Time to Defend Science

I have set up a petition so those who object to the traducing of climate science can make their voices heard. I hope even opponents of the mainstream position will sign, because the issue is not whether the science is correct or not, but whether science should be tested by weighing evidence against theories, or by personal attack on scientists.

Even The Guardian, a paper that usually breaks from the herd and attempts to report matters like this accurately, has jumped on the bandwagon of attacking climate scientists personally. In this article, reprinted in the internationally distributed Guardian Weekly with the title Research red in tooth and claw journalist Fred Pearce makes numerous errors. I was particularly annoyed at this reprinting of the article, since the weekly paper has more time to review content for errors. Here is what I've written to the editor:

Fred Pearce's long article "Research red in tooth and claw" is riddled with errors and misinterpretations. First, he identifies a paper by Lars Kamel as the one referred to in stolen CRU emails because "It is the only one published on that topic in the journal that year" then goes on to explain how the paper was never published. Either it was published, or it wasn't. Second, he goes on at great length about how CRU's temperature analysis has not been recreated by anyone else, and cannot be because of lack of access to the data. While I have problems with the whole concept of lack of access to data, this is denialist spin. NASA's Goddard Institute for Space Studies has independently constructed a temperature record using mostly the same data sources, and all of its data is public, and there are other independent temperature records against which CRU's can be tested. Pearce's understanding of peer review is also confused. There is nothing improper in asking other academics for inputs into reviewing a paper, and this happens very widely. Nowhere does Pearce present evidence that confidences were breached.

As Pearce could easily have discovered had he done the job he's evidently being paid to do, despite Jones's threat to keep junk papers out of the IPCC, they are in fact cited as references. The "MM" paper referred to is unlikely to be the McKitrich and Michaels paper on urban heat islands because this was published in 2007, 3 years after email Pearce quotes. The paper concerned is more likely to be that by McIntyre and McKitrick of 2003 (also rubbish), which is in fact cited in IPCC reports.

In one of Pearce's most serious blunders, he fails to report the fact that half of the editorial board of Climate Research including the newly-appointed editor in chief resigned over the journal's lapse in standards at the hands of De Freitas.

Pearce says the fact that Jones was new to the IPCC process was no excuse for errors of judgement. How new is Pearce to journalism?

PS: has Rupert Murdoch bought The Guardian?

There is a deep problem here when a usually trusted news source starts behaving in this way. Science is traditionally conducted through the research literature for a good reason. Unqualified people can easily make mistakes when assessing a complex field. The hard thing is translating research into policy. Human society does not on the whole do that very well, and I am not proposing alternatives here. What I do however object to in the strongest terms is traducing science as a whole and vicious often ill-informed personal attacks on scientists.

If you agree, sign the petition.

Of Drama Queens and Climate Science

Yet another bunch of anti-science letters in The Australia, this time in response to an article reporting the UK's chief scientist as saying climate scientists are prone to dramatise claims.


It suits the tabloid press (including those who disguise this tendency with a broadsheet layout) to frame issues in absolutes and set up crazy debates. Anything any scientist says that sounds the vaguest bit extreme or contrary to known science gets blown out of proportion, and the 99.9% of scientific work that doesn't fit this model is ignored.

Try reading the scientific literature using Google Scholar instead of regular Google as your search engine. You'll be amazed to find the vast majority of accusations against climate scientists (here for example the claim that they underplay uncertainties) are totally unfounded. Try however to get an article into a newspaper that tells it the way it is rather than setting it up as a screaming match between drama queens and you will get nowhere.

I stopped buying newspapers that do this and only read them where I can find them for free (online, in coffee shops etc.). If more people did this they would get the message.

It's hard to get a real debate going in the online comments so I will repeat some of my points here.

Peter: your claim that the raw data is "destroyed and manipulated" is rubbish. Most of the raw data is freely available. The drama queen argument here is that if a tiny fraction of the data used by one research group isn't available then all of it isn't. NASA for example makes ALL their climate data and their computer code available (see here for theirs among others). Enough data has been available for years for any serious scientist wanting to undo the mainstream theory to do so. Instead, we get vicious personal attacks, theft of email and political arguments.

SW1: your claim that further flaws will come to light is not only plausible but almost certain, but so what? Published science is riddled with errors. A researcher develops a new theoretically superior instrument, and in the rush to get published, makes mistakes in calibration. A research group finds a new angle on a problem and in their enthusiasm at finding something exciting, miss an obvious error. Peer review is a filter but not a perfect one, and this kind of thing can slip through. Sometimes the reviewers are lazy, or have a predisposition to accept the paper's findings. That's why scientists check on each other. Let's take as an example the 2009 paper by Lindzen and Choi that purported to show a much lower climate sensitivity to CO₂ increases than the accepted range of 1.5-4.5°C. This paper was trumpeted over the blogosphere as yet another nail in the coffin of anthropogenic global warming. Now various analyses of their paper are showing that their approach was flawed, including one by Roy Spencer, who generally disagrees with the mainstream figure on climate sensitivity. There is at least one published response that shows the paper to contain serious errors. The drama queen version of events: Lindzen once again strikes a fatal blow against science, and the "corrections" can safely be ignored.

Another example is the claim that the oceans are cooling. NASA has an interesting story on their web site of how the science in that case unfolded. One of their researchers, Josh Willis, published results using a new technology, Argo floats, that showed the oceans were cooling. While he described this as a "speed bump" rather than a slowdown in warming, the blogosphere went gaga and many people still quote this result. What actually happened was some of the floats were under-reporting temperature, while the older technology used for older measurements was over-reporting temperature, and these two effects combined to create an appearance of cooling. Any cooling would contradict a range of other kinds of evidence, so had these results held up, it would have been a major challenge to the science. Instead, the problem turned out to be in the data. Again, nothing too unusual. But the drama queen interpretation is Willis overthrew the entire theory, and the conspirators worked hard in the background to reinstate it.

My focus here has been on attacks on the mainstream but of course there are people out there who make extreme claims about the risks of climate change. These people do not represent the scientific community, who as a whole are pretty loath to make extreme claims without strong evidence. But these stories too get the headlines.

The bottom line? Google Scholar is your friend. Whenever something implausibly extreme or contrary to accepted knowledge appears, go to the scientific literature. Preferably after waiting 6 months for the dust to settle.

Dealing with climate change

We have no option but to deal with climate change.

Either we deal with it or it will deal with us.

There is no serious challenge to the science behind climate change. The whole basis for the denial movement is not attacking the science, but attacking the scientists. Look for a real scientific paper attacking the science. There is a very small number of these, all of which have been debunked. Most other material is on blogs and the opinion papers of the mass media, replete with the language of personal attack. Climate scientists are "alarmists", "warmists", "catastrophists", and any other number of made-up appellations.

As with the tobacco and HIV denial movements, personal attack does not prevent the ultimate harm that denial of the facts causes.

Unlike tobacco and HIV, we are not talking about something that will ruin the lives of a few million people if not dealt with expeditiously. Rapid climate change is causing a rapid acceleration of extinction rates. The rapid depletion of the Himalayan glaciers threatens the food supply of a half a billion people: melt water from these glaciers feed major river systems of both India and China. And reduced agricultural output is likely to be worst in poorer countries.

Talk therefore of whether an emissions trading scheme should be contemplated in the absence of any serious alternative is crazy. The Australian government's scheme falls far short of any reasonable solution, but their main opposition is opposing it without offering an alternative.

Given that, how will climate change deal with Australia?

There are two possibilities: Australia's inability to face up to the problem could be a worldwide phenomenon, in which case we face a period of growing environmental disaster: death of the Great Barrier Reef, collapse of food supply in India and China, famine in Africa. If on the other hand the rest of the world gets its act together, Australia as the world's biggest coal exporter and the country with the highest per capita CO₂ emissions will be left high and dry, and as uncompetitive as a country that did not foresee the trend away from horse-based transport a century ago.

Climate of Fraud Part 3

Another round of letters in The Australian, another round of debunking.

Let's focus this time on the comments, since I had a letter published.

First, Greig, a regular contributor:

Philip Machanick in the past you have vehemently argued that it is CO2, and not the sun, which is the major driver of observed global warming. And now, year after year, global temperatures stubbornly refuse to climb in correlation with CO2, and so you revert to using the sun to explain. Like Dr Kevin Trenberth you are wedded to an opinion which you want the data to fit, and like him you saying “there should be even more warming . . . the data are surely wrong.”

Greig, CO₂ drives the trend for the simple reason that there is no trend in natural factors, whereas atmospheric CO₂ is increasing exponentially. The science says there's a logarithmic relationship between climate sensitivity and CO₂ levels, implying we should over the long term see a linear increase in temperatures if the basic science is correct. The purpose of detailed models is to narrow the range of uncertainty, not to predict the basic effect. We are currently at a solar low, but the multi-decadal trend in solar insolation is flat. Same for ENSO. We had a big El Niño in 1998 and more recently a La Niña. The net effect of these things is zero, but pick a time when they are all pointing up or all pointing down, and you get a short-term high or low. Trenberth's comments were in a private email, a musing with colleagues, who pointed him to evidence that he was wrong. Much as I am loath to use stolen material, since it's out there and being misused, here is one of the responses by a colleague to Dr Trenberth:

I look at this in two ways. The first is to look at the difference between the observed and expected anthropogenic trend relative to the pdf for unforced variability. The second is to remove ENSO, volcanoes and TSI variations from the observed data.

Both methods show that what we are seeing is not unusual. The second method leaves a significant warming over the past decade.

Next, Ian 11:39am (you will have to page through the online comments to find all of these):

Philip Machanick Did the models predict that the global temperature increase would lessen over the last 10 years?. That these years are the hottest 10 since whenever really isn’t a valid argument despite those such as v who claim it is. Fact is that the increase over the decade has been less than predicted. Now why is that? [snip] And incidentally Professor Latif who is a climate scientist, has made and is still making comments that suggest the world is not about to turn into a heat ravaged dust bowl just yet

No, the models do not make predictions for as short a period as 10 years. But see above about removing known effects like ENSO. See also here how solar output does not correlate with temperature as it should if there were no interfering artificial effect, and another very clear illustration here of how temperature and solar output have diverged. Also, I illustrate that you can add a strong artificial upward trend to a temperature record that is essentially flat and still find a 10-year period with a strong downward trend.

You are wrong in your interpretation of Latif's comments, as I've indicated before on this site and in more detail elsewhere.

Philip (not me) 2:27pm:

Gosh all these people ignoring the hacked emails and documents showing what nasty little con artists the CRU are. In ignoring them they are behaving just like ahh… I don’t know maybe like holocaust deniers.

Or maybe they don't agree with the principle of stealing private communications? How about challenging the deniers to reveal all their emails for public scrutiny? That would make for interesting reading. Things colleagues say between each other are often unguarded, said without thought as to correctness or unintended consequences. Why should we all take exquisite care in every private conversation that we don't ever say anything we don't really mean, in case someone is eavesdropping? What kind of world would that be?

de Brere 01:27pm

Philip Machanick. Spot on, you do validate a theory by testing its predictions against observations. However, the prediction-observation link needs to be 100% accurate once other factors are taken into account. Otherwise the other unknown or unaccounted factors may themselves be the source of the observed variation, and not the theorised cause; in this case, GHGs. In that respect, all the tens, hundreds of billions invested in the “science” of AGW in the last two decades has spectacularly failed to come up with models that provide even moderately accurate prediction (or explanation) of observed climate change; far less of the individual weather events that go to make CC up.

De Brere, I've covered your misconceptions about how science works earlier. For your benefit I'll summarise again briefly here. Science in the real world is never 100% exact. Handling errors and uncertainties is a standard part of applied science. If you want exact answers, do pure mathematics. As for these astronomical numbers for the cost of climate research, what are your sources? I found a comparison of current US military and climate change-related funding and:

1. excluding the recovery package, US military spending outranks climate spending 90:1
   
2. even within climate spending, much more is spent on energy that on climate science
   

And finally, climate science is not about predicting tomorrow's weather. ENSO and the solar cycle are not fully understood, and have a bigger short-term impact than increasing CO₂. However, if you keep increasing greenhouse gases long enough, you create a trend that breaks out of short-term variation. Where's your evidence that that is not happening? How do you explain that contrary to the strong downward pull of natural factors the last 10 years or so, the temperature trend is still slightly up?

Having dealt with comments about my letter and comments, I'll let Stephen Morgan 05:47pm have the last word:

Okay - so apparently these hacked emails are SO important that they change the case. So, can somebody tell me what they say. Only not just the ones that apparently give reasonable cause to doubt a few scientists, but every single one of these thousands of emails?

If you think the emails matter, then you will obviously review ALL of them, understand the context of ALL of them, be able to balance the views presented by ALL of them, and be able to come up with a reasonable theory that is supported by the contents of ALL of them.

To do otherwise is selective, prejudicial, and inevitably driven by a desire to support or discredit a specific cause rather than to seek any reasonable explaination.

It’s called denialism for a reason - it is the methodology of those who seek to deny rather than to discover, to obfuscate rather than to explain.

AGW IS widely accepted because it IS widely understood by those who support it. It is popular because it is far more reliable and consistent than any other option.

Is it proven - NO! Is it undeniable - NO! Just do as I asked - come up with something positive that offers an alternative, not just continued hackneyed attempts to sink a battleship with a pin-prick!

Well, quite.