Happy Birthday to Watts' paper!

[Don't miss the 2nd birthday!]

why-are-we-waiting About a year ago the entirety of the intertubes were rocked to their foundations by an announcement of epochal proportions: WUWT publishing suspended – major announcement coming. Or so we were told. Speculation was rife: had AW finally found those pix of Mann's Prince Albert? But exactly a year ago the Truth Was Out and it all turned out to be very dull - it was just a paper preprint [*].

Most scientists of any kind of quality manage to produce at least a paper a year, anyone with ambitions for their step up is looking at two or more, in decent journals. Even a blog scientist (a bit close to the knuckle there, but the bit about guest posts about climate elves being utter sh*t resonates. AFAIK the true Blog Science manifesto is to be found at denialdepot) shouldn't get too excited about a preprint.

However, a year has now passed with little public evidence of action. Perhaps the climate elves are working behind the scenes. I imagine that AW hasn't given up all hope, because the "2012" graphic is still proudly on his blog. I wonder if any of the Watties ever notice it and wonder, or is it like the underpants on my bedroom floor, once they've been left lying there for a while they become invisible?

[*] The disparity between the mighty trailering and the feeble reality has lead those few who still think AW has a strong grasp on reality to speculate that Something Else was originally intended, and when that Something Else went sour, a half-baked preprint had to be rushed out to fill the gap. We may never know.

Refs

* Top Physicist Withdraws Support For Climate Sceptic Professor Sacked By Australian University says JM
* 2014/06: we’ve spent two years reworking it and dealing with those criticisms. Our results are unchanged and will be published soon sez Watts in "Reason".

More like this

*puts his party hat on*

Maybe this time, because an "opponent" remembers, AW will try to rush something into E&E ?

You evil Englishman!

You know that congratulating someone in advance brings bad luck. :-) (Or is that just a German superstition?)

I hope Watts will submit, even if only to E&E, it would be interesting to see if there are any changes to the authors list.

By Victor Venema (not verified) on 29 Jul 2013 #permalink

Another overdue paper.
As far as I remember BBC weatherman Paul Hudson, once carried a report that private weatherman, Piers Corbyn , was just about to publish a paper on his own theory of global warming. Any news?

By deconvoluter (not verified) on 29 Jul 2013 #permalink

Now now, Watts is a computer guy and vaporware is a long and proud tradition. More seriously Eli has Willard Tony to thank for introducing him to Victor's blog.

By Eli Rabett (not verified) on 29 Jul 2013 #permalink

At the time, there was a rather pathetic or hilarious (depending on your viewpoint) "Anyway, now that I’m drawn into this, I’ll have carry out the TOBS analysis, which I’ll do in the next few days (at the expense of some interesting analysis of Esper et al.)" [Note: I de-href'd your text, because it was text, not a URL -W] about this at CA. Steve McIntyre, who was apparently added as a co-author to Watts's paper at the last minute and without his knowledge, expressed some consternation over the whole thing. He also said this:

Anyway, now that I’m drawn into this, I’ll have carry out the TOBS analysis, which I’ll do in the next few days

Does anyone know whether Steve ever got around to that? I don't recall ever seeing it mentioned on CA after that first thread. For sure there was nothing posted "in the next few days"...

Remember the context, though. These guys imagine they understand how the world works.
Ferrara
Kessler

By Hank Roberts (not verified) on 29 Jul 2013 #permalink

I see WMC referred my pointer to Graham Readfearn's latest article on the Salby affair, now "headed to court,," where alby ahs often been, although as far as I know, not before in Australia. He didn't make out too well in US Federal and Colorado courts, as per Murry Salby: Galileo? Bozo? Or P.T.Barnum?
The bozos were elsewhere. For 4 days, certain blogs had a party, with 1500+ comments.
Salby was: Galileo, Copernicus or Einstein.

Macqiurie (and/or mainstream science):
criminal, dictatorial, barbaric, Orwellian, Nazi, Stasi, Deutsch Physik, Lysenkoism, Stalin's Finest . ...
and I've only annotated 40% so far.

But then the party ended, and suddenly nobody wanted to talk about it any more. Bu as With Watts, the Internet Doesn't Forget, especially with a little help from WebCite.

By John Mashey (not verified) on 29 Jul 2013 #permalink

http://pps.sagepub.com/content/7/6/615.full

Restructuring Incentives and Practices to Promote Truth Over Publishability
doi: 10.1177/1745691612459058
Perspectives on Psychological Science
November 2012 vol. 7 no. 6 615-631

(hat tip to Metafilter)

By Hank Roberts (not verified) on 29 Jul 2013 #permalink

Maybe this time, because an “opponent” remembers, AW will try to rush something into E&E ?

Our last paper was in the Journal of Geophysical Science. And if you think we have "forgotten", please allow me to disabuse y'all of that notion. If you think I intend to let over half a man-year's effort slide off the table edge, you need to think again.

It's just a matter of getting it right. That takes time. If things work out, those who, oh, maybe don't like us very much are going to scrutinize my ratings and maybe even go online and vet the metadata. And I want the error rate to be both minuscule (I can't hope for 0%) and inconsequential.

As for the authors' list, we'll probably be subbing JN-G for Mac (the former of whom was part of Fall et al.). We need a stats guy with more chops than I got for our error bars; he'll do the monte carlo for us.

You know that congratulating someone in advance brings bad luck. :-)

Scoffing at someone ahead of time, OTOH, brings them good luck; thanks. #B^j

Does anyone know whether Steve ever got around to that? [Doing the TOBS analysis]

No need. Sometimes the best way to stay out of trouble is to stay away from it:

The erstwhile MMS (now HOMR) has TOBS stats. So I simply dropped all the PM - AM observation changes as well as AM - PM. Therefore, there is no need to analyze the vicissitudes of TOBS at this time.

It took a chunk out of our sample, but we still have ~680 samples remaining. I retain only AM - AM and PM - PM observation times.

By the way, I find that TOBS-bias is, indeed, quite severe, and needs to be factored in (or in the case of this paper, "out").

P.S., We also account for MMTS conversion, another legitimate (though a lot more minor) concern, using Menne (2009 & 20010) as a basis.

By Evan Jones (not verified) on 29 Jul 2013 #permalink

The disparity between the mighty trailering and the feeble reality has lead those few who still think AW has a strong grasp on reality to speculate that Something Else was originally intended, and when that Something Else went sour, a half-baked preprint had to be rushed out to fill the gap. We may never know.

The observation that heat sinks not only increase temperature (already well known) but also increase temperature trend (sic) is a very major finding.

And if true, then the current NOAA/NCDC homogenization procedure becomes farcical (obviously).

[If true, it will be interesting. But that's a big caveat. I still think it would have been better to get it right than to rush out a preprint. Given that its going to take more than a year to bash the preprint into publishable shape, I don't think "half baked" is too unreasonable. Note that is a description of the state of the preprint, rather than a judgement on the science it contains.

BTW - do you find it at all odd that there is so little discussion of this paper at WUWT? The last ref to it appears to be Aug 2012 - just below the mods snipping some other stuff. Are WUWT really so incurious about this major matter? -W]

By Evan Jones (not verified) on 29 Jul 2013 #permalink

Dear, I got quoted *giggles*

Mr Jones, I didn't say you forgot, I implied that I thought maybe Mr Watts wasn't expecting someone to remember and thus would work on the draft again to get an actual submission. It is not the best motivator, but if Mr Watts and you can find the moral strength to actually submit I'm sure you will get the cheerings you deserve for that.

However, I wonder why it takes so much time after the draft announcement to get something submitted - I naively thought that, since there was an announcement, there was something almost ready - after all, Mr Watts has clearly shown he frowned upon empty PR coups ... was there a problem with stats ?

And could you tell me why you don't seek the help of Mr McIntyre if you need a "good" stat guy ? He's the mann for the situation :]

I really hope to see the submitted paper. I will read it thoroughly with delight.

[I found a nice comment from McI here. I can understand his dismay at suddenly finding himself a co-author. But AW does have a tendency to take without permission -W]

Evan,

Could you elaborate on 'The observation that heat sinks not only increase temperature (already well known) but also increase temperature trend (sic) is a very major finding.'

I ask because ten months ago you seemed to be saying the opposite: 'I am suggesting that overwhelming presence of artificial heat overwhelms and dampens the trend compared with the trend of Class 3 or 4 stations.'

http://scienceblogs.com/stoat/2012/09/01/evan-jones-speaks/

Evan:

"By the way, I find that TOBS-bias is, indeed, quite severe"

That's good progress, indeed, catching up with mainstream over the last year!

I ask because ten months ago you seemed to be saying the opposite:

Not quite, but sharp observation. What I was saying was that the effect begins to overwhelm the trends with the very worst stations. Not so much that Class 5 trends are lower than Class 1\2, but they are noticeably lower than Class 3\4.

So, yes, bad siting does increase trend, but it is not a linear effect.

… was there a problem with stats ?

Well, we had to deal with TOBS and MMTS conversion. And, of course, re-review everything. That takes time. That's done now, so we are moving forward.

That’s good progress, indeed, catching up with mainstream over the last year!

TOBS was always an issue. The question was merely when and how to deal with it. we were going to tackle TOBS in a followup paper, but we decided to deal with it now.

By Evan Jones (not verified) on 31 Jul 2013 #permalink

BTW, we did not Include St. Mac as co-author without his permission.

[Interesting comment. I can't reconcile that with:

Steve McIntyre: If the paper is submitted anywhere, I will either sign off on the analysis or not be involved. I didn’t “allow” or not “allow” anything in respect to the discussion paper.

Can you? -W]

By Evan Jones (not verified) on 31 Jul 2013 #permalink

Ok, thanks for the clarification Evan.

EJ says: we were going to tackle TOBS in a followup paper, but we decided to deal with it now.

Probably a good decision. A paper along the lines of what you're doing seems unlikely to be worth publishing if it doesn't deal with TOBS one way or another.

EJ:

"TOBS was always an issue."

Not consistent with things said a year ago ...

Watts PR:

"The new improved assessment, for the years 1979 to 2008, yields a trend of +0.155C per decade from the high quality sites, a +0.248 C per decade trend for poorly sited locations"

If you guys "always knew TOBS was an issue" you had no business calling your first draft an "improved assessment".

Probably a good decision. A paper along the lines of what you’re doing seems unlikely to be worth publishing if it doesn’t deal with TOBS one way or another.

Mmmm. Perhaps.

Yet neither Menne et al. nor Fall et al. accounted for TOBS and they passed peer review and have been cited bu other papers.

But, yeah, it's best dealt with now.

If you guys “always knew TOBS was an issue” you had no business calling your first draft an “improved assessment”.

Whatever.

The "improved assessment" refers to the use of Leroy (2010) as opposed to Leroy (1999). So it was an improved assessment - and has been improved further since then.

TOBS does somewhat narrow the gap between Fully Adjusted and raw-well-sited-(with-an-upward-MMTS-bump). But there is still a wide gap. And it made near-zero difference between well and poorly sited stations.

Not consistent with things said a year ago …

Then read the actual paper. It says TOBS is an issue and that we would deal with it later.

By Evan Jones (not verified) on 31 Jul 2013 #permalink

Menne et al. and Fall, Watts et al. did account for TOBS. They used the standard NOAA homogenized data (USHCN) and that data is homogenized by first appying TOBS corrections and then performing statistical relative homogenization. Even if they did not explicitly perform the TOBS corrections, the statistical homogenization will take care of the inhomogeneities due to the time of observation changes.

Fall et al. also showed results for raw data, but it would likely not have been accepted for publication if it had not included the results for the homogenized data. If you want to write a paper that leaves out results for homogenized data (as the manuscript version of last year did), you will have to proof that your boss Watts is right and that "homogenization just smooths the data" or at least that homogenization does not improve the temporal consistency of the American temperature data.

If you manage to do that, please mail me, I would be very much interested, that would be a very interesting finding as it goes against everything we know.

Not taking the homogeneity of the data into account (including TOBS) is like making a study on the relationship between diet and cancer, but not studying whether the confounding factor smoking differs between the groups studied. You may get lucky and hardly have any influence of smoking or TOBS, but not taking it into account is bad science. Such confounding factors are not details to study in a later paper.

By Victor Venema (not verified) on 02 Aug 2013 #permalink

Evan Jones: As for the authors’ list, we’ll probably be subbing JN-G for Mac (the former of whom was part of Fall et al.). We need a stats guy with more chops than I got for our error bars; he’ll do the monte carlo for us.

Friendly advice, it is a good idea to keep Steve McIntyre on the authors list. Even if you could argue that he did not do enough work to be on the authors list, reviews can be very difficult if the reviewers know that an insider does not want to be on the paper.

Then the reviewers will wonder what inside information that former contributor has; inside information that made him prefer not to be on the author list.

It gives the impression that the paper is not honest. Especially when that happens during the review, the paper is typically rejected.

By Victor Venema (not verified) on 02 Aug 2013 #permalink