Watts reasons with Goddard

Over at Reason magazine, Anthony Watts is very critical of "Stephen Goddard" over claims he made that US temperature records have been fiddled.

Some segments of the Internet are abuzz with the claim by climate change skeptic Steven Goddard (Tony Heller) over at his Real Science blog that NASA/NOAA have been jiggering the numbers so that they can claim that warmest years in the continental United States occurred recently, not back in the 1930s. Folks, please watch out for confirmation bias.

Via email, I asked Anthony Watts, proprietor of WattsUpWithThat, what he thinks of Goddard's claims. He responded...

Read the whole thing.

Comments

[Richard Drake]

sHx: Disagree. Helpful discussion. Less chill is sometimes required.

NicFromNYC: Some interesting points, notably reputation cost to sceptics for endorsing the wrong allies (including for you as real person in the real world, despite using a pseudonym online) and the dangers of Holocaust analogies on either side of the debate. I'd forgotten Goddard on the second as I'd written about it recently. Not good.

[Will Nitschke]

I wouldn't disagree with NikFromNYC characterisation, although it comes across as rather hash,

Goddard's blog is a counter-propaganda site in the sense that SkS and RC are propaganda sites. Is that a good thing for skepticism? I'm inclined not to think so.

[Jimmy Haigh]

I'm also with Steve Goddard on this one. I don't agree with everything he says but raw data is raw data and it is all we have. Once you start fiddling with it, it gets fiddled the way the fiddler wants to fiddle.

[SandyS]

Steven Goddard has a post which I think may be in part an answer to Anthony Watts, and Zeke Hausfather who seems most upset with what Steve Goddard does. Precision Vs. Accuracy, are there any fatal flaws in his (Goddard's) logic?

In the comments there is an interchange between Steve Goddard and Zeke Hausfather, some of the answers from Zeke H strike me as interesting:

Infilling is actually pretty useless, as it just mimics the effect of spatial interpolation. Its only really there so you can average absolute temperature (if you really want to) without running into climatological biases due to missing stations. I’d personally prefer that NCDC eliminate it.

Its a combination of biases in your “measured” data due to changing station composition and the effect of various adjustments (TOBs and Homogenization). No one is questioning the fact that the data have adjustments; its just that infilling is not one that makes a difference.

No idea why it was flat in USHCN v1…

You have to remember that the zero-point on the y-axis is somewhat arbitrary. Because NCDC’s algorithm assumes that current temps are accurate and adjusts everything relative to current temps

If I had to guess I’d say that USHCN v1 didn’t actually update their TOBs metadata after setting up the network in the early 1990s, but I really have no idea. There are documented TOBs changes in stations up to present, however.

As a later comment by a third party says

I think Zeke is about to become a denier hahahahaha

To me it seems he has good cause to highlight what is happening and I'm still happy with my assesment that They fabricated a warming trend by adjusting the data as part of their iniquitous activities.

[Stephen Richards]

Anthony appears to be treading warily around NOAA and NASA just in case he might need them. His case for Goddard being WRONG is flimsy at best.

Shame on Anthony for this piece of criticism of a fzellow traveller.

"He's only having fun!"

Jun 24, 2014 at 11:51 PM | Unregistered CommenterAthelstan

As his site says.

[sHx]

Richard Drake:

"Disagree. Helpful discussion. Less chill is sometimes required."

Agreed. As Christopher Hitchens used to say often, "light comes from heat".

[Richard Drake]

Stephen Richards:

Shame on Anthony for this piece of criticism of a fzellow traveller.

You have to be joking, right?

As sHz says Hitchens was right. We have to take lumps out of each other - there's no avoiding it - it's the way we do it that matters. I tend to agree with Will Nitschke that NicFromNYC can be harsh. But the argument that's broken into the open between Watts and Goddard is for me wholly good.

[richard verney]

We will see the answer in early 2015, when the divergence with statellite data can be reviewed.

It is the satellites that are keeping the land based temperature data sets in check, and unfortunately due to the shortness of the data length, they are unable to keep in check the adjustments being made to pre 1979 temperatures.

There is a lot of evidence that past temperatures are being revised downwards, and the appropriateness of such downward adjustments is moot. The 1930s temperatures have been adjusted downwards, many times, whether those adjustments are sound is an issue, but it is one reason why temperatures today, may exceed those measured in the 1930s.

[Jabba the Cat]

The blog post at Lucia's by ZekeH titled, How Not to Calculate Temperature, deals with the flaws in Goddard's methodology and is worth reading...

http://rankexploits.com/musings/2014/how-not-to-calculate-temperature/

[SandyS]

it is one reason why temperatures today, may exceed those measured in the 1930s.

There has been no dust bowl so far in this warmer period; circumstatial evidence that it might not be as warm as claimed.

[Richard Drake]

SandyS: Steve Mc had some fascinating things to say about the 1930s dust bowl and similar events in his review of the IPCC work on climate impacts in AR4 WG2 and SREX at the GWPF in London on 16th August 2012. Was a recording ever released of that talk? My notes are cryptic indeed, though this is a favourite:

Mongolian Dzud disasters new to Steve. Climate extreme != disaster

Signifying 'not equal to' in programmer lingo. Among other things touched on, America was much poorer in the 30s and understanding and technologies of irrigation are much improved. Steve himself has never been too bothered about surface temperature arguments, while respecting Anthony's work on Surface Stations, and I'm exactly the same. The key issues in the scientific foundations of the climate debate surely lie elsewhere. All the same, cheating may have gone on. The debate is good.

[SandyS]

Jabba the Cat
That post was created on 5th Jun2 2014,
This is the follow up
http://rankexploits.com/musings/2014/how-not-to-calculate-temperatures-part-2/

[SandyS]

Richard Drake
I'm familiar with programmer lingo.

Unfortunately you and Steve Mc are in a very small minority with regard to surface temperature arguments. Personally I think there is an awful lot of cheating still going on. Steve Goddard does an excellent job highlighting this.

Climate extreme != disaster

Is not always true, and probaly requires at a minimum handling by a CASE statement.

[Richard Drake]

Unfortunately you and Steve Mc are in a very small minority with regard to surface temperature arguments.

I'm surprised to hear you say that. Amount of noise on the net != a scientific poll of sceptics!

I reiterate that there may have been cheating. But even with it the increase in the somewhat surreal stat we call globally averaged temperature anomaly, since 1850, 1900 or 1950, is really small. One only gets to posited disaster in the future through positive feedbacks in climate models - and no doubt by ignoring the economic and technological factors increasingly in mankind's favour highlighted by Steve two years ago - but not, it seems, in the real world, which is stubbornly taking a break from much warming at all, while atmospheric CO2 continues to increase. That's why I can't get too excited about cheating in temperature adjustments pre 1980, even if it exists, as satellites finally begin to inject more rigour into the record. I can't speak for Steve, needless to say.

[thinkingscientist]

I think I agree with Anthony Watts view that you should never assign motive when incompetence is sufficient as an explanation. Goddard is over the top in claiming motive without any supporting evidence.

However, there is a fundamental issue with the temperature data sets. I have repeatedly asked a series of questions, looking for Mosher or some such to provide an answer. The most fundamental question is why do the adjustments show a systematic temperature rise over time, for which the magnitude is 50% of the final warming trend over the 20th Century? And more importantly, what physical process or affect on temperature measurements requiring adjustment would have a positive sign over a long period?

Still waiting to hear from someone who can explain the physical basis for those adjustments.

[harkin]

I'll side with Anthony on this one. Any scientist who goes overboard because he's "having fun" and whose over-the-top or sloppy work (especially when errors are not admitted) can be used to smear all AGW skeptics is a minus, not a plus.

[Richard Drake]

I'm with thinkingscientist in case I'm drummed out of sceptic camp for heresy. It was mention of the 1930s dust bowl as evidence that got me onto wider issues.

[Bruce]

Obviously Carrick, you are on the approved list.

[Bruce]

Suyts has a good post to counter the Zekes.

" The first graph I posted is from The Blackboard. Essentially, it affirms what Steve is saying. The USHCN used to have 1214 stations which reported real data. Now, they have about 800 or so stations. But, we keep reporting temps from the dropped stations via gridded infilling. "

http://suyts.wordpress.com/2014/06/24/sigh-skeptics-doing-what-skeptics-do-best-attack-skeptics/

Try and get you head around that. The USHCN keeps the station count at 1214 but less than 900 of those are actually reporting data. The rest is "Estimated" or "Infilled" = Fabricated.

And the Fabricated data is warmer than the "real" data.

(Remember, I'm not talking about TOBS or other adjustments here) The infilling step is to the data after all the other adjustments have occurred.

[SandyS]

Richard Drake
I was thinking along the lines of the whole shebang rather than just those of a sceptical view. I haven't done a 97% analysis but I do suspect the majority of those who make a living from the environment would not agree with you or Steve Mc.

The problem is, as I see it, that by cooling the past while they can it doesn't really matter what the satellites, argo bouys, Arctic Sea Ice, Jason or whatever say the damage will have been done. Whether Steven Goddard is using a incorrect method or not he has highlighted something which nobody is able to deny, there are missing stations and the past has been cooled. The relevance of the measurement is of secondary importance to the politicians and public.

[SandyS]

Bruce
I've just finished reading that and was about to post the link, you beat me to it.

[Richard Drake]

SandyS:

I was thinking along the lines of the whole shebang rather than just those of a sceptical view.

Ah, that's a very different matter. The Great Misdirection: We say climate policies are immoral and cannot possibly be justified on the back of some model-based science that is inconclusive at best, they say that we say there's been no global warming. (No warming since when? They seldom say what we say to that level of detail. Given it's misdirection, the vaguer the better.)

More on The Great Misdirection in a discussion thread not far from here in the not too distant future. I feel I have something fresh to offer on that. But with it as background, here's the thing we shouldn't say: Steve Goddard should not criticise NASA/NOAA. What we can perhaps say is that he should be very careful as he does, lest he play into the hands of those pushing The Great Misdirection. Being unteachable and extreme in other ways, such as identifying sceptics today too closely with Jews in 1940s, would not in my view be part of being very careful. But the man's science could still stand up. I'm not the man to look into that.

[Bruce]

Remember when July 2012 was declared as hottest month ever in USA.

NOAA has changed their minds.

http://sunshinehours.wordpress.com/2014/06/25/noaa-usa-july-1936-is-back-on-top/

[lapogus]

The surface station datasets cannot be trusted, and it is not just raw verses value-added that is the problem, surface station selection and deletion are key, as McKittrick noted a few years ago:

GAT v No of stations

Is the step up around 1990 just another climate coincidence?

[sHx]

Richard Drake: "As sHz says Hitchens was right."

Yes, I must have said something like that in the past.

You know Richard for someone who cares so much to keep his name true and polished, you can be a bit careless about other people's choices for a name. We are not all vain those of us who choose monikers.

But I'll make an exception for you.

It is sHx, Richard. The x at the end of my seal is pronounced like the 'ch' sound in German. As in "Achtung! Achtung!"

[sHx]

The edginess has caught up. Sorry for my outburst ... on reflection.

[Richard Drake]

It was entirely accidental, sHx, but if there was Freudian aspect it surely has to do with Heinrich Hertz - quite a compliment as one experimentalist to another.

[sHx]

Ok Richard it wasn't on purpose but a compliment nevertheless. Sorry for the blitzkrieg.

[rw]

Since I don't have time to dig into all the in's and out's of temperature analysis, I tend to latch onto specific cases. Two come to mind: (i) Darwin, Australia, as discussed by Willis Eschenbach on WUWT a while back, where a decline in temperature from the late 19th century was turned into a 6 degree per century increase, (ii) sites in Reykjavik and other parts of Iceland, where there is a well-documented decline in temperature from about 1940 to 1980 (I know because I've looked at some of the climatological literature from that period), which disappeared in the GHCN record as discussed by Paul Homewood a couple of years ago. Both cases seem to go well beyond mere incompetence. So I think similar claims if they have some backing have to be taken seriously, despite the danger of falling down the conspiracy rabbit hole.

[Richard Drake]

The Smoking Gun At Darwin Zero in December 2009. Quite so. Published by Anthony Watts, of course.

[_Jim]

Nik - wrong "Jim". I am not JR nor Jim Robinson as asserted. Your assertions of bigotry on FreeRepublic are likewise spurious and in error. None of this likely will change your MO, but the facts of the matter are now a matter of public record.

[gregole]

Real Science is a great blog - I check it out at least once a day if just for infotainment; but for gosh sakes, it is a blog - not the final word. Don't like Steven's numbers? Post a rebuttal. He only bans egregious trolls. Don't like his politics? Post a counter point - he might call you a moron. It is after all, his blog. But it is comparatively quite open to all; so he attracts some shall we say, interesting posts. Better yet for infotainment. Just saying. Some of his most virulent critics are quite entertaining as a matter of fact.

I find the recent personal attacks on Steven by other climate skeptics disturbing. Are we not all taking ourselves a bit too seriously?

[TinyCO2]

gregole "I find the recent personal attacks on Steven by other climate skeptics disturbing. Are we not all taking ourselves a bit too seriously?"

Being sceptics we can be sceptical of anything we want, including each other. Part of the rotten core of warmism is their unwillingness to call out bad work/behaviour on their own side when they see it. He can of course post what he wants and have fun but then he's getting the fallout for it.

The reason I don't steer newbies in his direction is not because of his climate opinions (right or wrong) but because I don't want them to write off scepticism as a political standpoint. Goddard's other interests and his delivery are too polarising. We've judged warmists on more than the science and we can expect others to do the same for us.

[shub]

Vincent Courtillot calculates US continental temperatures using a different method and much fewer stations:

The graph looks like this: http://images.slideplayer.us/1/630582/slides/slide_6.jpg

Again, note the y-axis. These are average temperatures. Note also the shape of the curve - the multidecadal swings are much more pronounced. This is what people and ecosystems actually experience.

The US forms a small portion of globe. There is no reason to expect US continental temperatures to confirm to the shape of the global anomaly graph. The US could cool the entire 20th century and it still would not have mattered. The original global warming narrative, however, carried the construction that 'global' warming was seen on all continents and large countries. Guess what, ... the narrative builders were/are in charge of the temperature data. It is self-evident the keepers of the records would not have found it difficult to *rationalize* the adjustments they carried out.

[johanna]

Indeed, shub. We now have a new metric in Australia, courtesy of the warmies in the BoM. It is the "national average" temperature, calculated we know not how, from an assortment of weather stations across an area about the same size as the continental US. No prizes for guessing what comes out of the sausage machine. Yep, we're frying.

It's a meaningless metric per se, but when you throw in the adjustments, especially the retrospective ones going back years, and the secrecy about how it is compiled, it's even worse than that.

[SandyS]

An interesting comment by Willis Eschenbach, who is a regular contributor at WUWT

http://stevengoddard.wordpress.com/2014/06/27/anthony-watts-says-he-will-have-a-look/#comments

Willis Eschenbach says:
June 27, 2014 at 6:43 am
stevengoddard says:
June 27, 2014 at 3:01 am

The distinction is nearly meaningless, and I in fact clarified it on my blog within half an hour of the original post on June 1.

I see a clear difference between e.g. a station which is missing some data and a station which is closed and shuttered, although certainly YMMV.

But they do in fact post data for many stations which have been shuttered for years. The final monthly data count is constant for over 50 years.

I haven’t checked the data, I’ve never put together the software to do that and it would take a while. But given that that’s the case, and I see no reason to doubt you, I’d say you definitely are on the right track and should persevere with your investigations and revelations.

As a suggestion, it would be helpful and would make your presentation both more powerful and more understandable if you would display some examples, e.g. “Here is station X. Here is the documentation that it was closed and shuttered in 2006 … and here are the alleged records from the station from 2007 to 2014.”

My best to you and thanks for your clear answers, keep up the good work,

w.

[lapogus]

SandyS - I think it is also worth repeating the link (which you posted on Unthreaded) to Paul Homewood's new post on the Luling (Texas) station here:

Massive Temperature Adjustments At Luling, Texas - http://notalotofpeopleknowthat.wordpress.com/2014/06/26/massive-temperature-adjustments-at-luling-texas/

[Anything is possible]

......And Anthony Watts issues a mea culpa :

http://wattsupwiththat.com/2014/06/28/the-scientific-method-is-at-work-on-the-ushcn-temperature-data-set/#more-112127

[Richard Drake]

Thanks Aip. Worth reading AW in full but this I'll repeat:

All of that added up to a big heap of confirmation bias, I was so used to Goddard being wrong, I expected it again, but this time Steve Goddard was right and my confirmation bias prevented me from seeing that there was in fact a real issue in the data and that NCDC has dead stations that are reporting data that isn’t real: mea culpa.

But, that’s the same problem many climate scientists have, they are used to some skeptics being wrong on some issues, so they put up a wall. That is why the careful and exacting analyses we see from Steve McIntyre should be a model for us all. We have to “do better” to make sure that claims we make are credible, documented, phrased in non-inflammatory language, understandable, and most importantly, right.

Otherwise, walls go up, confirmation bias sets in.

The words of a great man and a wiser one today.

[Bishop Hill]

The thread was deteriorating into off topic and venting. I've removed the offending comments.

[Richard Drake]

There may have been some I never saw. Happy day :)

[robotech master]

The thread was deteriorating into off topic and venting. I've removed the offending comments.
Jun 29, 2014 at 9:53 AM | Registered CommenterBishop Hill

Missed some

[Vangel]

So what happens now? While Goddard has made some errors in the past this particular claim seems to be a very difficult problem for the alarmists who have been pushing the global temperature trends that have been fabricated from data that was not real? How do we look at any of the material published by the IPCC and claim that it has merit if it is based on data that was manufactured rather than measured?

[not banned yet]

Good piece here from Brandon Schollenburger:

http://hiizuru.wordpress.com/2014/07/02/laying-the-points-out/

Apologies if it has already been noted.

[johanna]

Anthony Watts has updated on this. Apparently, "our algorithm is working as designed". As an avid watcher of US crime shows, I would characterise that answer as "non-responsive."

It's been interesting to watch, because Anthony Watts obviously detests Steve G (Tony Heller).