass (not that I did) – while accusing scientists of exaggerating AGW for funding is de rigeur on every thread.

A bunch of clowns

For the last couple of years some bunch of madcap polar scientists have been taking bets on the arctic sea ice extent at maximum melt. Each time, Tamino at Open Mind has stomped the field with his predictions, and the Watts Up With That crowd have massively over-estimated sea ice extent, because each year they think is the year sea ice will recover. This year they have been a bit quieter, but they are currently hosting a thread about “when will the cooling begin” which includes a link to this doozy from February, in which some idiot deletes 40 years of data from a 100 year series in order to prove that a lake’s depth is related to solar cycles – without mentioning that the lake’s source rivers are heavily influenced by human behavior (a point raised in the comments). Gee, I wonder if they’d accept that from a climate scientist?

These people are basically completely ignorant of the scientific method, data collection techniques, the basic rules of classification and validation, and any statistics that isn’t built into MS Excel. They are ill-mannered, defensive, hypocritical and shrill, and they are consistently wrong about everything they touch. This year the arctic sea ice is going to reach an all-time record low, just five years after the last one, at the same time as Watts and his flying monkeys are postulating a rapid recovery, claiming all temperature records are fudged, and predicting a return to a cooling world next year. They are simultaneously sneering at any environmentalist ideas (witness this response to the suggestion of using urine for carbon capture technology), any science which suggests humans can affect the climate, and any solutions for same. In the long march of history, once global warming has begun to really eat into our environment, these people will be viewed not as harmless clowns but as criminal propagandists, the way we now view the people who protected Big Tobacco. They don’t don’t accept any kind of genuine scientific debate, and they will turn nasty and threatening if you even try. In my opinion they aren’t worth debating with, they aren’t set up to handle debate, and debate is not their interest. The entire site is an exercise in poisoning the well. Denialism has no credibility, and if Watts ever gets any of his work published it will be through deception and luck, and whatever journal publishes him will regret it.

So much for my first and only incursion into the denialosphere. Now I’m off to take a shower…

Advertisements