Anthony reports today on a NOAA “Talking Points” document which purports to rebut Anthony’s Is The U.S. Surface Temperature Record Reliable?, without deigning to cite
Watts, A. (2009). Is The U.S. Surface Temperature Record Reliable? Downloaded from http://wattsupwiththat.files.wordpress.com/2009/05/surfacestationsreport_spring09.pdf.
or using Anthony’s name. They seem to have adopted the Team tactic (Hansen in particular) of refusing to even speak the name of critics.
U.S. federal policy on research misconduct defines plagiarism as follows:
Plagiarism is the appropriation of another person’s ideas, processes, results, or words without giving appropriate credit.
I invite readers to review the NOAA “Talking Points” document with that definition in mind.
I have one very serious criticism of Anthony’s post on the issue. He left out an important reference to Knowles et al 2006, who provided a definite comment on the citation issue in a 2006 conference.
Climate Audit 
54 Comments
Since an analysis by your own readers confirms the talking points conclusion–no bias–what is the issue? The fact that the talking points replicate an analysis performed by John V, but that Anthony has been too scared to perform, even now that he’s breached the 75% of the network level (arbitrary in the 1st place), seems to argue against the plagiarism charge.
Re: bigcitylib (#1),
Regardless of whether the analysis in Watts(2009 is right or wrong, it is entitled to be properly cited. JOhn v’s analysis, such as it is, has nothing to do with that.
I’ve commented on John V many times and do not agree with your spin on the results.
The survey at that time showed a substantial difference between CRN1-2 and CRN5 stations, confirming – as against Jones et al 1990 – that there is a bias that is greater than 0.1 deg /Century.
In the U.S., NASA GISS made an attempt to adjust for this bias, but NOAA and CRU did not. As a result, as I’ve observed on several occasions, there is a very large discrepancy between GISS and NOAA trends in the U.S ,with GISS much lower than NOAA. This is the sort of thing that climate scientists ought to have tried to reconcile. Anthony’s survey gives objective evidence that the NOAA US index runs too hot.
In this case, the GISS adjustment was fairly decent. The GISS US index was not used in the EPA GHG study – it used the hot NOAA index.
I’ve also observed on many occasions that the GISS US adjustment is not the same as what it does in the ROW (where GISS runs hotter than other indices.) People have tried to bait-and-switch.
In any case, they cite the surface stations website. What does Anthony want? A gold star?
What does Anthony want?
Courtesy?
So, I’m looking at the study that the talking points do reference. (EXAMINATION OF POTENTIAL BIASES IN AIR TEMPERATURE CAUSED BY POOR STATION LOCATIONS BY THOMAS C. PETERSON).
This study looks at the 8 stations that Davey and Pielke examined in detail. They use two with “good siting”. They throw away two who’s poor siting were related to gravel roads adjacent, and use for comparison four stations that are close to obstructions.
The graph of the raw data for the obstructed stations shows a negative trend relative to the unobstructed stations, but with their sizable adjustments to location, the adjusted trend matches the unobstructed stations.
No error bars, or estimate of error involved (total of 6 stations studied).
They conclude “The results presented here clearly support the theory that, if poor siting causes a bias, homogeneity adjustments account for the biases and contradict the hypothesis that poor current siting causes a warm bias or even any bias in the homogeneity-adjusted U.S. temperature change record.”
Yet they intentionally disregarded the poorly sited stations that would potentially show a warm bias (nearby gravel roads).
From this study of 6 stations, NOAA is issuing talking points refuting a survey of 1200+ stations.
Interesting.
Steve,
Why cite the report? The data used comes directly from Surface Stations and the report itself could not have been used in the analysis.
Re: bigcitylib (#6),
A variety of reasons.
One of them is a policy that scientists should not “appropriate another person’s ideas, processes, results, or words without giving appropriate credit.”
The Talking Points are entitled “Is the US Temperature Record Reliable?” and is clearly directed against Watts (2009). Is The U.S. Surface Temperature Record Reliable?.
I do not see any citation of Watts(2009) and, without such a citation, I do not think that they have given “appropriate credit” to the person who raised the question.
Re: bigcitylib (#6),
It’d be interesting to apply this to the Mann data. After all the data Mann used is directly from others’ work. Would it be right to not mention Michael Mann or MBH9x when discussing temperature reconstructions? I think you’d be at the front of the line of people complaining. Next time, try thinking about how your complaints would look if the shoe were on the other foot.
Re: bigcitylib (#6), In addition to what Steve just wrote, the “study” would clearly be poorly researched if a major data source that addressed the research question was ignored. It would mean the scholarship backing the analysis was lacking in diligence. It is bad science. In my own field we talk of “gray literature” which is composed of technical studies that have not been formally published. There is simply too much work being done to publish it all. These documents often contain important primary data and even innovative thinking that could be important in the field. Scholastic due diligence means you have to research this. It may mean using email, or the telephone instead of a library, but it still will inform and improve a study. Anthony Watts’ study may not have been formally published, but it is widely known and contains a mass of primary data that work supporting the “Talking Points” lacks.
Also, if you look at the “Talking Points”, the title quite clearly evokes the WUWT investigation, while not using quite the exact title. Note what is italicized. This really poor practice and mannerless and rude as well. Even journalists usually know better. It smacks of condescension and politics.
…which is to say that the purpose of the report (it seems to me) is to discuss AW’s ranking methodology. The NOAA thing simply assumes that, and APPEARS to pull station data from the SS website. The report does not really provide detailed data on the stations that appear therein.
If the report pulls anything (data, or simply the idea that they feel needs refutation), then professional standards require a citation.
By not citing, they imply that they haven’t pulled any data from AW. If they haven’t pulled any data or conclusions from him, then they are refuting someone else, and AW’s data and conclusions can be considered to be unchallenged.
They can’t have it both ways. Either they are attempting to refute his data and analysis and they should say so in a citation (where did they get the data? where can an unbiased reader compare their refutation with the data?) or they are refuting someone else (and should cite that) or they are simply shouting into the wind.
Oh, wouldn’t I love to cross-examine an NOAA witness on the stand about that Talking Points memo!
Amazing. They’ve never done the quality control that any marginally competent scientist would require and they use that lack of quality control as a defense! In essence, they say “we don’t think there is any merit to the claim that the records are biased, because no one has ever bothered to find out!”
Re: stan (#10),
Actually, I would have said they are saying “You can’t prove we’re wrong!”, when you have asked for evidence they are right.
Re: Neil Fisher (#33),
Yes.
The line I use here to show this up is:
“Michael Jackson is the 2nd coming. Now you prove he’s not”
Proving a negative is impossible.
Re: ianl (#34),
Actually its easy.
“I am not a giraffe and I can prove it”
Proving a Universal negative for the existence of something, like Santa Claus or the Tooth Fairy, is impossible.
Also, its entertaining to watch RC commenters indulging in direct denial of prima facae evidence that the US Surface Temperature record is severely contaminated with warm biases. Clearly they haven’t been visited by the Irony Fairy in a long time.
A NOAA talking point:
Sure they do. They must have had to look long and hard to find the Tucson site.
They’d be better off just conceding the bad sites than arguing that they “worked diligently” to choose this site.
1) They ref. the SS site. That’s where the data has been pulled from.
2) The questions re. siting issues and their effect on the temp record do not origonate with Watts (Pielke Sr.s been on about this for years, as the Peterson paper notes). Hence, any refutation doesn’t have to be directed at Watts.
3) This is all blowing smoke to cover the fact to cover the fact that yet another analysis of the SS data shows no discernable bias.
Steve: please do not use this offensive acronym for surfacestations.org.
1) They mention surfacestations.org. They do not provide it as a citation or reference in any format that would allow an unbiased reader to find the original claims that they are refuting. …
Without the citation, you are simply blowing smoke. With it, you provide a means for someone to lookup what you might be talking about.
2) They don’t bother to cite Pielke either.
3) Your basis for this claim is? (John V’s conclusions from a couple of years ago? NOAA’s talking points? Divine inspiration?)
(Sorry Steve. I know I really shouldn’t feed the trolls.)
My favorite part of the talking points was this:
“However, they could evaluate urban bias and found that once the data were fully adjusted the 30% most urban stations had about the same trend as the remaining more rural stations.”
So they adjusted the rural stations upwards until they got a match?
But the “talking paper” itself raises new questions:
Work the numbers. 70 stations out of 859 (70 percent of the 1221 stations) could be classified as good or best. Still leaves 1151 stations (about 94% or so) listed as either poor (3, 4 or 5) or with unknown siting problems.
They agree that the ratings AW gives the stations are accurate.
Question, then: when will NOAA send out people, with the proper “level of expertise” to check out the 1151 stations that do not fall in the “best or good” ratings?
Props to the volunteer force that GAVE them the large scale site evaluation information.
Where is the third series that shows the comparison between the full data set compared to the remaining 789 or so stations that surfacestations.org classified with 3, 4 and 5?
In regard to the states listed: Is the lack of good or best stations because those states haven’t been fully surveyed, or have ALL the stations in those states fallen into the 3, 4 or 5 category?
But as everyone keeps saying, the US is only a small percentage of the world’s surface and cannot be representative of the Earth’s Total Temperature. What siting bias is the ROW’s temp stations imparting in the trend?
If we could find a list of the world’s stations used, we could see for ourselves.
One more statement from the “talking paper”:
Such as New Mexico, one of the states listed in the “no good stations” list. Or Tucson Az, as shown in the picture.
Somebody who went to Surfacestations.org could easily see just how many stations there are in the southwest, and how well their “modernization and maintenance” efforts are working. Any takers?
From the actual report – as linked in the first paragraph – the citation should be:
I suppose you’re free to adopt whatever parenthetical citation method you choose, but if I remember my MLA manual correctly, since this is a report published by a private organization, one can either use end-of-line [(Watts 2009, ##) or (Is the U.S. Surface Temperature Record Reliable?)] style or refer directly to the name of the paper in the sentence referencing it. I don’t have the manual in front of me though, so someone else would have to verify.
The plagiarism charge boils down to whether or not the NCDC is presenting Anthony Watts own, original (or duly referenced) ideas as their own. The issues addressed by Watts are presented by many others, and the NCDC talking points memo doesn’t seem to address anything that Watts pinpoints in particular. Furthermore, Watts himself cites NCDC and NOAA documentation when discussing certain points in his paper – in these cases, unless Watts is building on top of these documentation to construct new points, is a direct citation even necessary?
Sure, the talking points memo would be better served by a comprehensive citation of where all of its data come from, and sure, it should cite the report which it seems to be addressing. But it’s hardly plagiarizing… unless failing to cite your organizations own previous work is plagiarizing (which some do consider it to be).
As has been pointed out at WUWT, Thomas C. Peterson is apparently the unattributed author of the talking points “response.”
A comment of mine on realclimate.org wasn’t posted where someone on there said that surfacestations.org shouldn’t be trusted as the photographs could have been taken at any angle making things look closer than they really are. My post of “I never thought I would see the Capricorn One defense used on RC” didn’t make it past the censors.
They are feeling the heat from Surfacstations. I don’t know what the final conclusions will be but after seeing so many bad station sitings on WUWT, it can’t be good. These guys are nervous enough they don’t even want to mention WUWT, it’s no small point.
The conclusion that I gahter from reading the rebuttal is that their answer to the question they pose is:
We don’t know
Expertise in taking pictures of stationary objects form various angles. Gee, that’s a tough one. Is there a Ph.d for that?
It just keeps getting stupider and stupider.
You wonder where and what they spent their billions on don’t you?
And, just as clearly, there is no indication that the entire data set taken from 1221 stations is necessary (or building a whole new series of stations, either).
Since 70 stations, covering 43% of the country (and missing at least 10 states) matches the record, what is the data from the other 1151 stations being used for?
Somebody should take those 70 stations, and plot them out to show just how much U.S. coverage NOAA considers acceptable.
Re: henry (#25), 70 is fine. Hell, Steig did it with only 42 in Antarctica – and that’s 1.5 times the area of the US.
Re: Ryan O (#27),
Yeah, but try running that 70 station database through the Steig mill, and use the resulting maps to show the U.S. warming trend – Nature may just use it as a cover story…
Re: henry (#29), Peanuts, my man, peanuts. I’m going to try it with just 1.65 stations, baby. Teleconnections galore!
Re: Ryan O (#27), Don’t need any 70 stations. Four works: one in D.C. one in St. Louis, one in Denver and one in LA. We would know just about what we know now.
Re: PhilH (#35), Especially if we use the 1200 km smoothing that seems to be so popular.
If 1200 km smoothing is accurate enough to capture the Arctic’s warming trend, then it should be good enough to capture the U.S. trend…
Since it is obvious that they are sending this email in response to the Watts document, they should reference this. In my view, as an academic, it is not a marginal call. Howver, not referencing properly is a pretty common sin, in fact, so one should not be too surprised. Speaking out is the correct response—make a ruckus and you will be heard. Honestly, I cannot imagine how the climate-science establishment is not hugely embarrassed by the fact that Watts’ efforts are so obviously needed and had not been undertaken. ‘No funding’ is not an excuse unless they can demonstrate that specific requests for studies similar to Watts’ were requested multiple times and rejected. Otherwise their priorities have not been well-considered, at least so far as this goes. I suspect everyone thought Hansen was taking care of all this, so it was OK, and it is no doubt bad form in such circles to question Hansen.
Re: cal browser (#26), Agree.
It all seems very petty and pointless. In fact, if they were actually trying to be smart about the PR, they would lavish praise on the surfacestations.org effort for pointing out where research dollars are needed, and promptly apply for grants to survey stations, upgrade stations, and conduct studies and write papers attempting to quantify and correct inaccuracies that might or might not have been imparted through siting issues.
When playing hardball, pros impart spin.
But, as many different individuals participated in the site evaluations, with varying levels of expertise, the degree of standardization and reproducibility of this process is unknown.
Re: Tamara (#31),
You can say the same thing about the original measurements – so you’re probably proving a little more than you want.
What really stunned me from the “Talking Points” was this:
Q. What can we say about poor siting’s impact on national temperature trends?
A. We are limited in what we can say due to limited information about station siting.
Here is a government agency which is given millions of taxpayer dollars each year saying that they have “limited information about station siting”! Come again? I wonder what these people do all day long? They seem incapable of actually, you know, VISITING a site to find out all the information they need to know! Or maybe making, what’s it called, a PHONE CALL??
Another thought – why does NOAA even need talking points? Are people there that ill-informed of the condition of their historical climate network? Perhaps these were sent around so NOAA employees wouldn’t slip up and say the “wrong thing”.
All in all, truly lame and embarrassing CYA by Tom Peterson…
A couple of observations about the Talking Points graphic that need to be examined.
When I last visited this data in 2007 and briefly in 2008, USHCN v2 had not been released. It was scheduled for release in mid-2007 and was hugely delayed.
There are changes in the network for some reason. The number of stations is reduced from 1221 to 1218, but only 1159 stations are common. About 80 stations have been dropped and 80 added. Has anyone seen an explanation of this?
Second, the data at the USHCN v2 website no longer shows raw data or TOBS data, an Oct 2007 version remaining available at the FTP site
ftp://ftp.ncdc.noaa.gov/pub/data/ushcn/
All we are presently seeing is the new adjusted method. The readme says:
One of my adverse impressions of the predecessor FILNET method was that it sometimes allowed “bad” stations to adjust a “good” station upward. I recall seeing something like this Grand Canyon and Tucson, which was one of the sites that I examined in the Y2K thing.
USHCN v2 uses a homemade methodology for adjusting data. As I understand it, the method dispenses with TOBS metadata adjustments and the like, trusting in their new adjustment algorithm.
Until one can peel the data back to unadjusted data, you don’t really know what’s being compared in the Talking Points.
I’ve looked at a number of studies purporting to show that UHI is not a factor and so far every one has the same strategy: they compare two data sets and say that there isn’t any difference, ergo, UHI isn’t material (notwithstanding the obvious fact that cities are warmer). When the data is parsed, one invariably finds that there is no valid stratification between the two sets, that they are each jumbled and therefore no trend difference. We’ve seen that with Jones et al, Peterson et al, Parker et al. Maybe the Talking Points is the first one not to do so; we’ll see (or try to see).
Steve Mc – check your email
With due respect, BigCityLib, whether or not we agree with Mr. Watts conclusions, and I certainly do not the report is called a ‘response’ yet does not cite what it is ‘responding’ to.
As a concerned person in these difficult matters I find myself increasingly dissappointed with the behaviour of scientists and organisations on my side of the debate getting far too carried away with this sort of slight on the efforts of people like Mr. Watts and Mr. McIntyre who obviously put much work into their analysis.
It smacks of something to hide and its beginning to unnerve me.
Re: Paul Clements (#41),
If continue to look, you will eventually discover that this sort of thing is widespread, because there is a specific, documented, PR strategy behind it. Not science. PR. I don’t have the links handy at the moment but they’re available on this site. Most telling is a report commissioned in the UK.
Essentially, the entire situation is to be represented as accomplished conclusion rather than a matter of vigorous debate. To do that, skeptics are to be ignored in every way possible, and if not possible, to be derided. Again, that’s PR, not science.
Re: MrPete (#42),
Stopped by to say hi… a quick google search produced the link I was referring to:
Warm Words: How are we telling the climate story and can we tell it better?
Enjoy! (This is a 2006 document. It is nothing new at CA… but new readers might find it…enlightening 🙂 )
A money quote from page 8:
Re: Paul Clements (#41),
Paul my friend, you are a bit behind the curve I’m afraid. Mr Pete, he of great fame at this site, has given the succinct explanation.
This technique is being used in many more areas than just climate science. I’ll just leave it at that.
Keep plugging away Anthony Watts and contributors. You are manifestly making a difference. The talking points memo clearly indicates attention is being paid to the deficiencies you have highlighted.
Indeed, in the first talking point of the June 9 memo it is implied that they found the siting difficulties so egregious that they are developing a new data network. I note the web site says “The U.S. Climate Reference Network (USCRN) is a network of climate stations now being developed ” (my italics) .
Much of the memo is a struggle to reassure themselves with admittedly inadequate data and minimal, incomplete studies that things aren’t as bad as they obviously appear. But the real point is that AW has now got the agency paying attention to the mess the data collection has evolved to. Embarassingly, the only data they have with which to analyze this mess is, they admit, the data you collected. As you point out at WUWT, it would be nice if they took some minimal steps to get your data correct. I look forward to your further response to their graphic indicating there’s really no problem.
Funny though, that to answer “Is there any question that surface temperatures in the United States have been rising rapidly during the last 50 years?” The NOAA doesn’t ask us to rely on any surface temperature data, instead they resort to completely different, uncited proxies in which “…as many different individuals participated in the … evaluations, with varying levels of expertise, the degree of standardization and reproducibility of… [these proxies] is unknown.”
Ultimately, all we want here is reliable data that we can draw meaningful conclusions from and/or a clear understanding of the data defects. AW has clearly pushed the NOAA in that direction. Keep up the good work!
Any true lover of science celebrates significant amounts of new data acquired and made freely and publicly available like Anthony’s project is providing. Absolutely no one on either side of the issue should miss that very important point in their effort to claim whatever they care to claim after analyzing that data. Peterson was able to make his analysis and graph because Anthony and his volunteers collected the data to make it possible.
If Anthony’s project proves a significant bias in the US surface temperature data, that’s useful *to science and informed debate*.
If Anthony’s project proves that despite flaws in collection there is no significant bias in the US surface temperature data, that’s useful *to science and informed debate* too.
At least if you love science and informed debate more than axe grinding.
Anthony’s project has been right out in public from the very start and without any chance at all to hide results by ash-canning the project with no one finding out what the results were. I wonder how many projects by other “scientists” (kaff!) suffer a quiet unpublicized death midway if they begin to see it might not go where they were hoping? The obvious conclusion is Anthony is more interested in finding out the truth, whatever that happens to be, than in backing into some pre-determined conclusion.
Since ghost writer Peterson is using Anthony’s classifications to generate his graph, and lends formality to his note by having a list of references (including one to himself), he should also have listed Surface Stations in the references, with Anthony as author. The Chicago Manual of Style indicates how to properly reference a website that is used in a formal paper. Just because it’s a non-peer reviewed website doesn’t mean it doesn’t have to be credited in the references.
Although Surface Stations does not give Anthony as its author on its home page, the Contacts button clearly indicates the site is operated by one Anthony Watts.
Although Peterson’s title echoes that of Anthony’s report, with one word missing, I am told that titles are not copyrightable, so that this is not plagiarism in itself. It is tacky, however, that Peterson would not cite Anthony’s “Is the US Surface Temperature Record Reliable?” in addition to surfacestations.org, since this is obviously what he is rebutting.
Hu, I disagree that copyright is relevant here. Consider te literal definition of plagiarism in the misconduct regulations – it’s broader than the student copying situation. It says:
The Talking Points memo appropriated ideas, processes results or words without giving appropriate credit.
Interesting post from Jennifer Marohasy — http://jennifermarohasy.com/blog/2009/06/how-the-us-temperature-record-is-adjusted/ — on how the US temp record is “adjusted”.
I’m sorry stan. Everything is so pleasant and reassuring when one reads the Talking Points Memo. You and Anthony Watts are just confusing me:)Everything is so simple if I just accept the graphs in the Memo at face value and don’t let your referenced post confuse me.
If I were a responsible scientist working for the NOAA, how would I feel after receiving and thoughtfully reviewing the Talking Points Memo? What would I do with it? I’ll bet I would know about surfacestations.org and who Anthony Watts was without any citations in the memo at all. Your reference is an interesting post. It even mentions AW’s name.
Steve’s criticism of AW’s non-attribution of Steve’s original attribution of Knowles, et al, was a little harsh, I think – I personally couldn’t take more than about 15 seconds of it.
Today (7/1) Anthony’s study is clearly referenced on the Talking Points memo, still dated 6/9:
It wasn’t there a couple of days ago, was it?
(I’ve also noted this on WUWT)
.
Re: Hu McCulloch (#52)
The pdf I downloaded on June 24th does not show it.
.
Perhaps 1984 is coming early this year.
2 Trackbacks
[…] Pielke likely wouldn’t have commented on it, McIntyre wouldn’t have written about it, twice, and thus from all the pickups from those articles, Mr. Sinclair probably wouldn’t have ever […]
[…] the NOAA “talking points” memo on relaibility of the U.S. surface temperature record might constitute plagiarism, even though the whole memo was an evaluation of the data found at SurfaceStaions.org, which was […]