Tuesday, April 24, 2012

On Manufacturing Doubt for Levitus 2012

Willis Eschenbach, still trying to kill off Levitus et al 2012, asks:
Regarding the climatology, they say that it is from the World Ocean Atlas 2009 (WOA09), viz: ” … statistics at all standard levels and various climatological averaging periods are available at http://www.nodc.noaa.gov/OC5/WOA09F/pr_woa09f.html “

So I went there to see what kind of numbers they have for the monthly climatology at 2000 metres depth … and I got this answer:

"The temperature monthly climatologies deeper than 1500 meters have not been calculated."

Well, that sux. How do the authors deal with that? I don’t have a clue.
In fact, the authors say how they do this in their Supplementary Material, which has not been published yet. (I guess now it's too much to wait until scientists' results are actually published before attempting to destroy them.) But a preprint version is here. The authors write:
"The results describing the variability of ocean heat content shown here are based on 745 gridded (1-degree latitude-longitude grid), interpolated fields at standard depth 746 measurement levels have statistical estimates of reliability associated with them."
(emphasis mine). And while the the monthly climatologies below a certain depth aren't available, the quarterly climatologies are -- at least, for every depth I've checked below 1500 m.

Does it matter if quarterly climatologies are used instead of monthly climatologies? I guess it's a fair question, but it's sufficiently technical that I would be sure to first ask the people who know the nuts-and-bolts of the data and analysis before writing a few little subroutines, making a lot of assumptions, and then labeling their work with names like "goofy." (It's also exactly the kind of things that get picked through on peer review.) But then, that's why I enjoy primarily doing journalism and more in-depth writing than being a blogger, especially a blogger in a realm that is really only looking to manufacture doubt about the science rather than fairly evaluating it (which is always more complicated than a few hours work.)


Dano said...

Poor Willis. You have to give him props for trying, though.



Piltdown said...

Wouldn't it be just as fair to say I guess now it's too much to wait until scientists' results are actually published (and checked) before attempting to cite/promote them? :-)

Does it matter if quarterly climatologies are used instead of monthly climatologies if you said in your paper that you used monthly climatologies? It surely makes it harder to figure out what they actually did.

The suggestion to ask the researchers for the data and explanations is an excellent one, but I have heard that some climate scientists get upset to have climate sceptics calling them up with enquiries all the time. Why should they make the data available to us, when our aim is to try and find something wrong with it? Very reasonable. It has been suggested that the answer is to publish all the data and code with the paper so there can be no doubt about what was done, and so no questions should arise. But that upsets some scientists as well, who feel they have proprietary interests in their data.

I think it is a very interesting question. What do you think the ideal solution would be?

aaaaa said...

Ha pathetic trashing of papers WUWT doesn't like and so obvious.

I notice Svensmark's paper receives a very different treatment.

Piltdown said...


Quite so. Confirmation bias is a human universal. The scientific method is founded on this recognition of our own fallibility, and is in some ways no more than a catalogue of the extreme measures we are forced to take to reduce its effect. Double blind. Randomised trials. Testable predictions. Eliminating alternatives. Systematic challenge. Experimental evidence.

That's why it is so important to talk with people who disagree with us. They have different blindspots, and can reveal for us our own errors, as we reveal theirs. Only through discovering and correcting our errors can we progress; only through surviving well-motivated attack, in circumstances where we have reason to believe any flaws would have been revealed, can we gain justified confidence in our beliefs. We owe a debt of thanks to our opponents - without them we would all descend into dogma.

We trust you to find any errors in the Svensmark paper. You'll do it far better than we ever could. That's why we take part in the debate.

Anonymous said...


Strangely enough, Willis also criticized the Svensmark paper and seems to not really buy it:


Kind of dampens the narrative that he only disagrees with papers that promote the consensus.