Willis Eschenbach caught lying about temperature trends

As shrill and silly as Eschenbach is, you're jumping the boat yourself a bit, Tim.

Eschenbach's argument is basically, "I don't know what adjustments were made or why, so instead of asking somebody or applying the published homogenisation procedures, I'll assume they're fradulent."

While it's interesting to note that both the GHCN and the Australian BOM have made adjustments, it'd be better to get clarification on exactly what those adjustments were. Why not do the bit of work that Eschenbach was too lazy or incompetent to do, and ask the Australian BOM about what specific site changes happened at Darwin? For one, I've heard rumors that the station was bombed in 1941. Something to follow up on.

Ha - following your old link, it looks like Eschenbach still has the same compulsion to make any anomaly chart start at zero. I don't know why. I think he also fails to mention what baseline period he uses to calculate anomaly, though the definition gets obscured when he then nudges it so it starts at zero.

He's lying when he says that the nearest station with data going back to 1941 is 500km away, too; there's another about half that distance. He also strongly implies that the Darwin station couldn't be homogenised due to the lack of nearby stations. He conveniently neglects to mention that if GHCN can't homogenise stations, they don't:

Also, not all stations could be adjusted.
Remote stations for which we could not produce an
adequate reference series (the correlation between
first-difference station time series and its reference
time series must be 0.80 or greater) were not adjusted.

That's from the same GHCN document he quoted, incidentally.

Tim: Since this graph appears in a recent Volokh post that doesn't allow comments, I was wondering if you'd respond to that post here.

thx.

Tim,

I had to laugh when I read the link to your earlier Eschenbach post:

[quote]The [b]trick[/b] Eschenbach used was to use a single year for the baseline instead of the thirty year average that is normally used. Yes, it's another version of the disingenous baseline game that produced all those bogus "global warming ended in 1998" claims.[/quote]

I'm not a sceptic by any means; rather the point of this comment is simply to illustrate that I can see why those that leap on the 'climategate' emails and distort what 'trick' means simply have to go to that post for example to see where other scientists have used it in a context which implies dishonesty.

Just be careful is all I'm saying.

I demand to see his raw data, his MATLAB code, everything! Smile. Maybe Steve M will audit his work for us all.

Seriously though, Tamino has also weighed on this matter.

We have David Kane to thank for this elucidation, since he suggested it below.

Don't forget his email!

And I don't think Willis is intentionally fouling this stuff up. There may be some confirmation bias going on because he seems to be wrong the same way all the time :)

Tim,

Good stuff! Taking on the best arguments from the skeptic is a much more productive use of your time than taking on Palin.

Can we get a link to whatever Tamino has said?

I agree with Boris - I don't think Eschenbach is 'lying' at all; the title here is far too shrill. He's just somewhat less than competent, and in this case, amazingly fast to lob accusations of fraud. When shown how he's messed up, he's then rather stubborn until he doesn't have much choice but to admit he messed up.

IOW, he's not lying so much as bullshitting.

sorry David, but this "raw" data meme that is going on at the moment, is exactly the same as what Palin is writing.

Sorry, I messed up carrot eater, nothing on Tamino (yet), ThingsBreak has a short piece.

http://thingsbreak.wordpress.com/2009/12/09/the-smoking-gun-darwin-stat…

Personally I think that the denialists are trying to bait NOAA and BOM into making statement clarifying their position on this. The reason is clear, it makes them look defensive when they have to "defend" something. This is the same tactic used to perverse extents by McIntyre and his cohorts. Watts also employed the same tactic with the US with their stations. This is all about appearances and spin, and nothing to do with science or getting at the truth.

Tamino does have this interesting bit of research on the alleged cooling and questioning of the validity of the SAT data:

http://tamino.wordpress.com/2009/12/07/riddle-me-this/

Just how much more time do we ned to spend refuting these ludicrous conspiracy theories and myths? Are the radiosondes and satellites colluding now too? The oceans as well (i.e., OHC)?

These guys just keep spinning and obfuscating, it is quite pathetic.

Excellent work, Tim. What explains to your opinion the trend after the jump in 1940?

MapleLeaf: I think a response from the Australian BoM would be perfectly appropriate, just as the NIWA was on a similar matter last weak with New Zealand surface records.

In fact, I rather enjoyed the NCDC's response to Watts, showing what little difference it made if one only used the weather stations his photographers rated as being good. I think Watt's reply to that (after some irrelevant blather) was to mumble something about homogenisation.

But in any case, I'd suggest to the bloggers to either
a) do the legwork Willis was too lazy to do, and find out what adjustments were made and why
b) hold off on commenting until somebody else does.

I totally agree with Maple Leaf: we're dealing with people who steal internal emails, with no legal authority to do so, and then lie about what the emails say. If these so-called skeptics actually had a solid scientific basis for their denialism, they'd be doing the actual science, and not degrading their credibility by associating with thieves and liars. These people are are nothing but infantile, ignorant con-artists; and we do ourselves a grave disservice by letting them exert ANY control over ANY adult debate on ANY subject. Trying to defend the actual science in response to such people is no more productive than arguing geography with a flat-Earther, arguing human rights with a Nazi, or arguing citizenship law with a birther.

Carrot and Raging bee, good points. The NOAA and NIWA responses were good, but then Pielke Snr. dismissed the NOAA response as 'cavalier'. There is just no making these guys happy, period. Either they then just keep repeating their lies, or they move the goal posts. It is like gorilla warfare.

I suggest a class action lawsuit by scientists in another thread. Why, why do we continue to sit on our hands? I used to think that one could reason with a denialist and show them the way using facts and data. Nope. They only start paying attention when you mention "lawsuit". There must be endless material out there for lawsuits. For example, McIntyre stating at http://www.climateaudit.org/?p=6912 that:

"in contrast, James Hansen and his disciples have a more jihadist approach"

or at http://www.climateaudit.org/?p=7257 McIntyre states:

"However, as CA readers know, the resulting Yamal chronology with its enormous HS blade was like crack cocaine for paleoclimatologists "

And there are many more from him and others.

I suggest a class action lawsuit by scientists in another thread.

Wouldn't it be like the reverse of a class action lawsuit? (one to many, not many to one.)

There are plenty of bloggers who have outright accused the CRU researchers of fraud, based on little more than wording such as "hide the decline" and comments in code that was never used in any publication.

It would be interesting if such bloggers got sued, but I think the blogging community as a whole would back the bloggers, almost regardless of the merits of the lawsuit. They could even sue in the UK, where they don't even need to show reckless disregard for the truth. The standard is much lower.

And now the denialists are resorting to death threats:

http://www.abc.net.au/news/stories/2009/12/09/2766508.htm

None of the nearest "high quality" sites for Darwin airport show a significant warming since the 60s (HALLS CREEK AIRPORT (733 km away), TENNANT CREEK AIRPORT (877 km away)).

Expansion that has taken place at Darwin airport (check it out on google maps 12.42°S 130.89°E and it's discontinuity with its "local" peers, it would seem to be a prime candidate for UHI + jet backwash.

Willis' main point is that the whole trend is due to "corrections" to the raw data - that is a fact. The problem with human correction of data is that it's prone to confirmation bias, and very subjective. You stop looking for corrections when the data "looks" right.

As it stands, not only is the data out of sync with its peers, the whole trend is due to corrections. That is not healthy specially for a "high quality" data point. I would have expected high quality to mean very few corrections necessary.

The problem with human correction of data is that it's prone to confirmation bias, and very subjective.

You appear to be overlooking the entire field of statistics.

Additionally, this:

As it stands, not only is the data out of sync with its peers, the whole trend is due to corrections.

is just nonsense. The raw data for station 1 shows a clear increasing trend from 1940 to 1990, which is very closely matched in the raw data from station 2. Which is precisely why the station 0 data has been adjusted for that period, I'd imagine.

None, the data in Australia are related on a scale form 1-5 (see link Tim provided), Darwin is rated a "3" or fair (1 being the highest rating), so they know there are issues with that site. It is not a "high quality" data point as you allege.

Data homogenization is a very real and important issue, there are very few sites globally where homogenization issues are a problem. Scientists do the best they can with what they have. And they do a hell of a better job than Eschenbach.

Moreover, who has a better reputation and more credibility, NOAA or someone who has been shown to fudge data in the past to fit their agenda and support their ideology? I'll go with NOAA thanks, that is until someone can demonstrate convincingly using the proper stats and real science that their homogenization technique has serious flaws. I challenge Eschenbach to try and publish this in J. of Climate.

Ok MapleLeaf, I assumed Tims "Oh look, here's the Australian Bureau of Meteorology's HIGH QUALITY climate data for Darwin aiport", and the fact that it said "Australian HIGH-QUALITY climate site data" at the top of the web page implied it was indeed high quality data. Mea culpa, but I think you can sympathise with both Tim and myself being confused over the matter.

The point that the trend is due to human corrections rather than the raw data is important. Specially when the nearest stations (which would clearly not have UHI problems because they are much smaller) in the "high quality" (i'm now using that as a noun, rather than an adjective) network show only minimal trend.

I would be interested in seeing explicitly what individual corrections had been made to Darwin Airport's temperatures, and why (in an itemised list with rationale and effect, rather than the merely the absolute adjustment which was shown by Willis). I do not think it is wrong to ask for that, or even to harbour the suspicion that there might have been mistakes made on this station, given the difference it exhibits from it's peers. Any idea who I could contact to get hold of that information ?

The point that the trend is due to human corrections rather than the raw data is important.

...and wrong. This document describes the method used for homogenisation in detail. And, as I said, stations 1 and 2 at Darwin airport show a warming trend from 1940. That's the raw data, not the adjusted version.

Journal of Smoking Guns & Final Nails, more like.

Earlier I meant to say "Data homogenization is a very real and important issue, there are very few sites globally where homogenization issues are NOT a problem."

None, let me get this right. You are now suspicious of all the data and the homogenization technique because of a seriously flawed analysis for ONE station posted on a blog by a pseudo scientist? You also seem very focussed on the UHI, there are a myriad of other issues that come into play with homogenization issues. Hopefully BOM has been diligent of keeping record of those changes.

You also neglect the marked increase in global mid-trop temps. determined from the global radiosonde data and satellite data. Yes, it is all one MASSIVE "conspiracy", thank God enlightened pseudo scientists like Watts and Eschenbach have figured it all out for us. And yes, I am been incredibly sarcastic.

If you are in Oz (i.e., tax payer) you should be able to get the raw data easy enough.

Oh I doubt his suspicions extend to stations whose data have been adjusted down. Adjustments aren't the problem; up is the problem.

None writes:

"Specially when the nearest stations (which would clearly not have UHI problems because they are much smaller) in the "high quality" (i'm now using that as a noun, rather than an adjective) network show only minimal trend."

Really?

http://data.giss.nasa.gov/cgi-bin/gistemp/gistemp_station.py?id=5019423…

http://data.giss.nasa.gov/cgi-bin/gistemp/gistemp_station.py?id=5019423…

This is the unhomogenized data for this location (not a lot different from the homogenized). Just eyeballing the two records, I'd estimate a linear trend of around +1.5 C since 1926, which covers a sufficient part of the homogenization adjustment period for Darwin.

This isn't the first bit of disinformation from Willis Eschenbach.

http://rabett.blogspot.com/2006/10/rtfwr-or.html

Global warming denial has quite the media market. No coherent argument needed. Just scream "conspiracy" and the whole choir chants in unison.

MartinM
The raw data is here:
http://data.giss.nasa.gov/cgi-bin/gistemp/gistemp_station.py?id=5019412…

The document describing the method used for homogenisation is exactly what I am talking about - it's a human added correction, a program written to adjust the data. As a programmer myself, I am fully aware how easy it is for subtle problems to affect a small minority of data in a big way. Without knowing an itemised list of the changes this process makes to the data you are only left with faith in the guy who programmed it. There is no process mentioned for example, on how it adjusts for UHI, other than having a "preference" for rural data, as one of a set of criterias for initial selection of data.

None, Eschenbach's issue was not with the data post-1960. It was with the data pre-1940. That's the adjustment he focused on in his WUWT piece. But the answer to you is the same as the answer to him. Quit filling the gaps in your knowledge with accusations of bias, and find the evidence for yourself.

Incidentally, here are the stations you mentioned

[Halls Creek Trend](http://www.bom.gov.au/cgi-bin/climate/hqsites/site_data.cgi?variable=me…)

[Tennants Creek Trend](http://www.bom.gov.au/cgi-bin/climate/hqsites/site_data.cgi?variable=me…)

[Darwin Airport Trend](http://www.bom.gov.au/cgi-bin/climate/hqsites/site_data.cgi?variable=me…)

Where is your evidence that there is no significant warming trend since the 60s on Tennants Creek or Halls Creek? I admit to not having done any calculations, but unless you have and can show me it seems a very odd assertion based on eyeballing alone.

On the post itself, I think it is always very difficult to prove deliberate dishonesty, and I'm not sure I would have used the phrase 'caught lying'. Having said that, if you chuck accusations of dishonesty around yourself with far less evidence than what Tim has, you sort of don't have much of a recourse to complain.

Incidentally, is there a good reason why Eschenbach suddenly switches from graphs showing the mean temp to the temp anomoly about halfway through his post, other than the anomoly separates all the data points and makes it look visibly like the whole set has been fudged upwards? It's a genuine question, and my ignorance may be showing here, but it looks to me like he's playing on the gullibility of his readers.

None, Mapleleaf:

"Several high-quality (or homogenised) observational datasets have been developed to identify, monitor and attribute changes in the Australian climate. These datasets have been produced using a variety of quality control and correction techniques."

Just so we're all clear on the meaning of [high quality site data](http://www.bom.gov.au/climate/change/hqsites/about-hq-site-data.shtml)

All I did was point out the Willis is correct that the huge trend at Darwin Airport was a result of adjustment of the data, an inconvenient truth ;-) perhaps, but a truth.

Yet you come back with "Yes, it is all one MASSIVE "conspiracy", thank God enlightened pseudo scientists like Watts and Eschenbach have figured it all out for us. And yes, I am been incredibly sarcastic."

Talk about hysteria. Slap yourself across the face and tell yourself to "get a grip man!".

MarkB, your links don't work. Following the links to "nearest alternative sites" at the "Australian high-quality climate site data", you'll see they do not have anything like a similar trend than Darwin Airport data is showing.

Off to bed!

Maple Leaf: I will say the NOAA response to Watts was presented in a somewhat 'cavalier' way (though still better than his blog science); I do wish they'd take the time to publish a short paper about it.

Meanwhile, talk of lawsuits? Come on. That's not how these things are handled.

I love it when the programmers feel the need to act as if they're an authority, just because they write code. They haven't a clue about the physics or the specific code involved, but since they know programs can be buggy, well of course [insert name here]'s code must also be buggy.

MapleLeaf - "If you are in Oz (i.e., tax payer) you should be able to get the raw data easy enough."

Actually it is really difficult:-) You have to get up off your lounge chair and ring the BOM and go through to the data area where someone will ask if you are part of the conspiracy or not as they keep two sets of data.:-)

Then you have to specify the stations you want and time that you want it then go to all the trouble of walking over to your PC and getting the email with the data attached. Also you do have to a pay a nominal fee.

I recently asked for and obtained all the wind data for WA for the last 10 years. Mind you they did not ask if I was in the conspiracy then as this was before Climategate.:-)

All this trouble to actually get the raw data is just too hard then you have to analyse it so most deniers would rather just sling off at a few scientists instead.

Thanks Bud.

None, this "hysterical warmer'" will now go and take their valium to calm down ;o) Do you not grasp the concept of sarcasm?

No None, people are actually pointing out inconvenient truths to you, and instead on conceding anything you just keep repeating the same old mantra like a broken record. Stop pontificating.

So Watts has another acolyte....good grief.

To others in the know about these things-- any idea how many stations homogenization resulted in a reduction or reversal of a warming trend evident in the raw data?

none:

All I did was point out the Willis is correct that the huge trend at Darwin Airport was a result of adjustment of the data, an inconvenient truth ;-) perhaps, but a truth.

This point has been acknowledged and explained so many times in the last day or so that your 'pointing it out' is entirely redundant. Adjustment of the raw data is neither inconvenient nor convenient. It is simply, on occasions necessary.

Again, are you going to keep throwing around allegations of bias, or are you going to find evidence?

The document describing the method used for homogenisation is exactly what I am talking about - it's a human added correction, a program written to adjust the data. As a programmer myself, I am fully aware how easy it is for subtle problems to affect a small minority of data in a big way.

"[H]uman-added"? Only if following statistically valid and scientifically justified corrections is human addition.

Presumably, as a programmer, you use logic and perhaps even mathematics in your programs. Is such logic and mathematics rendered illogical simply because you yourself are a human, and thus any logic and mathematics are merely "human additions"?

Here's a term for you to learn about today - "a priori". Think about how it applies to the statistical processes used in data analyses, and about what it implies for the veracity of the processes used.

Oo, and if you can point me to any correction that is not ultimately human-added in one form or another, I'd be most interested.

Following the links to "nearest alternative sites" at the "Australian high-quality climate site data", you'll see they do not have anything like a similar trend than Darwin Airport data is showing.

I get

• Darwin Airport: linear trend of +0.13C/decade
• Halls Creek Airport: linear trend of +0.13C/decade
• Tennant Creek Airport: linear trend of +0.06C/decade

looks like a similar trend to me.

I'm answering questions on my posting at

http://wattsupwiththat.com/2009/12/08/the-smoking-gun-at-darwin-zero

Please post any scientific questions and objections there, I will answer them as time permits.

Thanks,

w.

PS - I may be wrong, but I'm not lying. I reported what I did, and the results. Your first step should be to replicate them.

Tim,

Precisely what are you accusing Eschenbach of "lying" about? Your post does not make that very clear.

PS - I may be wrong, but I'm not lying

Willis lied when he claimed he'd proved that the adjustments were done to fraudulently show a warming trend where there is none.

Then he lied when he said he didn't lie.

Shouldn't you fuss with adjustments on the actual temperature scale, as opposed to the anomaly scale? Every time you make an adjustment within the baseline period, the baseline also changes, so all the anomalies everywhere would change. No?

Willis Eschenbach:

PS - I may be wrong, but I'm not lying. I reported what I did, and the results. Your first step should be to replicate them.

No. Replicability does not come into it, as indeed it wouldn't come into a creationist study which played with some evolutionary numbers and concluded a 'god of the gaps'. The situations are directly analogous.

No-one, as far as I've seen, has questioned your calculations per se. Just why you bothered - or rather, why you bothered to work out the difference in raw and corrected data without actually making an effort to discover the reasons for the corrections.

I'm on record on this thread as doubting whether you deliberately lied. But you are the one accusing dishonesty in others with no evidence, so you don't have much moral high ground on the matter.

"None",

Asks a [reasonable](http://scienceblogs.com/deltoid/2009/12/willis_eschenbach_caught_lying…) question:

>*I would be interested in seeing explicitly what individual corrections had been made to Darwin Airport's temperatures, and why (in an itemised list with rationale and effect, rather than the merely the absolute adjustment which was shown by Willis.*

This seems an appropriate piece of evidence to seek when questions are raised. However it is not something Willis Eschenbach needed to pronounce he'd found the "smoking gun" nor was the needed for Willis to:

>"declares the NOAA's adjustments "blatantly bogus" that created a "false warming".

Thanks to Tim and the other posters who have started gathering and presenting documentation on how and why BOM calibrate the data.

Looking at the BOM data (from Bud) my eyeball suggest that the trend for Halls creek is 0.8 C per century; Tenant Creek is 0.5 C/century; and Darwin is 1.3 C/century.

MarkB, it would be interesting to read your links if you repost them contained inside angled brackets <>.

Perhaps calibration is done with other local stations; stations with shorter histories, but which overlap the Darwin changes in a way that make them appropriate for use as Darwinâs calibration?

Shall we wait and hear a response from the people who do this work? Or shall we blow the whistle on this massive hoax?

Willis Eschenbach, I've seen no evidence for serious truth seeking on his part. He looks at half the case and sees what he wants, then flings poop.

And "useful idiots" like Lank always laps up these premature claims, never learning from the string of similar claims that have been so overblown beat-up, or more usually erroneous.

I notice el gord is still mentioning Yamal, as though it was some kind of skeptics victory. So create a bit fuss, make lost of noise, make lots of claims, and the evidence can take a back seat. The useful idiots remember that they were always right! And after beeing right on everything, the only explanation is a massive conspiracy, where BOM and CRU and cook the books to pretend the globe is warming.

And the "skeptics" revelations are always on TV and the radio, so it must be true, and did you hear even Tim Flannery is a skeptic now! (Don't worry I'm sure its only members of parliament that got fooled by Bolt's dog whistle on that last one).

The obsession with "discovered" minutia about specific temperature stations and such is completely spurious, in my view, and basically a game of "gotcha" the deniers like to play. Bias and errors in temperature measurements can't possibly explain a number of observations, such as this.

@Janet:

Looking at the BOM data (from Bud) my eyeball suggest that the trend for Halls creek is 0.8 C per century; Tenant Creek is 0.5 C/century; and Darwin is 1.3 C/century.

Check the scale for [Hall's Creek](http://www.bom.gov.au/cgi-bin/climate/hqsites/site_data.cgi?variable=me…). Its trend is identical with that of Darwin, but with greater year-on-year variability above and below the mean.

You actually don't need to eyeball. The trend is given on the graphs (bottom right).

btw

Mea culpa 1: To be fair to none, s/he did a reasonable question which I didn't spot when I was having a bit of a pop.

Mea Culpa 2: For some reason I linked the 13y average for the Halls Creek site before. No idea how that happened.

His plot of Darwin Airport raw data is accurate.

This is the GISS plot for Darwin Airport

http://data.giss.nasa.gov/cgi-bin/gistemp/gistemp_station.py?id=5019412…

Referring to his imposed chart, the question is how does the raw become the adjusted?

The 1940 event obviously requires adjusting, and you would hope to be able to reconcile that with a known real world station event.

But after that? Why is the adjusted almost linearly diverging from the raw up until about 1985? What kind of station anomaly manifests itself in a linearly changing fashion over a period of 40 years?

The BOM has some explaining to do here.

I suppose the next argument is that the NOAA and the BOM are conspiring together to falsify the temperature record.

Hasn't that always been the only argument?

But after that? Why is the adjusted almost linearly diverging from the raw up until about 1985? What kind of station anomaly manifests itself in a linearly changing fashion over a period of 40 years?

I think if you ask Eschenbach why he compared the anomaly trends rather than the annual temperature means you will find the answer to your question. Ask him about the differences between pre and post-corrected annual mean temperatures after around 1940.

Bud,

Not cherry picking, just random picking, as you have to start somewhere on this v. interesting stuff. I only see a trend on the linked Hall's chart if you include the cold anomalies in the 1920s. But I thought that I saw a citation in the Della-Marta, Martin and Braganza (2003) article that there was a problem with bad thermometers back then that lowered the recorded readings. Or is it claimed that correction has been made in the homogenized data to the 1920s etc. readings? Without those early readings, the trend seems pretty fuzzy.

Janet Akerman writes:

"Looking at the BOM data (from Bud) my eyeball suggest that the trend for Halls creek is 0.8 C per century"

Actually, it appears to be about 1.3 C (see Mark Nettle's post), matching the Darwin trend. And there's no homogenization adjustment apparently. Bud's link for Hall's Creek didn't show the linear trend.

"MarkB, it would be interesting to read your links if you repost them contained inside angled brackets <>."

Let me try that...

[Tennant Creek Series 1]

[Tennant Creek Series 2]

My eyeballing is off on this one too.

Willis Eschenbach's line of thinking is similar to that of any global warming denier or conspiracy whacko: all science is fraudulent unless proven otherwise, and then deny the proof. Indeed, without knowing anything about the issue, he declared the adjustment "blatantly bogus" - one that created a "false warming" - something not at all supported by rational analysis, but something gleefully reported by drooling conspiracy nuts. I wouldn't take the bait and run over to the WUWT and give Watts more web traffic (of which he proudly touts regularly). Eschenbach's nonsense is taken apart nicely here. He's free to respond.

With apologies to Tim and readers if I am off-topic here, but I need some quick help regarding Willis Eschenbach. A "sceptic" correspondent has e-mailed me a pdf titled "Willis Eschenbach's FOI Request", and dated 25 Nov. '09. This appears to have been created from posts at CA. It supposedly details Willis's attempts to extract climate info from the Hadley Centre. I am asking for help here because I had the feeling, though not a strong recollection, that there had been a recent thread at Deltoid dealing with Willis's FOI requests. However, a quick google search on the name and "FOI" yields only links to the usual contrarian sites. Can anybody help me locate such a thread? Thanks, in advance.

"I think if you ask Eschenbach why he compared the anomaly trends rather than the annual temperature means you will find the answer to your question."

He's used the same yardstick from start to end. That is the only period where it doesn't produce a flat line. (apart from the very last of the upward trend where it bounces about a midpoint, say for 10 years).

I'm not sure what you're getting at. The gap between the two series' increases in line with his plotted adjustment. The anomaly is a direct derivation from the temperature so I can't see why that would display a different relationship to plotting a temperature difference between the raw and adjusted series'. To me it doesn't look like it would.

Looks like a whole bunch of posts went missing...somebody exercising their lawyers, eh?

Bud, MarkB,

Thanks for the pointers, I agree, Halls Creek has the same trend as Darwin, 1.3C/century.

Bud, with so many insincere trolls around, and with only text to give us prompts, its understandable that on occation one keeps swinging and finds some other poor sod on the end of one's "pop".

;)

I did some calcs and plots re Darwin which I posted at WUWT. First was the corresponding GISS data:
raw and homogeneity-adjusted plots. And there's very little adustment at all. Then the newly posted CRU data from this MetOffice site, which plotted against raw showed this plot. Again, no big adjustment.

So I wondered if in the GHCN dataset, other NT stations had big adjustments. I did a test on a block of stations in the v2.temperature.inv listing, which are in north NT. I noted that wherever there was an adjustment, the most recent reading was unchanged. So I listed the adjustment (down) that was made to the first (oldest) reading in the sequence. Many stations, with shorter records, did not appear in the _adj file â no adjustment had been calculated. That is indicated by âNoneâ in the list â as opposed to a calculated 0.0. Darwinâs 2.9 is certainly the exception.

In this listing, the station number is followed by the name, and the adjustment.

50194117000 MANGO FARM None
50194119000 GARDEN POINT None
50194120000 DARWIN AIRPOR 2.9C
50194124000 MIDDLE POINT 0.0
50194132000 KATHERINE AER 0.0
50194137000 JABIRU AIRPOR None
50194138000 GUNBALUNYA None
50194139000 WARRUWI None
50194140000 MILINGIMBI AW None
50194142000 MANINGRIDA None
50194144000 ROPER BAR STO None
50194146000 ELCHO ISLAND 0.7
50194150000 GOVE AIRPORT None

So Darwin is clearly an outlier there.

carrot eater said:"I love it when the programmers feel the need to act as if they're an authority, just because they write code. They haven't a clue about the physics or the specific code involved, but since they know programs can be buggy, well of course [insert name here]'s code must also be buggy."

Perhaps it is projection on their part. :)

After really looking at the graph the only conclusion that I come to is that the thermometer was moved in 1940.

The point is that the acutal temperature measurement only matters to the inhabitants of Darwin. What climate scientists are looking for is the difference from the previous year and the long term trend. You can see that the temperature measurement figures step change in 1940 however if you graft the start of the time after 1940 onto pre 1940 the trend is pretty much in line with the region.

What Willis needs to do is interview people that lived in Darwin during 1940 and ask them if the temperature suddenly dropped as he claims the temperature record shows. If heaps of people remember a sharp drop in temps or some other indicator of climate like bird migration or plant maturation also indicates a step change then he has a case.

However I am absolutely sure that Occums Razor should apply in the absence of such data and we can conclude that the thermometer was moved.

I also am appalled that ANYONE could take this as a smoking gun. Honestly even with my small knowledge of global warming there is no way that this could be an actual drop in temperatures. There would be so many other lines of evidence to support such a large drop. Why don't the skeptics ever think of this? 1deg drop in average temperatures is a big deal and I am sure that somebody else would have noticed. Also there is not corresponding drop in stations close by - need I go on......

David, Willis is lying by presenting the raw temperature graph, which mixes together several records, as the true temperature record of Darwin. He doesn't get the benefit of the doubt because of his previous dishonesty about temperatures.

He's quite transparent about that Tim. You can't be more transparent than this,

"I always like to start with the rawest data, so I can understand the adjustments. At Darwin there are five separate individual station records that are combined to make up the final Darwin record. These are the individual records of stations in the area, which are numbered from zero to four:"

If you source Darwin Airport data from GISS (http://tinyurl.com/ylpd7eh) you find the same plot as the what he used.

He also provides a long discussion regarding rational for and against adjusting that series.

Actually his analysis is all about discovering what adjustments are required to turn the raw into the (is it BOM?) adjusted series, so naturally he has to start with the published raw to do that.

The final question is whether or not the adjustments made can be justified in terms of the stations known history and proximity to other reference stations. He says no, others might say yes.

I posed this question above which I believe is pertinent,

"Why is the adjusted almost linearly diverging from the raw up until about 1985? What kind of station anomaly manifests itself in a linearly changing fashion over a period of 40 years?"

Raw data for stations from Bureau of meteorology can be found at http://www.bom.gov.au/climate/data/weather-data.shtml

There are two significant stations, one Darwin Airport starting in 1941, which shows an increasing trend, and the other is the post office, which ends in 1940, and has a decreasing trend, mostly before 1910.

The post office record ends significantly warmer than the airport record trend starts, so reasonably obvious that the step change in 1940 is due to the station move.

What is also interesting is the GHCN record for 'Darwin Airport', which I looked up from GISS at http://data.giss.nasa.gov/gistemp/station_data/

This record has 5 stations. The first seems to correspond to the post office, however ends in 1985 and includes the step down in data after the change to the airport location

The second station would seem to correspond to the airport station, and starts in 1940. However there is a significant step down around 1990 which is not in evidence for the BOM record which shows a continued rise through the 90s and 00s.

The other three GHCN stations appear to be shortened versions of the airport record.

This is then interpreted by GISS as an overall decreasing trend since 1960 (GISS throws away the earlier data).

It appears that GHCN have included an error in their station data which inadvertantly attributes data from both airport and post office to one station, and has also spliced recent data from an unknown (to me) source into the airport record. Maybe due to sourcing the data from a party intermediate to the Australian Bureau which had amalgamated the stations?

And if you look up the data for Darwin Airport station after the homogenization and adjustments have been applied, you see a cooling trend, not a warming trend.

The end result is a cooling trend in GISS when the raw data shows a warming trend. So much for the smoking gun that proves that climate scientists are always adjusting the data for warming....

[Ender](http://scienceblogs.com/deltoid/2009/12/willis_eschenbach_caught_lying…).

Your last point is a telling one.

If one wanted to confirm the station shift in the forties, there would surely be plenty of amateur records around that could be used to corroborate the anomaly trend at the time. Farmers, gardeners, and even old dudes who are the meteorological equivalent of twitchers or train-spotters keep such records, and there are bound to be old boxes, shelves and drawers that still hold this information.

In fact I'd be surprised if the BoM hadn't already long ago sourced such material with which to confirm any calibration.

Ender,

Can't comment on what the weather was like in the 1940's, but just out of interest, last night Darwin had it's equal record overnight minimum - a lovely 29.7 degrees. I didn't even have the a/c on!

Tim, thanks for bearing this idiocy in such an insightful manner, or as my English teacher said, "with relish and aplomb."

Eschenbach needs to choose between his being dishonest or his being ignorant.

"That blue line for raw temperature in his graph combines different records without any adjustment, even though Eschenbach could see that there was a step change between record 0 and record 1."

What! The blue line for raw temperature presents temperatures that have not been adjusted?

Scandalous!!

Keep up the good work Tim. Dont let those denailists succeed in passing off raw temperatures as if they were ... raw temperatures.

You aren't in any position to be calling anyone a liar.

David, Jones is lying by presenting the tree proxy graph, which mixes together tree rings and temperature records, as the true proxy record of Briffa. He doesn't get the benefit of the doubt because of his dishonesty about confidentiality agreements restricting data to academics only.

Posted by: Tim Lambert | December 9, 2009 11:33 PM

Why do the adjustments move up a full degree from 1941 to 1995, or so in a haphazard way? If you have to make that many corrections to data, then you have no data. What you have is fiction.
The raw data looks pretty good from 1941 to present. The earth is saved. Sleep well tonight.

Ryan asks: ""Why is the adjusted almost linearly diverging from the raw up until about 1985? What kind of station anomaly manifests itself in a linearly changing fashion over a period of 40 years?""

Gee - I wonder. Perhaps some trees were planted nearby?

How about you whiners stop whining like girls and just go and do your own primary research and then publish it in Nature so we can all benefit from it?

In case you _don't_ want to be sued for slander, you might want to note that the 'high quality' BOM data is already massaged.

See:

http://tinyurl.com/yeoom2b and http://tinyurl.com/ybfeba4

Since getting sued for slander sucks.

[Steve Mosher](http://www.sourcewatch.org/index.php?title=Steven_W._Mosher),

Have you ever herd someone claim "overpopulation is a myth"?

That would have to be one of the nuttiest of religious wack-job ideas (up there with condoms are wrong, and condoms won't stop HIV). All of which will continue to bring terrible suffering and scaring on the ecosphere and humanity.

I can't think of a worse set of lies to perpetrate. Denying climate change will delay action and produce a compounding the effect ombined with over population.

Instead of a soft landing, this compound denial will cause the worst type of crash, famine, dislocation, and violence.
The most vulnerable will suffer, as they are suffering with HIV due to lies put out by the head of the Catholic church.

A price on carbon will bring some beneficial social opportunities; I will also bring us closer to longer term sustainability. Thought this also involves bring forward realisation of the some of the real costs of living that we have been able to defer. That will mean that the lies about the population level not being a problem will be harder to maintain.

If you read Eschenbach's original post, he realises there was a station move at 1941 within that record. He thinks about making an adjustment for it, shows what such an adjustment might look like, and then decides against it for no particularly rigorous reason. It's that decision that needs to be revisited.

Further, Willis goes to GISS to get data, notes 5 different series for Darwin Airport, and notes how similar some of them are. Well, they seem similar because they're probably duplicates; the same data over and over again. But he uses that as a reason to not want to adjust them, because they "show exactly the same thing". Did he consider that was because they were the same measurement?

In any case, it is interesting that GISS avoids the mess at 1941 by simply leaving it out.

Janet: I'm not sure what you're ranting about at 6:07, but it seems off-topic. Let's stay on track.

Steven Mosher believes in the science. Still, his cheerleader briefs have "GO Skeptics!" stenciled on the rear. :)

73 Carrot,

Yes, Janet was OT but did you click on the link at the top of her post?

75 TrueSceptic,

I don't read links unless I have some idea of why they're there, and what manner of site it's taking me to. Now looking at that one, I have to agree with Mosher on some points: population control in China absolutely has involved violations of human rights and various ill consequences. I hope he doesn't generalise that to family planning efforts elsewhere, though - the natural trend is for fertility to decrease as nations become wealthier and as fewer people are directly involved in agriculture. You don't need to jump to China's draconian policies in order to decrease fertility.

52 MarkB,

(And anyone else who's interested.)

Some URLs contain underscores. Markdown reads these as paired tags to convert to and from italics so if you post the raw URL the underscores are lost and that part of the URL appears in italics.

_ between underscores _ becomes _between underscores_ and

http://data.giss.nasa.gov/cgi-bin/gistemp/gistemp _ station.py?id=501941200004&data _ set=1&num _ neighbors=1
becomes

http://data.giss.nasa.gov/cgi-bin/gistemp/gistemp_station.py?id=5019412…

The way to post URLs is as described just above the comment box: square brackets around the description and round ones (parantheses) around the URL, like this
[ description ] ( URL ), e.g.
[GISTEMP station](http://data.giss.nasa.gov/cgi-bin/gistemp/gistemp_station.py?id=5019412…)

(Of course, I've added spaces so you can see the underscores and brackets when I want you to. Do not include the spaces immediately inside the brackets.)

What's truly ironic about the "they're adjusting the raw data!" meme is that the raw data is what Watts et al attack as unreliable (SurfaceStations.org, the urban heat island effect, etc.). Deniers say that siting problems render the raw data unreliable, and then they say that adjusting for those siting errors is evidence of fraud. You know deniers are being truly disingenuous when they attack both the raw data AND the use of statistical techniques to adjust for the errors in the raw data.

Of course, that's just the start of deniers' hypocrisy. I've started a list of all the examples of this hypocrisy. I've come up with 11 for a start - feel free to add more.

http://akwag.blogspot.com/2009/12/climate-of-hypocrisy.html

What Willis needs to do is interview people that lived in Darwin during 1940 and ask them if the temperature suddenly dropped as he claims the temperature record shows. If heaps of people remember a sharp drop in temps or some other indicator of climate like bird migration or plant maturation also indicates a step change then he has a case.
However I am absolutely sure that Occums Razor should apply in the absence of such data and we can conclude that the thermometer was moved.

It certainly did move since the Post Office where it was housed was destroyed by bombing in Feb '42! [PO]
[PO2] Also as a result of that bombing (and subsequent attacks) most of the population of Darwin was evacuated (most women and children ~2000, in Dec '41).

WAG, I noted that above. They only like adjustments that have the most recent data go downwards. No other adjustment is allowed.

The methods used for the adjustments are published. If there are specific factors at a given site (station move, time of observation change, etc), they can ask the relevant weather bureau. They can try to work out all these things for themselves, to see what's going on. Maybe they'll actually find something strange here and there. But to do a half-hearted analysis and then just accuse everybody of fraud?

I don't understand what could have happened to a station to even adjust the temperature upward.
It' the heat island effect that would skew a station, and you'd have to adjust the data downward not up.
What could have cause a station to show a colder reading do to bad station citing and therefore have a rationale of adjusting the temps up?
Site relocation? Still you'd adjust the older site down and not mess with the new sites data.
It doesn't look good for AGW activists.

78 WAG,

You might like my [HTBAGWS](http://scienceblogs.com/deltoid/2008/06/eli_rabbets_guide_to_climate_t…). Feel free to use anything there that's of use. :)

BTW. All five stations, including Darwin were adjusted dratically so the notion that this issue is the result of cherry picking by skeptics is irrational.
A vast magority of stations have been adjusted upwards.
There are ones that haven't ,you say? You don't need every station adjusted upwards to make the average show a warmings when their isn't one.

I'm a little late, but Willis and I have gone at it numerous times.

I don't think Willis is a liar, he is a narrow-minded ideologue injuneer, who seems pathologically incapable of widening his extremely narrow view. 'An idiot' is apt as well. But he is so pathologically narrow-minded he can't see the truth to lie about it. In my view.

Best,

D

WAG
"What's truly ironic about the "they're adjusting the raw data!" meme is that the raw data is what Watts et al attack as unreliable"

What you say is true, but I don't know why you think it's ironic or hypocritical. The surfacestations project claims fairly convincingly that over half the sites in the US are so badly situated that their readings are probably affected by 5 or more degrees C. They are already reading too warm. Then they have a number of additional adjustments made which in almost every instance appears to decrease older readings and/or increase more recent readings.

Shawn Sene | December 10, 2009 10:55 AM

"I don't understand what could have happened to a station to even adjust the temperature upward. "

Have you spent any time at all, trying to find out? Any number of things - site changes, changes in the local environment, changes in the time of day the observations are recorded, changes in how the mean is calculated - all these things can have you go upwards or downwards. There is more to the subject than just UHI, despite the impression you might get in certain circles. Heck, even Willis's post on WUWT gives a quote about that.

"All five stations, including Darwin were adjusted dratically so the "

Which five?

You know deniers are being truly disingenuous when they attack both the raw data AND the use of statistical techniques to adjust for the errors in the raw data.

...especially when they make absolutely no attempt to describe, or show they understand, the aforementioned statistical techniques used to adjust for errors. Seriously, if you honestly think there's a problem with the adjustment of raw data, then you're pretty much obligated to understand how such adjustments are done, and to base your criticism on such understanding. All the denialists do is shout "It's adjusted, so it can't be trusted!" over and over; which proves they have no intention (or ability) to bring honest criticism to the grownups' table.

'He's quite transparent about that Tim. You can't be more transparent than this,

"I always like to start with the rawest data, so I can understand the adjustments. At Darwin there are five separate individual station records that are combined to make up the final Darwin record. These are the individual records of stations in the area, which are numbered from zero to four:"'

No, that is not transparent. Transparent would be showing the five different stations' temperature records in different color. It'd also make for better analysis, as the adjustment would be apparent because of the change between temperature stations. Instead, Wills lumps them together in one great turd of a graph, confusing himself and those eager to be confused into thinking there's a great conspiracy when in fact *the old temperature station got bombed*.

This is why technologists like Wills shouldn't try their hand at science.
He's either incompetent or deceitful or both.

"David, Willis is lying by presenting the raw temperature graph, which mixes together several records, as the true temperature record of Darwin. He doesn't get the benefit of the doubt because of his previous dishonesty about temperatures.

Posted by: Tim Lambert | December 9, 2009 11:33 PM"

Are you claiming then that the whole IPCC process is a lie... because they often take different records and mix and match them together... and unlike Willis who clearly writes thats what he's doing the IPCC has a habit of not writing any notes on that habit next to the graphs...

None,
John V took the stations rated as a 1 or 2 by the surface stations project and showed that the temperature curve found using only those in the US matches the published curve. This means that the issue of siting has been dealt with in the published curve. This result was of course promptly ignored and poo-poo'd by the CA and WUWT crowd. They have declined to make their own analysis (probably because they know the answer would not support their ideas.)

"It certainly did move since the Post Office where it was housed was destroyed by bombing in Feb '42! [PO]"

You mean Keith Briffa and Phil Jones went back in time and convinced the Japanese military to bomb the weather station, knowing it would show cooling instead of warming.

The fiends! This conspiracy goes deeper than we thought!

"None, John V took the stations rated as a 1 or 2 by the surface stations project and showed that the temperature curve found using only those in the US matches the published curve."

Wasn't there also a published article taking 64 of Watt's approved stations and the general temperature trend which also found no difference?

Also, there was a comparison in a Nature article of temperature trends in high-prevailing wind speed versus low prevailing wind speed stations. The idea being that if there was a UHI effect causing the temperature trend such a trend would be lower in stations with higher heat transfer because of higher wind speeds that stations with lower heat transfer. Bubkes again. But nothing's going to dissuade Watt's from his mission.

"This result was of course promptly ignored and poo-poo'd by the CA and WUWT crowd."

IIRC, it wasn't ignored by WUWT. Watts threw a hissy fit at not being included as a reviewer for the published article, and I think he gave John V the cold shoulder.

Sock Puppet of the Great Satan | December 10, 2009 11:38 AM

"Transparent would be showing the five different stations' temperature records in different color."

They are in different colors (GISS IDs 501941200000,...1,2,3 and 4). You just can't see them all because I'm guessing they're all the same data from the same station, aside from the station move in 1941. I don't know that for sure, but it sure looks that way; it's worth asking somebody. One of the things Willis might have thought to ask somebody, before making assumptions?

It's mainly the decision to just flat-out ignore a station move that bugs me. At the very least, he could have kept his simple correction for that in there.

In any case, I don't think he's lying. He just has a track record of being error prone, and it's amplified because he puts his work where a lot of people see it.

90 Gator,

And where _is_ Watts's epic Surface Stations publication? Surely he would have got it out in time for Copenhagen?

The surfacestations project claims fairly convincingly that over half the sites in the US are so badly situated that their readings are probably affected by 5 or more degrees C. They are already reading too warm.

Then maybe they should use temperature anomaly.

As bad as the disinformation was, it was also a great opportunity for me to learn more about climate science - particularly how physical changes at weather collection sites affect the data collected and how scientists need to adjust for those changes.
Keep up the good work Tim!

Re #57. Hmm, wonder why they chose the Darwin area when GISS (which is one) shows that +2.9C. Can you say "cherry picking". Look at that cursory list, very few stations have any adjustment at all.

It would be naive to assume that the folks at WUWT stumbled on Darwin randomly. Now of course ALL the data records are fudged (sarc). That said, I would be surprised if these global SAT data were 100% error free; they are just too complex to have that expectation. The big question is are the errors so serious as to affect the global SAT record? Willis has shown unconvincingly that there may be a problem with one site. Funny how he happened to choose one that fit his hypothesis off the bat.

Seriously though, had I been looking at those adjustments for Darwin and saw that +2.9 C in the GISS then I would have been forced to delve a little deeper to try and figure out what was going on. Those "human" adjustments are a lot more reliable than WUWT would have us believe, the code just follows instructions, but the code can't always account for all possibilities. A suspect that there may be a few such examples in the SAT data, but that they will not affect the overall global trends. Wonder how many records in there which might have been adjusted significantly downwards?

Anyhow, worth repeating. Willis did not follow the official procedure as set out in the journal papers, as far as I can tell. Sample size of one. He obviously went into this hoping to find something, so bias.

But hey, we find one problem and then before determining whether or not it really is an issue, we blast it all over the internet and make premature and serious accusations. And they wonder why people think that they (WUWT) have no credibility. This is the same tactic CA use, find one mistake, no matter how trivial and regardless of whether or not it affects their conclusions, and then use is to character assassinate and undermine someone's credibility. Nice way of fabricating doubt.

If Willis can do a proper work up, clearly display his methodology for others to replicate instead of some rambling text, then maybe others can help.

Again, I challenge him (one of his moles here can relay the message) to invest some time and effort on this, develop a superior method and publish it in BAMS or J. Climate. Until then, the current data hold.

What you say is true, but I don't know why you think it's ironic or hypocritical. The surfacestations project claims fairly convincingly that over half the sites in the US are so badly situated that their readings are probably affected by 5 or more degrees C. They are already reading too warm.

Thanks for reminding me to watch climate crock of the week again, the one about the surfacestations project is one of the best.

<http://www.youtube.com/watch?v=P_0-gX7aUKk&feature=player_embedded

""Transparent would be showing the five different stations' temperature records in different color."

They are in different colors (GISS IDs 501941200000,...1,2,3 and 4)."

Yeah, but Willis doesn't show them that way. Or show the correction adjustment in different colors according to station.

I'm inclined to incompetence rather than mendacity, but it is remarkable sloppiness on his part.

And where is Watts's epic Surface Stations publication? Surely he would have got it out in time for Copenhagen?

Now ... that is a very good point ...

"Yeah, but Willis doesn't show them that way."

Huh? What are you looking at? In his Figure 5, he's clearly using different colors for the different raw data sets he found in GISS. You can't tell the difference between them because they pretty much overlap exactly.

Which would indicate to me that they're probably duplicate records from the same instrument at Darwin Airport. I rather doubt there were four different weather stations, all at Darwin Airport, in 1970. Though I suppose I could be wrong.

OK, I am going to ask some really dumb questions.

Say a station "X" has a systematic warm bias, then isn't the temperature anomaly with respect to the climatology at that site independent of the bias? So even if a thermometer is located next to a parking lot, so long at the parking lot has always been there for the duration of the record, then the warm bias is not going to affect the anomaly, right?

Yes, of course, the situation gets complicated if they build a new parking lot next to it after 7 years, or they change the screen or they change the sensor. But I am trying to keep this simple.

Just how do they calculate the global SAT anomalies? Is it the mean anomaly for all station anomalies, or do they calculate the mean global temperature and compare that value with the mean global value for the baseline period? Told you they were dumb questions.....

Anyhow, WUWT if what I said in the beginning is true, then the folks at WUWT are being misleading when they show those pics of thermometers near buildings etc.

Just caught another odd error in Eschenbach's post:

Another odd fact, the GHCN adjusted Station 1 to match Darwin Zeroâs strange adjustment, but they left Station 2 (which covers much of the same period, and as per Fig. 5 is in excellent agreement with Station Zero and Station 1) totally untouched. They only homogenized two of the three.

That's just not true. Station 2 has been homogenised too.

To 102 MapleLeaf

I think the main issue being raised by the by the surface site pics is that

1. Adjustments have to be made to make them work. Which in turn means that your not getting the best data possible.

2. Even if we use your parking lot argument and no other parking lots were built... that doesn't account for how many cars using the parking lot and many other factors such as if trees were planted and thus over 15 years grow and added shade to the parking lot(thus causing cooling) or if the parking lot started out with trees and then they were cut down(which would cause warming). Or big buildings being build that block wind to the area and hundred if not thousands of other major changes.

3. The adjustment data and reasons for adjusting the data on a station by station basis have not been released in any detail. Thus we don't know what effect or counter effect have been adjusted into the data and why they were done.

4. When dealing with science one should always be trying to use the raw data straight from the object used to measure it. When you have to make adjustments you are by default saying the raw data is flawed. A classic case would be setting up guide lines as to where/why/how stations are placed/setup(which places like NOAA have created).

The reasons for these guide lines is to ensure that the raw data is the best data and that little to no adjustments are made to it. When people go out and find stations that don't meet these guide lines then you have to play the adjustment game... and it is a game of mostly opinion. One person says XXX effects the station, another says YYY effects the station, a third says neither XXX or YYY effect the station but ZZZ and AAA effect the station and so on and so on.(which is really what this debate is about).

The problem is that some people are claiming that XXX, YYY, ZZZ are effecting the stations however they refuse to give a complete detailed report to why they believe XXX, YYY, ZZZ are effecting the station... and why the station wasn't moved to a more suitable place to get better raw data.

The problem many ppl have with the adjustment is that the urban heat bubble effect is pretty well known and understood... which means that as urban areas encroach on stations those stations should produce warmer reasons... and thus should be adjusted down. That however does not appear to be the case... as many stations are either unadjusted or adjusted to make them warmer.

However once again it is a matter of opinion and a guessing game because the complete detailed reasons for adjustments per station are normally overly simplified(thus hard to confirm or deny), have no complete records, or they have been refused to be released to the public.

@Sock Puppet of the Great Satan #92:

Just to clear some things up on the John V / Anthony Watts issue:
NCDC wrote a "talking points" memo:
http://www.ncdc.noaa.gov/oa/about/response-v2.pdf
which originally only referred to the surfacestations website. JohnV did something similar and got the same result, but he only reported it on WTFWT (sorry, I can't write that properly).
Watts has tried to talk his way out of it by complaining about all kinds of small issues, and claiming he has new data, so the old data is wrong.

Of course, as indicated by others, he never did the real analysis himself. I doubt he ever will, it's better to attack small parts at a time. After all, when you get rebutted, you always have the next thing to attack. And then the next. At one point in time he will find something that stands the scrutiny, upon which his minions will roar in unison that they've shown it all to be a fraud...

Thanks, TrueSkeptic. Giving it a try...

[Darwin unhomogenized data](http://data.giss.nasa.gov/cgi-bin/gistemp/gistemp_ station.py?id=501941200004&data_set=1&num_neighbors=1)

I like WAG's comment #78.

"What's truly ironic about the "they're adjusting the raw data!" meme is that the raw data is what Watts et al attack as unreliable (SurfaceStations.org, the urban heat island effect, etc.)."

Contrarians only think adjustments that result in more warming are fraudulent and the raw data should be left alone in those cases. The opposite adjustments are perfectly fine. Hypocrisy? Sure. It's more revealing of extreme bias, not associated with good skepticism. I just saw Carrot Eater wrote essentially the same thing.

MapleLeaf: You're on the right track. Heck, forget the parking lot; just changing the altitude of the thermometer above the ground will affect the absolute temperature you measure. But the anomaly would be the same. So long as you don't change the altitude over time, you don't need an adjustment. (And yes, I'm pretty sure that to compare anomalies across different sites, you have to calculate them using a baseline specific to each site).

All that said, you still do adjust for UHI; the method is described by GISS and the code is available at GISS. Anybody saying "but we don't know how they do it" is just being lazy. It's there for you to look at. Go study it, if it means so much to you.

It's typical laziness all around in the sceptic circle, actually. They stockpile a list of things that might affect the results, and then don't do the work to see what the effect is; they're just happy to create doubt and stop there. Happily, others do the legwork. Want to see the results using only rural stations? That's been published. Want to see windy days vs calm days? Been published. Want to see only the stations rated by WUWT to be high quality? Also available, though not yet formally published (as far as I know).

If the surface record was just noise, then the anomalies wouldn't spatially correlate so well. But they do: look at the anomaly at site A, and compare it to neighboring sites, and it'll be similar.

I'm still waiting for pictures of parking lots in the Antarctic Peninsula, Siberia and northern Canada, where the warming has been fastest.

Thanks carrot eater. Yes, our North is warming rapidly as has been determined by satellite, radiosondes and thermometers.

This "work" by Willis just seems to be more obfuscating on their part. I honestly believe that they have no interest in moving the science forward, just doing their best to muddy the waters.

If I were to go on the net and claim that NOAA or BOM screwed up, I would want to have done a lot of work and to have all my ducks lined up. Actually, scrap that, I would simply not do that, I would try and get it published and/or go directly to NOAA or BOM with my extensive evidence.

Speaking of data adjustments, the following 2008 study concludes (as a bit of a sidenote to the main thesis):

[Thompson et al.](http://www.nature.com/nature/journal/v453/n7195/full/nature06982.html)

"compensation for a different potential source of bias in SST data in the past decadeâ the transition from ship- to buoy-derived SSTsâmight increase the century-long trends by raising recent SSTs as much as 0.1 C, as buoy-derived SSTs are biased cool relative to ship measurements"

It was also detailed in Smith et al. 2008:

"Because ships tend to be biased warm relative to
buoys and because of the increase in the number of
buoys and the decrease in the number of ships, the
merged in situ data without bias adjustment can have a
cool bias relative to data with no shipâbuoy bias. As
buoys become more important to the in situ record, that
bias can increase. Since the 1980s the SST in most areas
has been warming. The increasing negative bias due to
the increase in buoys tends to reduce this recent warming.
This change in observations makes the in situ temperatures
up to about 0.1°C cooler than they would be
without bias. At present, methods for removing the
shipâbuoy bias are being developed and tested."

[Smith et al.](http://www.ncdc.noaa.gov/oa/climate/research/sst/papers/SEA.temps08.pdf)

Any update on when SSTs will be corrected for this cool bias? I imagine the fits the conspiracy nuts will throw.

To 107 carrot eater

"It's typical laziness all around in the sceptic circle, actually."

I find the ignorance of this statement very impressive. One need only review a post posted doc to get this statement from the NOAA

"Surfacestations.org has examined about 70% of the 1221 stations in NOAAâs Historical Climatology Network (USHCN)[...]at the present time this is the only large scale site evaluation information available so we conducted a preliminary analysis."

Yes clearly conducting a massive survey across the whole US is being lazy... more so they did it completely free and on personal time.

Most people want information directly from the source so that way they don't get into the very arguments being discussed in this very comment section... that they did the process wrong, forgot something, etc.

This also becomes an issue with keeping track of the data. One shouldn't have to shuffle through thousands of papers to create a data base... that stuff should be handled by the people collecting the data as they log the data in. Back in the 60s you could argue that logic but today with computers and the internet you have zero excuse to require people to search through hundreds of papers to get tiny bits of from each. This data should be readily and easily available. The fact that it isn't shows not only a lack of book/record keeping but in fact slows down science by a massive amount as it takes a long time to gather needed info.

The fact that CRU "lost"/discarded decades of raw data just highlights this simple fact.

robotech master:

You completely missed the point. The volunteer effort to go look at all the stations: that part wasn't lazy; it was actually somewhat useful. (Though it'd have been better if there was some check on whether all the volunteers were being consistent).

The lazy part is not using the information they found. The lazy part is going: "look at all these stations that aren't perfect", without doing the additional work required to see what difference it actually makes.

As it happens, it doesn't make much of any difference at all. But it seems the people involved were quite happy to not do that analysis, and simply assume that it made a big difference.

#10 Carrot Eater: "[Eschenbach is] just somewhat less than competent, and in this case, amazingly fast to lob accusations of fraud. When shown how he's messed up, he's then rather stubborn until he doesn't have much choice but to admit he messed up."

#40 Willis Eschenback: "I'm answering questions on my posting at [wutssup]. Please post any scientific questions and objections there, I will answer them as time permits."

# 92 Sock Puppet: "IIRC...Watts threw a hissy fit at not being included as a reviewer for the published article..."

If so, then there it is: In the United States a few decades ago, the slang term was "hot-dogging" -- drawing attention to yourself by jumping out -- in these cases without having the training, nor doing all the research.

The goal is to gain entry into the discussion, and to reserve an importance place for yourself within it.

As with Lomborg, it's a version of policy-entrepreneurialism, and it is not far from the method of the old think-tank chop-shops, formed to produce "studies" to arm the lobbyists.

Here, your problem is: climate is a complex system science, and there is already a multitude of experts at work -- but then again, that complexity means that the public can't detect your intellectual shoddiness, and allows you more avenues to wiggle out.

So: make a noise, then backtrack if you're discovered to be in error. But always keep those eyeballs coming!

To 111 carrot eater

Give them millions in grant money and I suspect they would quickly get that done for you... you seem to forget that many of the people that did that survey solely in they're free time at personal cost to themselves. Plus add in ITS NOT THEY'RE JOB. The simple fact is that data collection, storing and keeping of the temp records need a massive review... and yet no one has any interest in doing it. The people who main job is doing just that are failing at it. You can't call people volunteering on their own time to do something lazy when people who are payed to do it full time aren't doing it. By defacto you are calling everyone at the IPCC, CRU, NASA, etc lazy as well.

As MarkB already posted above a study that was done in 2008 which once again shows that data collection has been put into question. Every time someone looks into the data something needs to be fixed period. Thats not a sign of good science. Its very clear that all the data should be reviewed... and not by unpaid volunteers...

The adjustment data and reasons for adjusting the data on a station by station basis have not been released in any detail. Thus we don't know what effect or counter effect have been adjusted into the data and why they were done.

GISTEMP source is available. Your answers are there.

To 114 dhogaza

You got a link?

Try [here](http://data.giss.nasa.gov/gistemp/sources/)

For the record, MartinM's link @#116? In response to Robotech Master asking for a link to the GISTEMP source code? First hit on Google for "GISTEMP source code".

Robotech Master, you bemoan others as "lazy" when you are too lazy to do a fraking Google search?

Robot Master, you appear to be clueless as to how grant money is spent and on whom it is spent. Skeptic myth number 2457. Are you talking research chairs that are funded or some professor or someone working for NOAA? You seem to have this wild idea that every professor's salary is paid for by grants. Umm, no. They use grant money to purchase equipment, pay for field work or data gathering, and most importantly these grants are used to help grad students etc. Maybe it is different elsewhere, but in Canada unless you are a research chair, you do not benefit directly (in fiscal terms) from research grants.

You also seem hopelessly naive as to the complexities of these data sets. If you expect 100% error free data, then I honestly have no idea on earth where you would find such a data set. Each and every data set has its issues. As I said before, scientists do the best they can with what they have. Is there room for improvement, oftentimes yes. But Willis is not moving the science forward, he is just trying to muddy the waters and telling Anthony's devote followers what they want to hear.

As someone pointed out above, you guys need to make up you minds, you contrarians are arguing that the raw data are crap, then when someone tries to make scientifically-based corrections to those "crap" data they are accused of "fudging/massaging" the data. The truth is that cynics like you will never be happy. Time to move on and start fabricating the next myth or "controversy" pertaining to AGW. In the meantime, the planet continues to warm-- and we do not need a single thermometer to demonstrate that.

Lee Arnold explained WUWT's motive perfectly. Can I please use that Lee?

robotech master | December 10, 2009 1:49 PM

Give me a break. I'm not talking about the volunteers. I'm talking about Watts and some of his friends. If he's got time to run that blog, and time to publish a little book about the surface temperature record, and others there have time to come up with all manner of blog science, then somebody over there has the time to do a bit of analysis, to see what effect the surface stations project actually would have on the results.

As it happens, not much, as the NOAA and apparently somebody named John found. But this isn't surprising, because (perhaps unnoticed by you), those who get paid to do these things have already been looking for problems like UHI, and seeing how robust the results were. That's the whole point of the homogenisation that you guys are suddenly so upset about. They wouldn't be doing the homogenisation if the raw data were perfect.

We can't go back in time to 1850 and put all the weather stations in all the perfect places, and then post guards there to make sure nothing ever changes nearby, and that nothing in the procedures change, either. We have to work with the data we have. So make up your mind - do you want only raw data (and greatly limit what can be used), or do you want to adjust for stuff like UHI? Or do you only want to make adjustments that go one way?

I will review it but at first look this appears to be software and source code not station by station adjustment data with reason for it.

I going to try to get it up and running do you have any idea where the station by station info is listed?

I'm talking about Watts and some of his friends. If he's got time to run that blog, and time to publish a little book about the surface temperature record, and others there have time to come up with all manner of blog science, then somebody over there has the time to do a bit of analysis, to see what effect the surface stations project actually would have on the results.

Not to mention he promised he would. Not to mention that over the summer he claimed to have some people schooled in stats working on it.

those who get paid to do these things have already been looking for problems like UHI

Someone recently did some detective work I wish I'd saved somewhere - they found a first reference in the literature to the UHI 'way back in something like 1937 ...

"You can't call people volunteering on their own time to do something lazy when people who are payed to do it full time aren't doing it."

Yes they are lazy. Lazy and arrogant. Too lazy to do adequate research on the topic, and too arrogant to realise they need to do more research to be qualified to comment. Just like you.

Yes clearly conducting a massive survey across the whole US is being lazy... more so they did it completely free and on personal time.

No one can compete with the stamina of a crank. Their energy to deny, confound and exasperate is boundless. However, I think he meant intellectually lazy.

I will review it but at first look this appears to be software and source code not station by station adjustment data with reason for it.

I going to try to get it up and running do you have any idea where the station by station info is listed?

Just like a denialist, wants everyone to do the work for him ...

Look, lazy ass, you're the one making ignorant claims about the results being bogus, along with whining that code and data aren't available.

You're wrong on all counts. Knock yourself out ... learn the Google happy dance and find the stuff yourself.

To 118 MapleLeaf

Let me see if I understand this correctly...

I claim "give some grant money to the ppl do the survey."

You say and I quote "you appear to be clueless as to how grant money is spent and on whom it is spent.[...]pay for field work or data gathering."

Now forgive me but last time I checked the survey WAS FIELD WORK AND DATA GATHERING...

I also find your implied beliefs about me pitiful at best... You claim all kinds of things and insult me because I dare question you, you ignorant little whelp please. You want this to turn into a insulting match go for it you will lose you tiny little fool.

To 119 carrot eater

Its beside the point running a blog is a minor distraction... how many working ppl blog every day... lots. So thats a pretty weak argument.

I also don't disagree that homogenisation needs to happen its just that the data needs alot more homogenisation from the fact that its not being done properly, completely or being kept up to data. Once again it seems every time someone looks at the data more issues are found... which is good. However a more complete review process should be done instead of the current piece meal which is in fact making things more confusing for everyone involved.

I also don't disagree that homogenisation needs to happen its just that the data needs alot more homogenisation from the fact that its not being done properly, completely or being kept up to data.

You've got the source code. Please show us where it's not being done properly, completely, and how that affects the temperature product in a statistically significant way.

robotech: You very much will want to read Peterson, "An Overview of the Global Historical Climatology Network", BAMS 1997. Surely GISS has published their methods somewhere, too; I just don't have the papers handy.

What you're asking for is called 'historical metadata' - the actual history of changes at the station. This isn't generally available to NOAA or NASA, so the homogenisation is done statistically. Though I think that metadata is available and is used for the US reference network; check me on that.

If you want the historical metadata for a station, you'll need to ask the national weather service of the country. In this case, the Australian Bureau of Met. I don't think it's put online, so you would have to ask after it. Based on material I saw at their web page, there might be some detail in somebody's PhD thesis from a few years back.

For the record, Robotech Master, the "adjustment data" is the raw data plus the described adjustments.

The raw data is publicly available.
The adjustments are described publicly.
The code to apply those adjustments *to* the raw data are publicly available.

But, of course, the raw data + the peer reviewed literature describing the adjustments + the raw code for applying said adjustments is not the same as "station by station adjustment data" in the same way that your complete banking records are not the same as a present balance.

"Huh? What are you looking at? In his Figure 5, he's clearly using different colors for the different raw data sets he found in GISS. You can't tell the difference between them because they pretty much overlap exactly."

Not in his figure 7 (reproduced by Tim above) he doesn't. If he'd color-coded the adjustments or raw data for when certain station series start, it'd be obvious that there was some alteration to the station when the second series starts in 1941.

I agree with you they're duplicates from the same instrument (or duplicate instruments at the same station)

To 128 carrot eater

I'm shocked someone in here who understands science.

I have been looking for it and have many papers including some by Bams(think so anyway it will be added if not) to my list of reading...

When I normally quest for that data I run its complete retards like dhogaza who wouldn't understand science if it was stuck up his @ss.

I have requested some of the data before however since like many ppl I do this part time they have thus far not release a meaningful amount of info to me. I'm always on the look out for source info from other places though... once again sadly I run into far to many retards in that quest.

106 MarkB,

You added (or left in from my example) a space between gistemp_ and station.

Giving another paper:

For Australia, you'll want to read Torok (1996) and Della-Marta (2004), both in Aust Met Mag. I think the latter is linked above. These describe how the Australians do their adjustments, and leading to plot Tim Lambert put above from the BoM. You'll see that they use both statistical methods and the historical metadata. You'll see that the statistical methods are able to independently detect changes confirmed by the metadata, but the metadata aren't good enough to rely on it alone. They do acknowledge a bit of subjectivity in the process.

But in any case, it then becomes interesting to compare the adjusted data of the Australian BoM, GHCN and GISS. They're all doing their own thing, and the Australians have access to the metadata. It's perhaps worth nothing that the GISS homog data doesn't start until 1963, so perhaps their algorithms found a reason to reject the earlier data as being unreliable, I don't know.

Robotech Master -- the idea that scientists should only use "raw data" is nonsense. For example, I worked on a particle detector. My raw data was a time series of voltages. I need to use these data to extract a temperature rise and ionization inside my detector. Volts != temperature or electrons. How do I do this with raw data?

In fact, one needs to fit data (for example to extract pulse heights out of a noisy signal), one needs to calibrate (for example fit pulse height vs. temperature) etc etc. Some of the temperature sensors we used in our systems were highly non-linear, not to mention we needed to account for other pieces of the measurement chain.

How the heck could someone use the "raw data" only? The practice of science is all about using what data one has and getting the most one can out of it. You are obviously not a practicing scientist and you should think hard about what you are writing before putting your "name" to it. Where did you copy/paste that list from? Cause you oughta think twice, hard, before letting them fill your brain with crap again.

@Sock Puppet of the Great Satan

Ah, you should have been more clear then that you were talking about the 1941 discontinuity, and not the overlapping records in the 1970s. Your comment at WUWT was similarly confusing.

I agree he made a very bad decision by treating record 0 as a single record, when he knew it was two different sites spliced together without correction at 1941. If the GISS link had divided those two records into two different files, he wouldn't have treated them in this way (I don't think). Just because all the data was in a single file, he decided it wasn't worth correcting for the site move.

Really, if he does nothing else, he needs to apply some sort of correction for the site move at 1941. Well, he already did, but then didn't use it. The rest of the project of understanding the homogeneity adjustments would take more time and effort.

Some here might say he doesn't have the time and effort to do so without being paid, but without having done so, it is extremely irresponsible for him to be accusing people of fraud.

125 dhgoaza:

Oh, look: yet another amateur wanting the code and data to check out. This must be the 15-20th time here that someone harrumphs and thinks the data are flawed and wants to check it out...but doesn't know where to look...

Comedy gold. Must be a new graduating class of credulous rubes.

Best,

D

I will include them on my list as well.

In the end though the problem still is the data needs to be reviewed and most of all organized better. I have done reviews for sociology and grown to have a great distaste for the current way many people "abuse" the "reference system". The whole debate on both side abuses this system and creates untold confusion to everyone. Many of the argument both for or against would be fixed if they did a better job reviewing, updating, organizing and recording of data and the process.

Wow, calm down Robot. I appear to have misread your post. My apologies. You are, it appears, naive about the data. I stand by that.

Your rants and juvenile name calling, while quite entertaining, mean that by this point you have zero credibility.

Regarding your lovely comments about dhogaza. Dhogaza is actually very well informed and very familiar with the science.

Several people here have provided links to data sources so quite whining. How about a "thanks mate"?

109 MarkB,

I remember the whole buckets vs intakes vs buoy issue being discussed and I'm surprised that we have not seen an adjustment announced by now. The effect will be to reduce the apparent drop in temps in the 1940s, which has always seemed to be more than can be explained by increases in aerosols.

Carrot re #135. Bravo.

To 134 Gator

I find your ignorance impressive... where did I say you should only use raw data...?

Your also using the wrong analogy. The correct analogy you say volt=temperature... the problem is your argument is based on that you get a raw data reading of of 8 volts from your equipment and thus since this raw data is wrong you then adjust it to say 1 volt and thus use temperature at 1 volt as the observed temperature...

"Really, if he does nothing else, he needs to apply some sort of correction for the site move at 1941. Well, he already did, but then didn't use it."

Well, because it'd wipe out his supposed "cooling" trend and leave a moderate warming trend. The 1941 adjustment looks like its about 1.2 deg C (eyeballing it), so adjusting for the 1941 station move, but making no other adjustments, would give a ~0.4 deg C/century warming trend. Which wouldn't be as popular with the fanboys.

"The rest of the project of understanding the homogeneity adjustments would take more time and effort."

To 138 MapleLeaf

No one has provide any links to the data I have requested... only carrot eater even understands the data I have requested...

Robotech, 131:

If you ask in good faith, without making unwarranted accusations and assumptions, people will work with you. For Willis E., I'm afraid he's probably already burned his bridges in that regard, but we'll see.

The Peterson paper in BAMS is a great place to start on the issue. If nothing else, I hope you see that the data you want (specific history of site changes) is most likely not available to the NOAA unless the station is in the US; they must resort to using statistics to infer site changes. The Australians do have that site history data (see Torok), and therefore use it, in addition to the statistics.

Torok has a nice example of a station, where the stats picked out site changes, and looking in the historical notes, they could figure out the reasons why.

Therefore, it's interesting to see if the NOAA or NASA corrections are similar to the Australians'.

"Just because all the data was in a single file, he decided it wasn't worth correcting for the site move."

Possibly, being charitable, he was looking for differences between series, assuming that each series corresponded to a separate instrument or station move or instrument replacement/adjustment, rather than some of the series themselves being kludged together across discontinuities, instead of terminating one series and beginning another.

He assumed the data was cleaner than it actually was.

To 144 carrot eater

Sadly their is no such thing as "good faith" any more... as evident by many of the posts in this very comment section that make huge assumptions based on a very tiny base of information. Which is also why its very hard to get any data...

The more closed the information set becomes the more you create problems and the easier it is for either side to dismiss the other.

TrueSceptic (#139),

That's the main thesis of the Thompson et al paper, but I was referring to a separate issue. See the quote from the Smith et al study in #109 and the Thompson et al quote. This refers to a much more recent cool bias in SST measurements. The adjustment will mean the recent decade's trend will be more positive.

Robotech at 141, look at your post 104, point 4.
"When dealing with science one should always be trying to use the raw data straight from the object used to measure it."

My point is that raw data is just that, raw. It almost is never used directly simply because it can't. Your retort that the proper analogy would be adding a voltage offset is simplistic -- what about changing voltage to temperature using a non-linear equation based on a measured, arbitrary curve? There is a huge chain of inference and measurement needed in that case. There is a similar issue with using temperature measurements made over a century, by different people, with different methods and equipment.

Your use of the word ignorance is confusing... I'm not sure if you are ignorant of the meaning of the word or otherwise... Because you surely are ignorant of how science, measurement and data analysis works.

PS Robotech -- the data is there but you have to get off your ass and get it yourself. No one is going to (or should) get it for you. If you are really interested you would be going to the websites of the national institutes and looking up papers etc, not carping here on a blog.

To 148 Gator

No your argument is when you take a voltmeter and hook it up to a 9 volt battery that when it only read 4 volts you should simply adjust that to 9 volts because that what the battery says it should be...

robotech master | December 10, 2009 3:12 PM

I beg to differ on the data. In the field of climate, data (raw and adjusted) and code are amazingly available. I've not seen anything like it, in any other field, including my own. I think it's remarkable. Most everything you could hope for is available to anybody online, and if you can't find the exact code, then the published literature will describe the method well enough. About the only thing in this episode that isn't available online is the station historical metadata; you would presumably have to write somebody in Australia to obtain that.

Regarding good faith: I see a lack of good faith by Willis. He jumped to explicit accusations of fraud, without taking the time to understand what he was looking at. This isn't reasonable behavior.

Likewise, we see a lack of good faith in the people who come around, posting things that can be shown to be false with just a moment's research. Since this happens with regularity, forgive the natives for being a bit jumpy. To some extent, data-whiners are lumped in here as well, because most all the data you could want is available anyway. See RealClimate's new page on links to data. But if you just politely ask for a link, somebody will oblige.

To carrot eater

I try to aviod going to realclimate just as i try to avoid going to climateaudit.

While they may be/have sources due to the express dislike by both side of each other avoiding the "main sites" of either side seems to be prudent at the moment.

If you ask in good faith, without making unwarranted accusations and assumptions, people will work with you.

You're all a bunch of frauds and jerks and you make adjustments to make your junk theory look true so you can get grant money then rule the world through cap and trade. Now, can someone please help me find the data and code so I can make you look stupid? I'd use Google myself, but the fact that the data and code aren't falling into my lap is a sign that you're all frauds and jerks.

Hey! How come nobody's helping me? Once again, proof that you're all frauds and jerks.

While they may be/have sources due to the express dislike by both side of each other avoiding the "main sites" of either side seems to be prudent at the moment.

Yeah, avoiding the leading scientists in the field is always the best way to learn about the science they work on.

"To some extent, data-whiners are lumped in here as well, because most all the data you could want is available anyway."

It's like telling them there's free ice cream, but them complaining they have to go down the street to get it and put the chocolate and nuts on it themselves.

Data-whiners are one step removed from concern trolls.

In case it wasn't clear, I'm speaking of Real Climate.

McIntyre, who runs Climate Audit, isn't a scientist. Like me, he has a BS in mathematics.

Watts, who runs Watts Up My Ass, has a high school education.

These are the people you imagine might be disproving the work of thousands of scientists.

It's like telling them there's free ice cream, but them complaining they have to go down the street to get it and put the chocolate and nuts on it themselves.

Take that analogy a step further, since robotwit is asking to be spoon-fed the data...

To dhogaza

Yes because "science by the educated" has never been wrong...

The vast majority of the greatest scientists even history didn't even have a college educate and yet proved their far "better educated" counter parts wrong.

Science by diploma is by far the weakest science of all.

I think I said this at WUWT, but I think one useful thing might come of this: it looks like GHCN, CRU, GISS and the Australians have four independent homogenisations for this station. It would be interesting to see the similarities and differences amongst the bunch. Nick Stokes has touched on this.

Willis's entire post was inspired by the IPCC diagram that used CRU as a source, so he's barking up the wrong tree looking at GHCN adjustments in that context.

@robotech: It won't kill you to go to RC just to reach the data sources page. It'll save you the trouble of asking anonymous strangers on blogs for links. Anyway, I think it's silly to avoid the site; why read somebody else's (mis)representations of the science, when you can get it directly? It's by far and away best to read the primary academic literature; but still useful to read RC. Likewise, I went to WUWT to read Willis's post for myself, instead of relying on Tim Lambert's description of it.

"Yeah, avoiding the leading scientists in the field is always the best way to learn about the science they work on."

Avoid the studies. Avoid the qualified scientists. Avoid the data. The thermometer in my back yard indicates global warming is a hoax.

Once again, as others have pointed out, this is a very good starting point for those truly interested in the data (raw or processed):

[Data Sources](http://www.realclimate.org/index.php/data-sources/)

The vast majority of the greatest scientists even history didn't even have a college educate and yet proved their far "better educated" counter parts wrong.

Science by diploma is by far the weakest science of all.

Sigh, next he'll be telling lies about Einstein.

"The vast majority of the greatest scientists even history didn't even have a college educate and yet proved their far "better educated" counter parts wrong."

That's sheer fantasy. Post 1900 AD, at least.

But in any case, it's your contribution that matters. If Watts came up with some sound and original analysis, it would be recognised.

The vast majority of the greatest scientists even in history didn't even have a college educate education and yet proved their far "far better educated" counter parts counterparts (unless you meant sinks and faucets) wrong. Science by diploma is by far the weakest science of all.

I do understand why you dismiss the value of education and credentials, though ...

But in any case, it's your contribution that matters. If Watts came up with some sound and original analysis, it would be recognised

Yes, true, credentials only speak to the odds ... I'd bet on a thoroughbred to win a track race over a donkey with a lot of confidence, but it's always possible the horse would break down and the donkey would win.

To dhogaza

Their is a huge difference between education and credentials... but hey you probably don't even have the slightest clue as to what it is.

To carrot eater

I do go there from time to time however as both sides claim they are posting fake data and distorting it... well I'd much prefer to seek the data out then go from there sites...

Riddle; consider two words and a date: Lindzen, Choi, Aug 2009.
Hint GRL

No your argument is when you take a voltmeter and hook it up to a 9 volt battery that when it only read 4 volts you should simply adjust that to 9 volts because that what the battery says it should be...

bad example. a better one is this: the clock on the other side of the bus station is always 10 minutes late.
you keep missing the bus, because you keep using the UNADJUSTED data..

here is a funny little look, at how [electronic thermometers](http://www.ehow.com/how-does_5162287_electronic-thermometer-work.html) work.

a microcontroller ... measures the resistance and converts it into a temperature.

converts? what does that mean? i want the raw data!

Once the microcontroller has processed the correct temperature, it creates a reading.

processing the temperature? why would we do that?

But for the user to be able to gain any useful information from the electronic thermometer, it must send the information to an LCD screen. The microcontroller will apply a charge to the appropriate liquid crystal molecules in the screen, which causes them to untwist.

crystals untwist? that sounds as if they ADJUST!
i want the unadjusted crystals on my computer monitor!

Their is a huge difference between education and credentials... but hey you probably don't even have the slightest clue as to what it is.

Credentials in science require an education. You can safely assume someone who has a PhD in physics has a decent physics education - assuming it came from a credentialed university and not a mail-in fake certificate mill.

Their There is a huge difference between education and credentials... but hey you probably don't even have the slightest clue as to what it is.

Why do I get the impression you have neither?

To dhogaza

Lol nice deflection... expect that most ppl with a PhD in physics disagree with each other all the time over all kinds of things in physics...

Their are ppl who wrote in the IPCC reports who disagree with the complete IPCC report... so while arguing in the case of climateaudit or watts you maybe correct but arguing the science as a whole it is once again meaningless...

To sod

Nice try but your not doing so well...

expect that most ppl with a PhD in physics disagree with each other all the time over all kinds of things in physics...

Yes, but it's informed disagreement.

Their are ppl who wrote in the IPCC reports who disagree with the complete IPCC report...

Yes, a very small percentage. There were also scientists involved in the creation of that report who disagreed in the sense that they believe the IPCC report is too conservative, and recent data bears that out.

Robot "To sod. Nice try but your not doing so well..."

Actually sod hit the mark, but you are getting your butt kicked. But you seem oblivious to that fact.

Isn't this when we start ignoring the person (Robot) craving attention?

Most of this is really interesting. Really! I do note, though, that the arguments "raw data has problems" and "adjustments have problems" do not necessarily exclude or contradict each other even though several commenters have said they do and those comments have been accepted uncritically. Each of those arguments may have other problems and I find that discussion valuable which is why I keep reading.

Hey wait a second.

These guy's game is use different thermometers without regard to things like altitude. Naturally if they pick the "right" ones they can hide the increasing temperatures. But one has got to suspect that using the raw data in this way has got to generate plots that show increasing temperatures including temperature increases far more that have actually happened. Indeed if temperature is rising (and it is) then data unadjusted for altitude, etc. will more often than not show increasing temperature.

So unless these guys go "lucky" (or "unlucky" depending on your point of view) then they almost certainly have seen increasing temperatures and failed to tell us about it.

I would demand to see the emails of Eschenbach, etc. but I suspect they won't give them up and would scrub them before hand anyways.

OK, robot, now you're talking about nothing. Instead of making weird statements about science, why not follow up on this data.

You've got

a) the raw data for that station or stations (with a site move in 1941) (see Aust. BoM or GISS page)
b) the raw data for surrounding stations (ditto)
c) the homogenised data by GHCN, GISS, CRU, Aust BoM (see each respective page)
d) Publications describing how data is homogensised, even when you don't know what was happening at the site.

I haven't the time to mess around with this. So if you want to take this a step further than Willis, and do what Willis should have done, please proceed.

I would be interested to see the four different homogenisations overlaid on the same plot, and then a homemade attempt at a fifth, using published procedures.

Heck, I'd be curious to see just the first four adjusted sets; you could probably make that in half an hour.

To carrot eater

dhogaza dragged the debate into the gutter... not that he didn't start in the gutter to begin with...

As for the data I'm more interest in collecting and checking before using it to form an opinion about what it does or doesn't state.

hmmm | December 10, 2009 4:45 PM

Adjustments are not easy; nobody should say that they are. But it's asinine to not adjust, when you know the station moved because the previous station got hit by a bomb. Or if the observation time changed within the day. Or if somebody put a screen above the instrument. And so on. If you can't come up with a reasonably rigorous adjustment, then throw the data out.

As it happens, it looks like GISS just tossed out the data before 1963, so maybe using their method, the older data just weren't usable.

robot:

Of course; I don't want your opinion on anything just yet. But if you could post a graph of the raw data and then overlay the 4 different versions of adjusted data, you'll have done more to advance this discussion than anybody else here. Then maybe add some surrounding stations.

Heck, at this point, I might just do it for myself, instead of blathering on about it.

Beware differences in how anomaly is defined. GISS and CRU have a really annoying habit of defining it differently.

I won't be plotting any graphs for awhile... my mine goal is to create a data base with easy to search info with complete data on a station by station basis that goes into the exact process at which the raw, adjusted, graphed data was reached.

Thus making the argument made above vs watts moot since anyone will be able to easily search the data.

You're clearly wasting your time with Robot.

His english language education is primary level, which indicates he is simlarly uneducated in mathematics and science. He proves this by apparently not understanding the difference between "voltage!=temperature" and "voltage=temperature".

This is the problem with the democratic nature of the internet:
every idiot believes he's entitled to inflict his cretinous opinions on everybody else.

Vince Whirlwind

:
This is the problem with the democratic nature of the internet: every idiot believes he's entitled to inflict his cretinous opinions on everybody else.

What do you recommend? One child policy, procreation licenses, and internet censorship?

My raw data was a time series of voltages. I need to use these data to extract a temperature rise and ionization inside my detector. Volts != temperature or electrons. How do I do this with raw data?

If I understand you correctly, you should keep the raw data, including the compensation algorithm, or have your sensors calibrated by an independent electronics service firm with a good reputation. That is, if you want to be able to undo any mistakes made in implementing your compensation algorithm.

I don't understand why credence is given to something calling itself "robotech master" as knowing how science functions. It has demonstrated no particular expertise and doesn't understand that adjustments are made for many reasons, none of which have anything to do with being duplicitous but merely in the quest for a more accurate mapping of measurement to reality.
As for the voltmeter/battery example, that is not even wrong as an analogy to what has occurred with the adjustments made to the raw data which are completely explained in the literature if as Gator wrote you look it up. That is unless the goal of robotech master is to waste everyones time and draw you away from working on solutions to this worldwide threat to human existence. Can we please share and apply knowledge on fixing the problem and stop wasting time with idiots who will never, for fanatical and ideological reasons, admit to the problem in the first place?

No, I recommend everybody go to school.
...and that cretins voluntarily spare us their illiterate nonsense.

For example, let's compare:

CRETINOUS NONSENSE:
"I love how, back in the 70s, we had the same types of ominous guilt inducing music and narration, but the boogie man was exactly opposite of what it is now. Back then, it was the coming ice age, now it is global warming. "

REALITY:
"A survey of the scientific literature has found that between 1965 and 1979, 44 scientific papers predicted warming, 20 were neutral and just 7 predicted cooling."

http://www.newscientist.com/article/dn11643

> I try to aviod going to realclimate just as
> i try to avoid going to climateaudit.
>
> While they may be/have sources ....
> robotech master | December 10, 2009 3:40 PM

This is how databases work -- one record and many pointers to the record. RC has a recent thread collecting pointers to climate data sources. If you don't like that one, ask your local reference librarian for help reinventing that wheel. It will go 'round just the same.

If the robot wants to re-invent the wheel, so be it. He might learn something in the process.

Free the data! Free the code!

180 Robotech,

And you _will_, of course, ensure that your database is updated daily (or nightly) with the latest versions of all those datasets?

After all, mistakes and corrections can be made at any time.

(WoodForTrees does that, but then it only deals with the global data, not individual stations.)

181 Vince,

(I was tempted to call you VW but stopped right there.)

Anyway, all you need to say is Dunning-Kruger. :)

Lordy lordy, wtf happened to this thread? Yesterday we could keep up with the stupid. Today I don't know where to start.

Darwin is on the coast. How far would you have to move a station to get those sorts of adjustments? Would you still be in Darwin?

>Darwin is on the coast. How far would you have to move a station to get those sorts of adjustments? Would you still be in Darwin?

You wouldn't need to move it at all if you changed the housing of the termomemter. That can change the temperature by degrees by itself.

So after being temporarily derailed by a robot, it seems that we are more or less back on track. The robot incident was rather amusing and gave us a good insight as to the workings of a denialists's brain, or is is that lack of cerebral function?

To be honest I'm a tad more confused now than yesterday. Is Tim's assessment of what Eschenbach did (or did not do) correct? Also, is Tim justified in saying that Eschenbach lied? Eschenbach certainly made some slanderous and unjustified statements/generalizations, and seemingly undertook some sloppy pseudo science, but was that the end of his transgressions?

Can someone summarize why what Eschenbach did (or did not do) was wrong, point for point. I had a go very early in this thread, but it seems that there is a lack of focus in the critique. Anyone interested? And just to be fair, did he get anything right, and if so what? And what are the implications of those findings?

This is OT, but you have to watch the Monckton video at desmogblog. He refers to some youths at the Copenhagen talks as "Hitler youth", and then when one of them extends his hand to introduce himself Monckton refuses saying "I do not shake the hands of "Hitler youth"", at which point the youth tells Monckton that he is Jewish. Monckton still refuses to shake his hand and still keeps referring to him and his colleagues as "Hitler Youth", while also entering on a long diatribe about how they are killing millions of people.

I find his behaviour towards the youth disgusting! And yes, they were polite and respectful.

At this rate Monckton may have to spend a night in the local cooler.

I demand to see actual electrons !!!

194 posts later and someone is asking what's wrong with the original article. I guess it couldn't have been too erroneous.

Ryan, maybe my posy was not clear enough for you. I was inviting people to summarize the transgressions, to focus the critique. The 'analysis' in question is flawed, without a doubt. It is just a question of how bad it is.

Ryan sets us straight,

Apparently seeking evidence and asking for questions shows flaws in your argument.

On the other hand; pronoucign guilt without seeking appropriate evidence, and prematually claiming records are "*blatantly bogus" and created a "*false warming*", this is proof of the strengh of your case.

Leaving off the adjustment upward which I don't doubt has a legit basis - why does the blue line in the Eschenbach figure look like post-1941 the trend is pretty much a wash whereas in that same figure, the red line post-1941 is so markedly on the upward bound, even if you start the line at 1941.

Maple,

The first point in summary is that Darwin appears to be not representative of other stations in terms of the number of adjustments. (Hat tip [Nick Stokes](http://scienceblogs.com/deltoid/2009/12/willis_eschenbach_caught_lying…)).

2nd This is not some NOAA GISS conspiracy as the BOM come to similar calibration adjustments.

3rd Darwinâs warming is consistent with Halls Creek (0.13 C/decade) ([Mark Nettle](http://scienceblogs.com/deltoid/2009/12/willis_eschenbach_caught_lying…))

4th BOM and NOAA publish their adjustment protocol, and Willis has not demonstrated that this is different to what has occurred with Darwin, not that this is inappropriate.

5th, BOM and GISS provide the raw data. ([Michael Hauber]( http://scienceblogs.com/deltoid/2009/12/willis_eschenbach_caught_lying…))

Iâm sure others can add more points.

I demand to see actual electrons

Interesting post from Blair at LP:

Roger posted a link to my post from a couple of weeks ago so I wonât repeat the information there. As it happens, weâre currently reworking the Australian historical temperature data set, using the more complex adjustment scheme outlined in that post rather than the single annual adjustment used at present, and also incorporating a fair bit of pre-1960 data that was effectively unavailable for use last time round because it was only available on paper and hadnât been entered into the computer database. Hopefully that will be done earlyish next year.

In the specific case of Darwin, while we havenât done the updated analysis yet, I am expecting to find that the required adjustment between the PO and the airport is greater in the dry season than the wet season, and greater on cool nights than on warm ones. The reason for this is fairly simple â in the dry season Darwin is in more or less permanent southeasterlies, and because the old PO was on the end of a peninsula southeasterlies pass over water (which on dry-season nights is warmer than land) before reaching the site. This is fairly obvious from even a cursory glance at the data â the record low at the airport is 10.4, at the PO 13.4.

Darwin is quite a difficult record to work with. There were 12 months of overlap between the PO and the airport, but the observing site at the PO deteriorated quite badly in what turned out to be its last few years because of tree growth overhanging the instruments. Fortunately, we recently uncovered some previously undiscovered data from 1935-42 from the old Darwin Airport (at Parap) and should be able to use this to bypass the promblematic last few years at the PO.

The post-1941 adjustments (all small) at Darwin Airport relate to a number of site moves within the airport boundary. These days itâs on the opposite side to the terminal, not too far from the Stuart Highway.

As an aside, while the âurban heat islandâ is a real issue, the biggest influence on temperature records within urban areas is the land surface in the immediate vicinity of the observation site. The current Perth site is on a golf course and on cold winter nights is often one of the colder sites in the region; on the other hand, even the tiniest hamlet can have a large âurbanâ signal if the instruments are too close to a building or a bitumen car park. Also, a stable urban area will have little effect on a long-term trend; thereâs been some work done which shows that, while London and Vienna are both significantly warmer than their surrounding countryside, there is no evidence that the differential has increased over the last 100 years.

Knowing as much as you can about the sites youâre working with is definitely important. The historic documentation is obviously critical for that but knowing them at close hand is useful too. I have a long-term ambition of visiting all 112 locations in the data set we use for long-term monitoring and have 24 to go. Perhaps I was a bit too enthusiastic about this because I managed to drown my last car in a washout in the process of trying to get out of a remote weather station in the north Kimberley

http://ow.ly/KQWg

Darwin also copped a fair bit of bombing by the Japs during the Second World War - that might have had some effect upon priorities. Therefore all of the Darwin data is fraudulent! Every last bit of it. It's a leftwing Commie conspiracy, I tell ya!

BOM (see how the Bureau of Meteorology has an acronym sounding very similar to "bomb"; they are playing a joke on us the scumbuckets) are in cahoots with the UN since 1920 to change all data so that there is a global warming trend (the UN was a secret organisation for much of the 20th century). That's why we must privatise all data collection agencies. It is too dangerous to trust guvmint fraudsters with our data. Our TAXES paid for it, guddamn it!!! Our opposition leader Tony Abbott and his front bench will tell you the same. And since Tony Abbott has read at least the first couple of chapters from Plimer's tome, he is an intellectual authority on climate change, unlike the Prime Minister who has read the IPCC flawed science, spoken with guvmint lying scientists (Commies, boo hiss) and who has talked with Tim Flummery. And it's cooled since whenever Bob Carter says it has - 1998, 2000, 2002, and "a few years ago". Isn't Carter great?

Oh, and "NAZI!"

Dano wrote: > I'm a little late, but Willis and I have gone at it numerous times.

Why is that a but?

Dano wrote: > Comedy gold.

Not really. This hair is comedy gold.

Janet Akerman (comment 72), you got the wrong Steve Mosher. That's Steven W. Mosher, a Catholic polyglot, who's also president of the Population Research Institute. The Steve Mosher you see commenting here and on other climate blogs is an engineer who worships at the shrine of Tyrone Slothrop.

You're wrong also about overpopulation. The truth is that the population of the entire world could fit shoulder-to-shoulder in a space about the size of Jacksonville, Florida. The truth is that 97 percent of the earth's land surface is empty. The truth is that if you allotted to each person 1,250 square feet, all the people in the entire world could fit comfortably into the state of Texas. The truth is that the problems blamed on overpopulation are almost invariably the result of bad economic policies -- such as, for example, the great Ethiopian famine, which was created and exacerbated by that leftist Ethiopian government confiscating the food stocks of traders and farmers and exporting them to buy arms.

Tom,shut up and go and stand with all those people in Florida.

Nexus 6,thank you for the details from Blair Trewin. His comments about tree growth confirm what I suspected of the old Darwin PO site after seeing the 1890 photo of an early Stevenson screen close to a large driveway and near a young Poinciana that would clearly have spread over the box within a few decades. The microclimate differential between the airport and the PO should have been obvious to Eschenbach,but hell who needs details?

Tom,

You are full of it, man. REALLY full.

First of all, the 1980s famine in Ethiopia had nix to do with a "leftist government confiscating food stocks"... it had everything to do with the World Bank and IMF (read: US Treasury) foisting structural adjustment policies on the country, forcing them to grow cash crops (e.g. cotton) in lieu of food crops to pay back the interest sccrued on the "loans" given to the country. This is hardly new; it has been repeated many times in underdevloped countries before and since.

Second, you have no idea whatsoever about the relationship between population and consumption. Your rant was out of a comic book in terms of facts, in other words about as deep as a puddle. It does not matter whatsover that you could cram the entire world's population into the state of Florida. What really matters is the *per capita* ecological impact of given populations. Mathis Wackernagel and William Rees have produced a number of interesting studies exploring the human ecological footprint showing that every country in thge developed world maintains an ecological deficit - in other words, our countries depend on resources originating outside of our own borders to maintain affluence as currently defined. This may explain why economic policies generated by countries in the north (i.e. the "G-8") have usually been based on expansionism and plunder from the south. There are volumes of evidence to back this up - methinks you ought to read some declassified planning documents, which pretty well lay out the whole, ugly story in detail. You also ought to read quotes originating from the likes of Kennan, Kissinger and Brezinski to name a few influential planners and politicians - they expose the absurdity of your last post.

I will not waste any more of my time on your pedantics or grade school level tosh.

Post # 200, hmmmm I have made the same observation, will re-quote again as I believe this phrases the problem well.

"Why is the adjusted almost linearly diverging from the raw up until about 1985? What kind of station anomaly manifests itself in a linearly changing fashion over a period of 40 years?"

Much of the critique here focuses on the failure of Willis to account for an obvious station event at 1940. It is obvious and noted by himself. If you want to contrast raw with the altered data then you show those two series.

I had one answer which suggested that trees would be an example of such a cause for making that adjustment, growing over a few decades. Logically it can make some sense but when would they ever try to account for that?

You know, I think it is funny that so many in the denialist camp have such apparently serious doubts about the ways that climate data have been gathered over the years when they habitually ignore all of the biological indicators which reveal clearly that rapid warming is occurring in many temeprate regions. If we look at mean early season growth rates and flowering times in plants, the distribution and migratory patterns of many insects and birds, egg-laying dates in birds, phenological interactions etc., then the picture becomes quite abundantly clear.

Many volumes of peer-reviewed ecological and environmental journals are literally full of papers in which these processes have been investigated. Here in Holland, we are recording generally thermophilic insects and other arthropods that are coming from the south for which records before 2000 are scant; breeding cycles of migratory birds that have been monitored since the 1960s began changing dramatically in the 1980s, in line with changes in recorded temperature. These obsewrvations are being repeated with other organisms in other temperate parts of Asia and North America.

Yet, as Dano said, we have armchair experts here suggesting that they can find glitches in raw temeprature data that ignore other mitigating factors such as the urabn heat island effect. So why do these people not spend some time perusing the empirical biological literature instead of wasting time on pedantics? Warming is a fact. It is happening. And the view of most scientists, inmcluding myself, is that there is very strong evidence that humans are the primary culprit.

The key lesson out of this is that - if something a professional meteorological/climate body has done isn't immediately obvious to denialists - it clearly must be fraud! They just know better and more about the data and how to handle it than these bodies and all the so-called "experts", "scientists" and "statisticians" they hire!

Heck - just make wild assumptions, jump to conclu... no, make that hurtle headlong to conclusions, slander their professional integrity, and make liberal use of those "fraud" and "bogus" words!

Come to think of it - I don't quite understand how it is possible to derive real solutions from complex roots of second-order linear ODEs. But now I know - these "mathematicians" are engaging in fraud, and their "mathematics" is clearly bogus...

Shamelessly stolen from a comment at Larvatus Prodeo:

âHighly trained scientists wrong about science, claims unqualified bloke in streetâ

@Tim: you can let this post on Eschenbach follow by one on Richard Keen:
http://noconsensus.wordpress.com/2009/12/08/alaska-bodged-too/
At least Eschenbach can claim some level of stupidity as excuse, Richard Keen actually works as a lecturer at the Department of Atmospheric and Oceanic Sciences of the University of Colorado.

Zoot, spot on! This comment is accurate.

I recall a few years ago writing into a well known denialist blog where I was drwn into a debate on the possible costs of losing biodiversity. I did my BSc and PhD in zoology and population ecology, and I had worked in the field at the time for a few years, when a high school student (or he said he was that) entered the debate claiming that biodiversity was not really all that important, using the most wafer-thin and ignorant arguments that he could muster. Of course it was easy for me to dismantle his arguments one by one, but the majority of contributors to the blog, also what I would refer to as anti-environmental denialists, rallied to the student's cause, claiming that his arguments made good sense and that I was the one who was wrong. I recall one person suggesting that the planet could sustain a population of one trillion people - pure lunacy, but most of those responding agreed with him.

In the end, it wasn't the empirical evidence that mattered bu the ideological bias underlying the debate. Since the web site was a contrarian one and because the arguments of my so-called opponent, however gumbified they were, provided support for the ideological biases that most of those contributing on the thread shared, then I was the one who was out of his depth and not the high school student. This explains why veritable laypeople write in here and elsewhere condemning scientists who have spent years reesearching their field of expertise.

The other reason is that it is easy for one to stick their finger to the wind and claim to have a grasp of the underlying science. As that now-famous study published a few years ago found, the less that subjects knew about a field the more they claimed they knew about it and *vice-versa*. As a population ecologist, I realize exactly what my strengths and weaknesses are in my own area of expertise, and with respect to climate science I defer to those who are trained in that field of endeavor. But this deference does not appear to apply to many of those including non-scientists and journalists who are often the most vocal in attacking the broad scientific consensus on climate change. Whilst claiming that scientists are politicizing science it is, in fact, these people who are generally promoting their own political agendas in rebuffing or distorting the science underpinning the current warming (e.g. far right and/or libertarian).

By the way, a new petition has just been posted in the UK that shows overwhelming support amongst university academics in support of the IPCC position. I am not a fan of petitions but if this process was repeated in every country throughout the world there would be hundreds of thousands of names on it.

Forget insects; I thought the sceptics love satellite readings (they haven't yet caught on that using the MSUs requires knowledge of the radiation physics that they deny, as well as all sorts of corrections and adjustments). I'm surprised nobody over there has dug up UAH or RSS data for that region.

Hey, carrot eater, let's not forget insects. Without these critters our planet's ecological life support systems would inevitably collapse. The loss of pollinators, seed dispersers, decomposers, predators and parasites would be catastrophic.

carrot eater: "I thought the sceptics love satellite readings... I'm surprised nobody over there has dug up UAH or RSS data for that region."

Well, first they have to get those loud, widely-distributed accusations of fraud and cover-up out into the media. Because clearly the records of satellite readings covering 1900-1950 were so incriminating that they had to be destroyed

Because clearly the records of satellite readings covering 1900-1950 were so incriminating that they had to be destroyed

you hot that completely wrong. "hide the decline" was an attempt to erase the RAW satellite data from the MDW period.

the "trick" was to tell people, that there weren t any satellites available at that time.

Huh? I heard the Chinese had a satellite up in 1000 AD, and it clearly showed that the Arctic was ice-free at the time. The EPA and UN used secret lasers to destroy it, though.

And where is Watts's epic Surface Stations publication? Surely he would have got it out in time for Copenhagen?

I'm guessing that Watt's is realising that he has no real story and that he is either looking for a way to create one, or he is simply stalling and hoping to sink the hook of the UHI meme further into the mouths of the Ignorati before letting his 'project' taper away.

Or perhaps he has a semblance of a story, but is afraid that his 'paper' will be as mercilessly deconstructed as he imagines that he and his guest posters are currently doing to the orthodox science.

Whatever the outcome, Watts' impact is bound to be anticlimactical.

One thing that strikes me in all of this is the absolute certainty that Watts, and indeed any and all of his Denialati compadres, have absolutely no understanding of the statistical properties of large and complex datasets.

In my PhD work I collected repeated (often for a score or more occasions) measurements of around a dozen physical characteristics of thousands of individual animals. I am meticulous in my field recordings, and I spent weeks at the end of my fieldwork double checking my transcriptions to datasheets, and thence to computer databases, and although my error rates were exceedingly low, I still found some errors that were individually quite ridiculous in magnitude.

Even so, correction of all of the errors in my databases made no significant difference to my final results. If any of the denialists here dispute that this can be so, I suggest that they enrol themselves in the nearest beginners statistics class and learn why.

Climatological records have a similar scrutiny for error, and they contain orders of magnitudes more datapoints than I have ever worked with. Of course there are bound to be errors, but they are highly unlikely to occure at any greater level than I found in my own work, and they will have just as little impact on the global climate record as my own recording errors did in my own work.

If the error rate was any greater than mine, the errors would have been found and publicised years before now, either by an astute denialist (yeah, I know, an oxymoron), or more likely by the scientists who analysed the data in the first place, and who would have sufficient pride in their work that they would not countenance leaving obvious errors in it.

The denialists do not have a case. As so many have pointed out, the biological and geophysical evidence is irrefutable. The direct climatic data is robust. The physics has no been rebutted. Even the denialist ideology is not logically tenable.

Of course, if the denialists disagree with me, they have only to provide the evidence that supports their case, and that I have [repeatedly](http://scienceblogs.com/deltoid/2009/12/firedoglake_book_salon_on_jame…) [requested](http://scienceblogs.com/deltoid/2009/11/new_zealand_climate_science_co…) over the last several days.

[Jeff Harvey](http://scienceblogs.com/deltoid/2009/12/willis_eschenbach_caught_lying…).

I recall one person suggesting that the planet could sustain a population of one trillion people - pure lunacy, but most of those responding agreed with him.

Two former Australian premiers, Bob Carr (NSW) and Steve Bracks (Victoria), discussed what might constitute a [sustainable population for Australia](http://www.abc.net.au/lateline/content/2009/s2768391.htm) on Lateline last night.

It was quite interesting to see actually - Carr is an intellectual, and well-versed in the use of fact and of logic, whilst Bracks is a much more market-based ideologue. Bracks obviously comes from the Chicago school of economic fantasy, and just as obviously has no idea about thermodynamics, biological equilibria, trophic cascades or resource debts.

At several points in the debate (and they both come from the liberal side of politics, oddly enough) Carr confronted Bracks with the question of limits to growth, and of carrying capacity, and Bracks' cognitive dissonance was peeking out as his grasp of the concept of biological sustainability slipped as much as if he has been clutching at a bar of wet soap.

Growthers are the economic equivalent of lynch mobs, and their attachment to their ideology is as rabid. It is one of the many reasons why the recent global finacial crisis will not be the last kick up their collective arses - their ideologies are not sustainable under the dispassionate gaze of time, and of the inexorable certainties of the externalities that they try so vehemently to ignore.

Bernard: I'm curious to see if Watts can work out how to do a spatial average.

If he ever comes out with anything, I'd guess he'd also skip the homogenisation step altogether, similar to what you see Willis struggling with. We'll see.

Yet, as Dano said, we have armchair experts here suggesting that they can find glitches in raw temeprature data

People like Eschenbach are what I call stage one denialists, i.e. they are still trying to deny that long term global warming is happening. I could generalize stage one denialism to denial of any kind of observed data, e.g. denying that humans generate far more CO2 than volcanoes do.

Stage two denialism is where there is denial of something not as definitive as an observation, e.g. denying that CO2 probably produces a 3 deg C/doubling greenhouse effect.

I'd say it could be useful to consider whether someone is a stage one or other denialist because stage one denialists are so obviously out of touch with reality to everyone but themselves.

Just to add another 1000 year hockey stick to the collection, and completely independent of any BoM, have a look at page 19 of [this document](http://arnoldia.arboretum.harvard.edu/pdf/articles/1893.pdf).

There's even a "MWP" for those who don't believe any record without one.

Benard, re #219. Excellent post. I am going through the same issues with my latest data right now. Found a couple of seemingly significant errors the other day, but the data set is so huge that correcting the errors had virtually no impact on the results, certainly not even remotely large enough to change my conclusions.

Anyhow the modus operandi at CA (and WUWT) is to find any error, regardless of its significance. They did it with Steig et al, they did it with Kaufman et al, they have done it with others (tried with Briffa but that ended up backfiring). Anyhow, their singular goal is to find any error in huge and complex data sets and complex methodologies (only in "warmer" papers of course), and then, advertise loudly that the scientists is crap and results meaningless. Actually, when Kaufaman et al. correct the errors pointed out to them is strengthened their conclusions. Corrections published in Nature by Steig et al. did not affect their conclusions.

At least at CA they found legitimate errors, WUWT that even get that part right. On a side note, I think these guys (Watts, McIntyre, McKitrick and others) have huge egos that need feeding and will do anything to be the centre of attention.

The irony is, every time one of these denialists trying to refute the data, they only increase my confidence in the data and science and scientists whom they are trying to slander. Ok, maybe not the raw data so much, but certainly the latter two points.

Recommended reading.

http://www.economist.com/blogs/democracyinamerica/2009/12/trust_scienti…

I took Willis's word that there wasn't another station nearby in the 40s, but here is one at Wyndham Port going back quite some ways.

Everyone on this thread, including Tim (if I may be so bold), need to read the article provided by Carrot at #225.

I hate to say it, but Mr. Eschenbach has indeed been caught lying.

Read the article and weep (on his behalf).

Why does almost everyone on both sides of the climate debate buy into the assumption that global warming would be bad? The examples of past warming episodes â medieval warm period, warm climate during the height of the Roman empire, Vikings first colonizing then leaving Greenland â suggest as strongly as any historical evidence ever suggests anything that warm periods in the earthâs climate are good for man and human civilizations. The predicted effects of AGW â tropics the same, poles warmer â suggest good effects from global warming. We could start with more arable land for growing food and move on to more inhabitable territory in newly ice-free areas of Greenland and Antarctica. There would probably be more mineral resources available for exploitation as the glaciers retreat, so global poverty would be alleviated. Even arctic animals might benefit -- just how many fewer would freeze to death each year if the arctic warmed up? You can't just assume that warmer temps wouldn't be a net benefit for them too -- certainly extreme cold has always been their traditional enemy. Indeed, wanting a warmer climate at the poles might be the true "green" position, considering how inhospitable polar temperatures are to life on earth (think of what a biological desert Antarctica is!)

By the way, if you could show beyond a reasonable doubt that global warming was seriouslly bad for us, then it would have to be fixed **whether or not** man had caused it. In fact the reasons it was occurring would be sort of beside the point unless it suggested a quick way to fix it â and last I looked Copenhagen-supporting climate scientists were saying that decreasing the industrial output of CO2 is a very slow way of reducing global warming, and we might even have passed the "tipping point" so that reducing CO2 would not have the desired effect. Climate scientists that are pushing this sort of fix show that they are not really all that worried about global warming as such, but rather more concerned about returning the earth to a "more green" past (while retaining their positions of influence, of course -- who do you think would be in charge of checking up that CO2 output was decreasing as intended?) Personally, if I were really worried about global warming, I would start thinking about developing serious climate-altering technologies. You could, for example, build lots of orbiting sunshades to deflect incoming solar radiation â itâs sure to work. If instead another ice age starts to look probable (weâre more or less due for the next one to start according to some experts), you could make the orbiting sunshades reflective and divert toward the earth sunshine that would otherwise miss the planet, which is also sure to work. If you really want to assume responsibility for the earthâs climate, this sort of sure-to-work technology is what you should be thinking about. It doesn't have to be space-based, it just has to be obviously adequate to the task at hand, obviously adequate to almost everybody and not just a small group of experts.

Ugh. Nobody's assuming anything; that was one long argument based on not much of anything.

First read WG2 of the IPCC, then discuss with a more firm understanding of what the impacts may or may not be.

[IPCC WG2](http://www.ipcc.ch/publications_and_data/publications_ipcc_fourth_asses…)

I really dig that Economist post, especially the last 3 paragraphs. The individual is engaging in true skepticism, which is the opposite of the behavior of those who declare themselves "global warming skeptics". I also like the irony of October 2009 being the warmest October on record for Darwin.

I agree, MarkB. It was careful, it covered more ground than we did here, and best of all, it recognised the author's lack of authority in statistics.

Who said it recently - everybody knows they don't understand general relativity, but somehow everybody thinks they understand climate science?

More examples of prevarication with Climategate (McIntyre again). Not strictly on topic - so maybe you want a start a new thread ...

http://deepclimate.org/2009/12/11/mcintyre-provides-fodder-for-skeptics/

This so-called Climategate is really getting out of hand, isn't it?

Steve McIntyre has published allegations - twice now - that an internal IPCC authors' debate about the inclusion of Briffa's tree-ring reconstruction in a key figure from the 2001 WG1 Third Assesement Report was driven by concern about the post-1960 "decline" in tree-ring widths, a decline that showed a marked divergence with the instrumental tempertaure record. McIntyre even claims that lead author Michael Mann worried that showing the series with this decline would give "fodder" to "skeptics".

But even a cursory examination of the emails in question shows that the discussion was really about other aspects of the reconstruction, specifically obvious discrepancies between Briffa's reconstruction and the other two under consideration over the major part of the reconstruction's length. Thus, once again, McIntyre's speculations are shown to be utterly without foundation. Even worse, McIntyre left out intervening sentences within the actual proffered quotes in what appears to be an unsophisticated attempt to mislead.

Bureaucratically minded people usually respond with an "Ugh" to thinking outside the box -- that's one of the things that makes them bureaucrats in spirit (no matter what they do for a living). It's very easy to come up with a list of drawbacks to any sort of change -- and bureaucrats almost always hate change, too, unless they think it will give them more power. Why don't you re-read the IPCC report and try to think of possible benefits they may have left out? I'd be surprised if they even seriously considered the plusses of an ice-free Antarctica -- imagine, a whole new continent, available for life, to do with as we pleased... What's not to like? I'll bet even bureaucracies of all types would end up larger and more powerful with all that extra land and a larger global economy to support them.

No way, SteveM feeding the denialists with misinformation? How can that be, he would never....oh hang, on he does this all the time.

This takes spin to a whole new level.

Please tell me they are going to have him cross-examined at the CRU investigation.....he was/is after all, one of those using the FOI to harass scientists.

As long as we're discussing denialism, the IPCC was not written by bureaucrats.

Quite a bit of the land mass below Antarctica would be under water if the WAIS and EAIS were to melt, see http://nsidc.org/data/atlas/news/bedrock_elevation.html

You also forget that an exposed land would be bare rock. No soil. Just a small problem, and add to that the fact that the weather would still be too cold to grow crops. You would also be waiting a very long time for all that ice on the EAIS to melt, if ever.

I could go on, but I hope you see the obvious flaws in your thinking.

'Ugh' is not a response to thinking outside the box.

'Ugh' is a response to somebody speaking from ignorance: somebody who thinks their idle musing is somehow worth listening to.

I'm sure there will be some positives to warming. I don't in the least assume there won't be. I also give you some reading material in which the issue is analysed. If you want to seriously discuss the matter, then read the sources, back up your points, and discuss it in the open thread. Not here; it's off topic.

The quasi-arbitrary data adjustments in Darwin are looking sticky.

Data should be sacrosanct, but someone thought a little fudging in their favor wouldn't do any harm. Let's not obscure the subtle nature of this fraud, the most dangerous lies are those closest to the truth.

Stop the propaganda now, free the data, free the code!

The quasi-arbitrary data adjustments in Darwin are looking sticky.

The totally arbitrary theft of internal emails, and the obvious lies being spread about what those emails really say, are looking slimy. So are the death threats at least one climate scientist is getting from denilaist assholes like el gordo & co.

Data should be sacrosanct...

...says the hypocrite who has no problem with stealing and misrepresenting data.

quasi-arbitrary

Justify that description, please.

Gordo "The quasi-arbitrary data adjustments in Darwin are looking sticky"

I'm sure that you mean those made by Eschenbach of course.

DC @#231. Excellent and essential reconstruction! Mann allegedly worried about feeding skeptics? Given McIntyre's mendacious editing of these emails exchanges alone, the worry is entirely justified.

Alaw'dd: "Why does almost everyone on both sides of the climate debate buy into the assumption that global warming would be bad?"

All our cities, forests, and farmlands have grown on climatic assumptions. When the climate changes, they will need to move. If a little warming happened over more natural periods (say, hundreds or thousands of years) this would be feasible. But since it's happening at speeds we haven't seen in geological epochs, neither we nor nature can move quickly enough (or adapt in other ways) without paying a high price. We in terms of money and lives, nature in terms of mass extinctions.

Consider that the reason why no one bothers with your question might be that your questions is really dumb (and has been considered already).

Oh, ignore el gordo. He clearly hasn't a clue what he's talking about (at least Eschenbach spent a couple seconds reading about the homogenisation procedures, I doubt el gordo has done that much); he's just trying to get a rise out of you lot.

el gordo has been in my killfile for forever.

dhogaza:

It's amazing, isn't it. McIntyre's always just called a "geologist." then again, he's way less educated in that than Plimer, but the latter is way more ignorant by his writing at least.

I wonder - and I really do wonder this - if Dr. Curry realizes exactly whom it is she's legitimizing, and on whose behalf she's being so condemnatory of RC and Cru and so forth?

I mean, he's not even got a Masters ... in Science!

FYI - The Australian Temperature record has been adjusted twice in the past.

In 1996 by Torok

http://134.178.63.141/amm/docs/1996/torok.pdf

In 2004 by Della-Marta

http://www.giub.unibe.ch/~dmarta/publications.dir/Della-Marta2004.pdf

The interesting aspect of both these papers is the Start date for measurements and a quick comparison with the current BoM data shows a lot has been dropped.

Della Marta also categorized the stations by reliability - rating them 1 - 5. They then discarded the 4 and 5s.

I might add that the Temperature anomaly used in this post is deceptive when you put it up against the monthly mean temperature over the same period.

http://www.bom.gov.au/cgi-bin/climate/hqsites/site_data.cgi?variable=me…

try this instead

http://tinyurl.com/ye57gxn

woops - wrong Tinyurl

this one

http://tinyurl.com/ydatxev

I'll believe step functions for site change adjustments when they are properly justified by reasonable comparison experimental data. This wouldn't be all that difficult to do.

I don't believe continuous upwards adjustments over decades based on dubious statistical comparisons with "nearby" stations hundreds of kilometers away which may have entirely different micro and meso climate issues.

Willis has been quite open about what he's done in his analysis yet you accuse him of lying and your basis for this is arm waving obfuscation.

If you want to compare what the effect of different thermometer exposures is just do the experiment.

Once you move the thermometers more than a few meters you have a new station. No continuous data.

Mike Borgelt @250: have you made your beliefs known to the BOM??

I wonder - and I really do wonder this - if Dr. Curry realizes exactly whom it is she's legitimizing, and on whose behalf she's being so condemnatory of RC and Cru and so forth?

I don't get her, honestly. The "can't we all get along" meme she suggests is refuted by her own treatment by those she tries to get along with.

Willis has been quite open about what he's done in his analysis yet you accuse him of lying and your basis for this is arm waving obfuscation.

Well, he started by saying "I don't understand" and based on this said "smoking gun of climate science fraud".

That will get him accused of lying every time (not about the "I don't understand" part, but the "my ignorance proves climate science fraud" part).

Is this really hard to understand?

Mike @250:

"Willis has been quite open about what he's done in his analysis yet you accuse him of lying and your basis for this is arm waving obfuscation."

Read the economics magazine article. Willis has made several false statements and assumptions, and this is not the first time he has fudged the data to satisfy his preconceived ideas. Yet, you continue to blindly defend him.

Jeez, I mean to say "economist magazine", it's late here. See post #225.

The thing with trolling such as robotech master above does is - it's not JUST trolling, it's highly abusive trolling. And it's crude divisiveness trolling on top of that. Even if an abusive troll praises some of the non-trolls and abuses others, we're still going to take everyone's statements on their merits. And we're going to killfile. And we're not going to bite on the hook.

By the way, and it's easy to forget this: An associate of Watts did go through the data they'd gathered after more than a year of harrassing weather stations. Just as everyone involved in the station data told them, his analysis showed that the surface station data wasn't distorted at all, and that Watts' claim was wrong. Then Watts clammed up for a while, and no one publicized the fact that they'd been shown internally that they were wrong, and they suppressed the facts. After enough cooling off time had gone by, Watts said his study needed more time.

It'd be interesting if I followed every climate denialist around (me and my droogies) and snapped photos of them all day and noted who they talked to, and so on. On the theory that wherever they went the number of peer-reviewed papers in climate science would decline slightly, because of their harassment, and weather forecasts would be late and inaccurate. And I did that for years. and I announced I was going to release results that would blow the lid off the climate denialists. and I was writing a book. And then I had some associates from the science blogs analyze my data. And then I hid it cuz the result was not what I wanted. I would be intrigued as to how Watts, McIntyre, Eschenbach would characterize all this.

You may well object that Watts downplayed it, because if he had hid it, how would I know? But I am using his standards, not mine. :)

I have calculated the bias of adjustment for the *entire* CRU dataset. You find the result here.

some people are doing a real analysis. like gg, who took a excellent look at ALL CRU adjustments.

or the [economist blog post](http://www.economist.com/blogs/democracyinamerica/2009/12/trust_scienti…) mentioned above, which completely takes apart this piece of unscientific work.

but this is the normal operation modus of denialists. just throw some rubbish into the public arena. then change to a new topic, before all errors have been reported. and repeat.

we had the new zealand bogus first. they ignored adjustments for thermometer moves. the resullts were rubbish.

eschenbach was the next. a completely stupid approach on the raw data of a single thermometer. what was he trying to show?

and via Lucia we got an "analysis" of raw Alaska data. a really short look showed that the adjusted temperature is in perfect agreement with UAH data, and the unadjusted data is not, because all thermometers are in the same part of Alaska.

i fear we will see more of this. stay tuned..

I strongly recommend looking at gg's post 258. It seems to be the conclusive refutation of the Eschenbach nonsense,

el fatso wrote:

Data should be sacrosanct...

For someone purporting to be interested in the science, deliciously ironic that he should choose a term with such strong religious overtones.

260 Nick,

Ditto. gg has done an excellent job, and he has made his code available.

Another ditto of Nick's comments, with the added value that a quick peek over there reveals more straw-grasping by a couple of denialists posting comments.

If the difference between the "raw data" and the "corrected" data doesn't make you look twice...then you must be crazy. Even by eye..if you look at the temperature record after the "adjustment year" [1941 I believe], there is obviously no warming trend. Why on earth would there be a warming trend created after adjustment?? At best, the unadulterated raw data shows no change in temperature at all!

If raw data is useless...then massaged data is even worse. Better to take nothing from bad data. Even worse is to create a problem from data that is discontinuous and cannot be trusted.

Gary B: If you'd spend three more seconds learning about the issue, in addition to the three seconds you spent looking at the graph, you'll see there are all manner of reasons why an adjustment can be necessary.

I'll add my voice to the chorus; gg's method of analysis is quite nice.

Gary B: self-taught Creation Climatologist!

Anyone here with any t-shirt graphics skillz so we can make a shirt out of Gary's...um..."logic"?

Best,

D

"I have calculated the bias of adjustment for the entire CRU dataset. You find the result [here](http://www.gilestro.tk/2009/lots-of-smoke-hardly-any-gun-do-climatologi…)."

You find my debunking of the result [here](http://i50.tinypic.com/16m51qc.jpg).

My detailed (second) comment is awaiting moderation there. He's using an optical illusion instead of stating his real results that he thankfully includes as a data file: The +/- adjustments both round to a magnitude of 0.1. But positive adjustments outnumber negative ones by 56% to 44%. The overall average adjustment is +0.017, which taken as a degrees/years slope (I assume?), amounts to 1.7°C warming per century due to positive adjustment bias.

268 Nik,

gg says

Not surprisingly, the distribution of adjustment is a normal distribution with peak pretty much around 0 (0 is the median adjustment and 0.017 C/decade is the average adjustment â the planet warming trend in the last century has been of about 0.2 C/decade).

What's your point? It would be surprising if the average adjustment were _exactly_ zero. The point is that the adjustment is very small compared with the trend.

The overall average adjustment is +0.017, which taken as a degrees/years slope (I assume?), amounts to 1.7°C warming per century due to positive adjustment bias.

0.017 per decade is 0.17 per century. There are tricks for multiplying by ten, you know, though I suggest you not use them now that denialists are anchored to the position that all tricks are fraudulent.

And that's the land temps, not the average global average temps...

Basil Copeland at WUWT has now found more 'evidence" of data being fudged. This time using ONE example from Nashville. Do these guys ever give up? Wonder how they "stumbled" on this site? Funny how they never stumble on a single one of those thousands of sites where temps have been adjusted downwards...

Anyhow, I'm too busy to figure out what they did wrong this time (not that I did much refuting the last time). Anyone else game?

Is an adjustment bias of .017 C/decade for n = 6,533 cases, statistically significant in a trend which is an order of magnitude larger?

Possibly, but if it is significant, is the bias valid or invalid? E.g. There is a bias in human controlled landscapes to plant ornamental trees and permit them to grow.

Hence there are potentially sound reasons for such bias. i.e. the bias for new stations having cleared land, and older station having maturing vegetation.

Regardless, GG has put any potential bias into perspective, and at worst any bias is an order of magnitude smaller than the warming trend.

i fear we will see more of this

I hope so. Stage one denialism ("there is no global warming") confers on its proponents a massive destruction of credibility.

Re #258: Kudos gg!

I got linked over here from the Economist, but got there from a link on WUWT. I try to read the other side now and then to see what their arguments are. I must say you guys have some facsinationg ways of rationalizing things and otherwise seem very willing to swallow illogical arguments.

Regardless of how strongly you want to believe (and/or do believe) in AGW, the bottom line is that NO temperature reading should be arbitrarily adjusted. I don't care how much you blather on about "homogenization" and how we must leave the statistics to the scientists, any and all temperature adjustments should have a clear and complete explanation based on a correlation exercise. In case you don't know what that means, that would be a direct comparison between one gauge at the same location and/or additional guages at any new location. If it is a new location in the case of temperature readings, that would mean within 50 miles absolute max (I don't care if Willis was wrong and there was a site within 272 KM. I pay enough attention to temperature to know 272 KM away can be vastly different when you're trying to resolve down to 0.1 C.), and would require a new gauge verifying the old site along with an aditional one at the new site. You obtain readings concurrent in time between the two or more gauges and then perform a Gage R&R analysis to make sure you have no bias. You might perform some other statistical test, but you don't make adjustments to the readings without having some valid reason to do so that is based on data taken at the existing site you claim is wrong.

If it happens that your measurements are in the past, you reproduce the above correlation exercise the best you can. You at least place temperature sensors in all of the previous locations and take readings concurrent with the new or current location. While you're at it you make sure you're taking readings from locations that are not affected by UHI, even if that means taking readings at two or three locations around a previous site that is now affected by UHI or otherwise does not meet criteria such as CRN1 in the U.S. That may mean taking readings at a few dozen sites if necessary to verify an adjustment, expecially one as large as that done at Darwin.

For those of you grabbing for the keyboard to begin typing a reply to tell me how brilliant and advanced some statistical temp reading homogenization routine is, instead tell me about some product sold on the market where the buyer will accept adjustments based on homogenization from readings elsewhere. How about the thickness of brake pads sold by 'ACME' to 'World Motors' for their new car? Do you suppose that World Motors would accept those brake pads based on data (on pad thickness) that they were told had been adjusted upwards based on some statistical routine??!! Or do you suppose they would tell ACME to measure them again with a gauge they had confirmed was accurate? Think of any other similar scenario (millions of possibilities at least) where a customer is buying a product based on the performance of that product. This would never fly. Yet you warmists want the world to buy your product for trillions of dollars.

I also note it's convenient how you left out Willis' graph showing the stepwise nature of the upward adjustments in temperature at Darwin. And of course you neglect to notice that there are no specifics in the eplanations given as to exactly when and why adjustments were applied. Don't you think you should have a specific reason for exactly why an adjustment was made and what the magnitude of the adjustment was that coincided with that reason, or are all of you a bunch of lemmings?

Have any of you inquired about that "old invalid" thermometer and done some research to confirm it was bogus? Have you attempted to confirm when the change occurred, and what sort of correlation was done between the old and the new? Or are you mostly all the type who sit around and buy things from infomercials based purely on the testimonials?

I also note a distinct lack of discussion of the tarmac and jet activity around the existing sensor location. Have none of you even boarded or deplaned on the tarmac anywhere? Asphalt? Jet wash? You can't seriously believe it's valid to adjust temperatures at a rural location vs. an airport (between runways/taxiways no less) downward.

Finally, the sensor at the old post office was in a Stevenson screen. See the pic at Warwick Hughes site.

Do you suppose that World Motors would accept those brake pads based on data (on pad thickness) that they were told had been adjusted upwards based on some statistical routine??!!

Well, they accept accelerated stress testing of components adjusted upwards to reflect longetivity under real-life operating conditions using statistical routines.

I suggest you stop driving and even more importantly, stop flying ...

Some information on accelerated stress testing.

Look at the end:

Residuals Plots: Plots of the residual values that have been assigned, via regression analysis, to each point in a data set. These plots provide a tool to assess the adequacy of the model (distribution and life-stress relationship) used to analyze the data set.

Regression analysis! Statistics? Whoa. Model! Whoa.FRAUD!

Do you people have any idea of how idiotic you really look? You're really expecting climate science to abandon tools and techniques that are pervasive throughout science and engineering?

Stress testing is usually required to estimate longevity, but longevity is not something that can be definitively measured at test time. Temperature is. Try giving an example of what I asked for instead of using a diversion that actually does nothing to support your argument.

I've been involved in writing software to take measurements, including pre and post stress. I've also been involved in many meetings discussing how various measurements should be taken and what sort of correlation exercises are necessary to verify the data is good. And I've written routines to perform Gage R&R analysis on data. My point here being your "Whoa" line is quite childish.

If you want to go down your oranges line any further to compare to my apples line, please quote the internationally recognized (e.g. ISO) document used that all the Darwin adjustments were performed according to. Or perhaps even a customer specification agreed on between the Darwin adjusters and their customers. Probably don't have that do you? Point here being stress testing is performed according to some specific procedure that is fully and openly documented between provider and customer. That way the customer knows exactly what he is getting.

275 Chris,

Wow. So many words. So little content. Do you somehow imagine that anyone here knows less than you do?

Oh, and "Willis" is not a plural noun; it's someone's name.

It appears that many if not most on here know less than I do, actually. If you aren't even asking questions about the when and why of the adjustments that's not going to change either.

Tha fact you focus in on a typo doesn't speak well to your knowledge either.

How about addressing my arguments?

OK Chris, I'll bite. Let me give you some examples:
1. Pharmaceutical companies (and regulators) determining the efficacy of treatments have to use sophisticated statistical models that homogenise data to determine outcomes;
2. All large companies use sophisticated statistical models to examine their financial future, all of which weight the inputs based on homogenised data;
3. Hydraulic engineers use homogenised models that weight factors to plan changes to river regimes (especially for example the affects of dam construction)
4. Mining companies use similar models to predict subsidence associated with different mining patterns;
5. In Europe at least, the regulators use weighted models to determine whether the exposure of the companies under their jurisdiction is appropriate;
6. Indeed, reinsurance companies themselves use black box catastrophe models, using homogenised data, to look at their risks, which run to trillions of dollars.

In every case "the buyer will accept adjustments based on homogenization from readings elsewhere". Your lack of understanding of this is very telling.

Chris<

I posted this in a somewhat different context, on a different blog - but it seems appropriate here.

[Open Mind](http://tamino.wordpress.com/2008/12/31/stupid-is-as-stupid-does/#commen…)

--

Lee // January 22, 2009 at 5:04 pm

Raven,

My dad was a planning engineer for the California Water Resources Department. A task he did repeatedly over the course of his career was streamflow analysis of a defined watershed, to determine water availability under varying precipitation amount/timing scenarios for uses such as reservoir storage, fisheries maintenance, etc. His work supported fisheries enhancement and restoration work, water storage and allocation decisions, and the last two major reservoirs the State of California built in the early 60âs. Stream flow projections based in part on his work are part of why California water planners know we have a serious water problem right now.

His data came from streamflow gauges â basically these measure how deep the water is and how fast it is flowing at defined stations in the stream. Stream surface speed was usually determined by measuring it a very few particular gauge heights, usually just 2 or 3, and then calibrating height to surface speed of the water flow. That data, and a measurement of the stream cross-section at that station, all fed into a mathematical calculation of the stream flow given that cross section and surface flow. California maintains a bunch of these across the state. The data were ALWAYS flawed.

For example â nonlinear stream-bed cross section and flow speed issues at non-calibrated gauge heights. Loss and replacement of gauge, with discontinuities in the
record at times of replacement. Gauge height being exceeded, with resulting truncation of high values at peak flows. Missing data for days, months or years when a gauge was lost or not being maintained or measured. Infill of surrounding man-made structures, or substantial erosion alteration of stream bed, altering the streamflow and therefore altering calibration â often with no re-calibration to compensate for it. Float gauges getting stuck, giving strings of âfalseâ data. On and on.

Almost always with no metadata to say âthese data have this problem on these dates.â

And yet, dad managed to use those flawed data, with checking for internal data inconsistencies (check â the climate people do this), cross validation with nearby streamflow results (check, the climate people do this), comparison with nearby unaltered gauges to check for calibration âcreepâ, (check, the climate people do this), and so on.

And somehow, using this often heavily flawed data, data it seems you would simply throw out and not use at all, data that does not meet your rigorous definition of engineering data â his streamflow analyses were pretty damn close, and very, very useful for making predictions of available water given a precipitation amount/timing input.

He didnât go take pictures of all the stations to prove they were flawed, by the way. He sometimes took pictures to validate a request for budget to replace or repair a station.

And he received commendations for his engineering work, based on his use of flawed data and using data-based corrections that most often had no real-world validation (no picture of a station alteration, no meta record of a stuck gauge, no recalibration to detect calibration creep, etc).

THAT is how real engineers use real-world data, Raven, your fantasies notwithstanding.

Chris is one funny person:

If it is a new location in the case of temperature readings, that would mean within 50 miles absolute max

The stupid burns. 50 miles is pulled out of your fucking ass.

Go to Salt Lake City, and 10 or 15 miles east will cause a discontinuity that explains why Snowbird offers Alpine skiing, while SLC offers temple tours.

the bottom line is that NO temperature reading should be arbitrarily adjusted.

And, fuckhead, they're not "arbitrarily" adjusted.

Not only are you a stupid fuck, but you're an outright lying one.

I don't care if Willis was wrong and there was a site within 272 KM. I pay enough attention to temperature to know 272 KM away can be vastly different when you're trying to resolve down to 0.1 C.

And, of course, the converse can be true. For most days of most years, Seattle and Portland have almost identical weather.

Fuckhead.

(sorry for my rudeness, the arrogant "I know more about everything than everybody, especially climate scientists" is particularly irritating this evening.

It appears that many if not most on here know less than I do, actually.

The brave words of the amateur presenting the latest perpetual motion machine to ignorant physicists ...

Sorry, it appears that your an idiot with an ego inversely proportional to your knowledge.

Thanks dhogza, you probably said to Chris what many is us would dare not say :) Chris seems to be another example of Dunning-Kruger, amongst other things.

Dhogaza, while you are all fired up and inspired, what do you think of Basil Copeland's "analysis" over at WUWT? They have found another "smoking gun", nay "smoldering gun" this time, Nashville is the cherry-picked example. Still waiting for them to discuss a station that should not have been adjusted downwards, but was.

I see Anthony W. is still moving Eschenbach's story to the top of his page, just to show how very important E's "seminal work" is in this area ;)

Lee, #282, your dad deserves a medal. He must have spent years working on that project and probably sprouted some gray hairs in the process!

Lee and GWB, infinitely better responses than the previous two.

GWB, drug efficacy is still really the apples and oranges thing as it can't be specifically measured with precision. (It would be nice if it could though.) None of the rest of your examples involve concrete measurements either. They are all estimates or forecasts and there exists no gauge to measure them other than an estimate.

Whether based on models or otherwsie I'll also bet they are all very well documented. Each distinct estimate or forecast will have a report, prospectus, or similar that spells out all factors, assumptions, calculations, etc. that were the basis for the forecast. Your lack of seeing my point on documentation is telling, as is your failure to distinguish between issues than cannot be measured with precision.

Lee, in your example of course stream flow at a given point can be finitely measured. Ask your dad if his data would have been significantly better if he had the resources to calibrate the gauges in a manner similar to what I suggest above whenever there was an issue. Also ask him for details. If your dad was as thorough as you say I'll bet he fully documented his rationale for adjusting readings, infilling data, etc.

I never said I would throw out data, by the way. I stated that the data should be correlated. With temperature readings it's pretty easy, so why not? Since your post is the best by far, IMO, what are your thoughts on a temperature sensor in between acres of asphalt with frequent jet exhaust vs. at an apparently rural or small town post office? What about the stepwise adjustments shown in the original story from WUWT that's being attacked here? Care to speculate?

If someone can show me true justification for the adjustments at Darwin, and the many similar stations, I'll be more than happy to join the club and agree everything is hunky dory with temperature adjustments in the GHCN and related datasets. If you can't do that with specific documentation for each station then at the very least admit that it is logical to question why the newer Darwin readings, for example, are being adjusted upward.

Just when there was a meaningful reply, the rabid dog jumps back in.........

Doesn't take much to push your buttons does it? Does this blog have no etiquette rules?

If someone can show me true justification for the adjustments at Darwin

Have you read the post Chris? Followed the links?

Chris:
"If someone can show me true justification for the adjustments at Darwin, and the many similar stations, I'll be more than happy to join the club"

Do your effing homework. The justifications are there - the BOM has published theirs. GISS has published their adjustment procedures in their various publications, linked on their web site, AND have released their entire code.

The fact that you have time to attack this, but not enough to education yourself - google is your friend - tells me a lot.

BTW, the daat my father used was HISTORICAL. Just like the temp data is. He couldn't go back and fix it, he couldn't calibrate readings from 50 years ago. So he corrected it, adjusted it, using the surrounding and corroborating data available to him. Just like GISS et al have done. His documentation, in many cases, was described IN THE FREAKING ALGORITHM. Or in the description of his procedure. Just like GISS has done.

Please stop pretending that your ignorance of the published and available documentation and resources means that somehow they are being hidden from you or don't exist. Educate yourself, and do your own damn homework. None of this stuff is hidden from you - you just have to have a minimum of competence, and look for it.

yeah that's great chris, your typology of measurements and feelings about precision are both simply fascinating, you must tell us more when you get the chance, but you know what else we can measure? we can measure the reliability and validity of our measurements. It's called statistics, and it is the language of science. Admonishing people, some of whom are in the past, to "measure with precision", whatever that's supposed to mean, is just casting spells, mate, please stop your gibberish.

Chris, perplexed by the frustration shown here? Yes? Though so. Scientists are getting fed up with every tom dick and harry lecturing them as to how they got it all wrong, and that you joe soap, can teach them just what they need to know.

I don't give a hoot what your credentials are, or how much you think you know about this issue-- it is like me thinking that I can start mouthing off at oncologists as to how they should do their job because I have some fancy letters behind my name. That would be ludicrous, naive, omniscient and arrogant of me, and I would not dare even contemplate doing so. This is no different.

You do not even begin to understand the basic problems associated with using heterogeneous data (in space and time), problems that have been worked on by many scientists for decades now, and they are continuing to refine and improve the process. It takes teams of scientists years to prepare these data. In fact, you need to be a bit of a masochist to undertake this kind of work.

The number of times dhogaza et al. have been told by some interlurker like you that "I demand the code, the data" and "I am going to show this all a farce", is just beyond calculation. They have had their buttons pushed for years now, and it is all getting rather silly, so they are tired of being nice and explaining patiently to every new interlurker why they are wrong. Do you understand now?

So instead of lecturing others on etiquette, first do your damn research, and show some humility. Then people might be more receptive. I hope this unpleasant experience does not throw you to the "dark side". Read the papers and documentation and then think about the problem some more. Go to the AMS website (ametsoc.org) and use their search engine, or Google scholar.

And the only reason I am being so polite to you is b/c I'm Canadian and we are, for the most part, pretty cordial :o)

DETREND THIS

(lead story here)

Then go here, and in the drop down menu click "All" to see how the temps have changed since 1948 at Darwin, ...NOT!

The pirates of CRU have you hoodwinked.

Chris:

Does this blog have no etiquette rules?

If you want a polite response then leave out the attitude.

Chris (#288) said "GWB, drug efficacy is still really the apples and oranges thing as it can't be specifically measured with precision. (It would be nice if it could though.) None of the rest of your examples involve concrete measurements either. They are all estimates or forecasts and there exists no gauge to measure them other than an estimate."

Absolute tosh. Let's take just one example that I know quite a lot about - mining subsidence. This can be (and is) measured on the ground, but to do so is very expensive. Unfortunately when subsidence is measured the dataset is surprisingly noisy because any given point is affected by a range of other processes in addition to the mining - movements on slopes, heave of the soil, etc. So, just as with climate science, the data are homogenised using statistical models to get a net effect.

Of course you could argue that this should not be done - i.e. that data are in some way sacred and should be preserbed unaltered. To do so means that the data are of essentially no use. You could also argue that more, better measurements are needed, but in most cases this is just not practicable. For measurements extending back in time (the mine that I work with has data that extends to 1972) this is not possible (you seem oblivious to the fact that it is impossible to go back in time to get better temperature data).

You appear to have a very idealistic view of data - that in some way that data are pure and that if collected properly data give you the answer without need for interpretation. This is almost never the case, either in the field or in the lab. From even the best experiments data are always complex and noisy, requiring very detailed analysis. This is often a shock to new research students, who believe that the experiment that they have designed will mean that the answer just falls out. The reality is that the answer is sometimes there, but hidden in noise that requires work to deal with. Of course sometimes the answer is not there at all, either because the experiment was poorly designed or because the hypothesis was misguided.

In a nutshell, cutting edge science is not like measuring brake pads on a production line, and never has been. Please stop showing your ignorance by claiming that it is.

Gee-dubya wrote:

You appear to have a very idealistic view of data - that in some way that data are pure and that if collected properly data give you the answer without need for interpretation. This is almost never the case, either in the field or in the lab.

Second-year bright engineering student, knows more than you do until he gits some more of th' learnin' then he's a-humble agin.

-------

And if I can expand on Lee's point in 282, we do that all the time in botanical sampling. It is done in wildlife management. Just about every scientific discipline wants more sampling. But resources are limited, thus the statistics an' stuff.

And if I may, the Russians use a totally different sampling method for plant censusing than the West does. It is not randomized but dependent upon the sampler's knowledge. And they have no problem gathering data.

Best,

D

Best,

D

So, we can rid ourselves of the evil sampling and associated stats by measuring everything. Great, we just need to cover the surface of the earth with thermometers.

Will the denialists suggest this?

Does this blog have no etiquette rules?

If I were to set them, I'd start by banning posts that demonstrate willful ignorance combined with an arrogant presumption of superiority over thousands of working scientists.

In my mind, that combination is far, far more rude than the word "fuck".

"I must say you guys have some facsinationg ways of rationalizing things and otherwise seem very willing to swallow illogical arguments."

Do you really expect gentle treatment when you lead with this?

Chris
>If someone can show me true justification for the adjustments at Darwin...

In your haste to spit up your ignorant screed, you must have missed [this](http://scienceblogs.com/deltoid/2009/12/willis_eschenbach_caught_lying…). Understandable.

Written by a physicist:

What isn't mentioned in this article is that the data can be "homogenized" to give a downward as well as an upward trend - depending on what is used as a baseline. This is not spin - just simple statistics. "Homogenization" of data is patently unscientific (this is not allowed in hard science). Just because the airport increased in temperature - does not mean we should automatically adjust every other inhomogeneous data set to do the same. This is guessing - not science.

Truthfully, such poor non-homogeneous data should be thrown out - or included in such a large sampling of data that both upward and downward inhomegenaities statistically cancel. The size of this larger sample would be determined by the amount of error in the calculations. No "homogenization" allowed - to easily biased.

Using the exact same argument as above - the data could be homogenized to the cooler, not the hotter temp. The auther is merely showing their bias and hoodwinking all your non-statasticians.

Willis Eschenbach's commentary remains more scientifically correct (no data manipulation). Although, as previously noted, a larger and homogeneous data set is needed to know what is really happening over a larger area with any certainty. Truthfully, we don't have anything here more than poor measurements - and possibly some wishful thinking - by BOTH political camps.

Chill out - all you children are far too excited.

sparrowfart,

Admit it. You haven't read this comment string, have you?

sparrowfahrenheit,

Please support this statement with evidence:

>*Homogenization" of data is patently unscientific (this is not allowed in hard science).*

No actual refutations of what has been shown to be wrong with Eschenbach's post then, Mr "Physicist"? Just think you can come in, throw some assertions around and expect your "I'm an (anonymous) physicist, therefore I have final say on what is and isn't science" line to impress everyone into submission.

Interesting that despite amusingly incompetent attempts to assert yourself as some kind of objective arbiter learning us all about science and statistics and other scary things, you can make a statement such as:

"Using the exact same argument as above - the data could be homogenized to the cooler, not the hotter temp. The auther is merely showing their bias and hoodwinking all your non-statasticians."

Now, I may not be a statistician, but even I know that that statement is meaningless. First, because speaking strictly regarding site movement across the single Darwin Airport station, it doesn't make a blind bit of difference to the trend whether you "homogenise to the cooler, not the hotter temp". Second, because there are not merely two arbitrary temperature readings to "homogenise" too. And thirdly, because if you check the [raw mean temperature data for Darwin](http://data.giss.nasa.gov/cgi-bin/gistemp/gistemp_station.py?id=5019412…) and compare it with the [annual mean temp](http://www.bom.gov.au/cgi-bin/climate/hqsites/site_data.cgi?variable=me…), you'll see that the most radical adjustment was indeed to "the cooler, not the hotter temp". Again, not that it makes a difference to the trend.

I'm not an expert, and neither, despite your pretences, are you. The difference between us is that a) I'm upfront about my lack of expertise and b) I have enough humility to be willing to learn, and to be wrong. Also, I bother to read people's posts.

I suggest you follow that example. I'd suggest it's the grown-up thing to do.

If you can kindly ignore the multitude of typos above, but it's late where I am...

sparrow, bud, et al.

Guys, have you BOTHERED to look and see what difference all this makes to trends calculated using the the raw vs adjusted data?

Try here:

[raw v adjusted](http://www.ncdc.noaa.gov/img/climate/research/ushcn/mean2.5X3.5_pg.gif)

That graph shows, for the US, the annual temps calculated using the raw data, and then overlain on it, the results of each successive adjustment to the data.

The total adjustment for the US causes an increase in trend over the 20th century of about 0.25 C / Century. Of that, about 0.2C is due entirely to correction for time of observation and or transitioning to MMTS sensors. These are necessary, well documented adjustments due to known changes in observing technique, and are not part of the homogenization itself.

The remaining 0.05C adjustment is from the homogenization - which is designed to statistically handle things like station moves and urban heat island encroachment. This means that the entire homogenization procedure applied to the entire US, leads to an increase in the trend of ~ 0.05C / century. Removing the homogenization step entirely, would cause a decrease in the trend for the US of 0.05C per freaking century. And you guys are up ini arms about how terribly misleading and unscientific and fraudulent this is? Get freaking real.

[USHCN1](http://www.ncdc.noaa.gov/oa/climate/research/ushcn/ushcn.html)

What isn't mentioned in this article is that the data can be "homogenized" to give a downward as well as an upward trend - depending on what is used as a baseline. This is not spin - just simple statistics. "Homogenization" of data is patently unscientific

Baseline, of course, isn't relevant for trend calculations.

Regarding homogenization being invalid ...

Tell that to doctors who homogenize ass-measured temps to oral-measured temps, and explain to them that the fact that ass-temps are lower than oral temps disproves everything medicine knows about historical records, because, you know, any adjustment is "fraudulent" blah blah.

What isn't mentioned in this article is that the data can be "homogenized" to give a downward as well as an upward trend - depending on what is used as a baseline.

And, of course, the poster doesn't know what homogenization means ...

Wow ... the illiterate overturning the work not only of thousands of climate scientists, but *every* scientist, and engineer, too.

God, it's a miracle his PC works ...

If anybody wants to see it, George Sherrington has found the historical metadata for Darwin. Keep in mind that the GHCN method does not use this information, but it is such changes that the method is trying to detect, statistically.

It is in alladj here.

ftp://ftp2.bom.gov.au/anon/home/bmrc/perm/climate/temperature/annual/

@ Lee:

It's possible that my typos and general English last night was even worse than I'd originally alluded to, but I'm fairly sure if you'd read my comment at #305 again (as well as various other comments I've made on this topic) you'd find that I was arguing against sparrow, and most emphatically not calling fraud of any kind.

Again, my point was that sparrow's:

"Using the exact same argument as above - the data could be homogenized to the cooler, not the hotter temp."

Basically demonstrates a lack of knowledge of what this whole subject has been about. For the reasons I mentioned. Take out the data pre-1940, and the homogenisation makes very little difference to the trend. Nick Stokes @#57 showed this better than I did, using the corresponding GISS data.

Thanks for giving the overall picture, but a little slower on the trigger next time please. :)

@ bud.

My apologies. It wasn't your typos, but my sleep deprivation at the time.

-Lee

Lee, no worries. Cheers.

Not only is his bogus 500 kms not in the original paper, but when he says there is no station for that period within 500 kms, he's wrong: according to the Excel file from the Aussie Climate data centre, the reference climate station at the Darwin Post Office -- basically right next to the airport -- covers the data from 1885-1941.

Hard to imagine how he could have missed that, since DARWIN AIRPORT and DARWIN POST OFFICE are right next to each other in the Excel file.

Michael Turton
The View from Taiwan

I posted this on the Monckton thread in connection with Monckton claiming that there has been no warming for 15 years, but it is perhaps more relevant here.

Monckton may have sourced this comment from something posted at The Air Vent on 12 November 2009. Jeff Id says at one point "so letâs have a little fun with the team". You can read the full post at:

http://noconsensus.wordpress.com/2009/11/12/no-warming-for-fifteen-year…

So it appears that Monckton's comment was made in reference to some stats analysis made by Jeff Id who concluded that "three of four measurements show no significant global warming for the last 15 years and came very close to clearing the 17 year mark."

The denialists were/are over the moon with this analysis. Unfortunately for them, Tamino has soundly scuttled their "brilliant analysis", today in fact, what great timing:
http://tamino.wordpress.com/2009/12/15/how-long/#more-2124

Tamino concludes that "That does not mean that thereâs been no warming trend in those 15 years â or in the last 10, or 9, or 8, or 7, or 6 years, or three and a half days. It only means that the trend cannot be established with statistical signficance. Of course, itâs another common denialist theme that âthereâs been no warming.â This too is a foolâs argument; any such claims are only statements about the noise, not about the trend. Itâs the trend that matters, and is cause for great concern, and thereâs no evidence at all that the trend has reversed, or even slowed."

The denialists really do know how to massage and distort data.

Why homogenize the data at all?
See the following at the link http://strata-sphere.com/blog/index.php/archives/11824

When a temperature station is moved it should simply become a new station at that point in time, with a new set of siting errors (and accuracy if the sensor is upgraded). It has a different time window than it previous incarnation â it is a new data set. You donât âhomogenizeâ neighboring stations into a mythical (and fictional) virtual station.

Basically what alarmists needed to do was not adjust data, they needed to create a thermal atmosphere model which would take into account siting characteristics both local and large. This would include distance from large bodies of water, altitude, latitude, etc. A three dimensional model that would explain why various stations have their unique siting profiles and temperature records. It would explain why temperatures near oceans fluctuate less than stations inland 100-200 miles. It would show how a global average increase of 1°C would result in a .6°C increase at high latitudes or altitudes. It would EXPLAIN the data variations in the measurements.

"But we donât have this model. Alarmists cannot explain with accuracy why stations 10 miles apart show different temperature profiles each and every day of the year. So they pretend to know how to âadjustâ the data..."

It would explain why temperatures near oceans fluctuate less than stations inland 100-200 miles

Yeah, it's really a scandal that scientists don't understand why this is true. It's also a pity that they don't know why this is more true on the western coasts of continents in the northern hemisphere. It's also a pity they haven't given these regions a catchy name, like "Maritime Climates".

Oh, wait ...

Here is how merging two stations 12 km apart and 62 metres different in elevation can distort things.

http://members.westnet.com.au/rippersc/hchomog.jpg

If you check fig 8 of the pdf http://www.giub.unibe.ch/~dmarta/publications.dir/Della-Marta2004.pdf, this should have set off alarm bells.

edward:

Basically what alarmists needed to do was not adjust data, they needed to create a thermal atmosphere model

So a science denialist thinks there should be a model?

Oh, the irony.

319 Chris,

Surely Edward is a Poe (although his punctuation makes it hard to tell what he's saying and what he's quoting)?

I urge people to read Eschenbach's actual posts, and decide for themselves who is actually "lying" here, Eschenbach or Lambert. This is a good place to start:

http://wattsupwiththat.com/2009/12/20/darwin-zero-before-and-after/

Uh, I'm sure most of us have.

again, in the comments to that article, Eschenbach admits, that he does not understand the data he is working with.

No, the âdayâ is midnite to midnite, so you will only get one high or low per day. The low is typically shortly after dawn, and the high somewhere in the late afternoon.

he does not know how a min-max thermometer works. he has not understood the time of observation bias. but he insists in using the raw data. pretty insane.

and the comments are even worse. they don t understand, that positive and negative adjustments nearly balance out over the [TREND](http://www.gilestro.tk/2009/lots-of-smoke-hardly-any-gun-do-climatologi…)

the discussion is worth a look at anyway. special thanks to Nick stokes and JJ for bringing some sense to WUWT...

The method used to correct the UHI is in error, yet we keep pushing predictions for a future that can't possibly be valid except by chance.

With so many uncertainties about global warming, it is remarkable that the IPCC is so imperious.

El Gordo - have you ever thought about getting a life? Rather than posting lies and insinuations on a blog you could be out there having fun.

With a never=ending supply of bullshit, it is no surprise el gordo is so imperious.

So who will be the reliable source now?

Lance, were you not around when El Gordo posted a link to Denial Depot, not realizing that it was a parody? That might give you some indication of the reliability of El Gordo's posts.

Wrong Brucy, I first found the parody at Eli R and dropped it here with little comment. It caught a few off guard and for that I apologise, especially to Jeff.

El Gordo, I don't doubt that you found the parody at Eli's. I'm just not sure you understood that it was a parody.

You are, after all, one of the people here who repeatedly posts weather reports, as though snow in London in the middle of winter says something meaningful about global warming.

Then again, maybe you only do that as a way of highlighting the absurdity of denialists. Maybe it's not that you don't recognize parody when you see it. Maybe I don't recognize parody. You've been spouting silliness for months, and here I was, reading your posts as though you actually meant what you were writing.

Eschenbach admits below that NASA GISS are down-playing climate change and that temperatures are rising!

Oh, and for what itâs worth, care to know the way that GISS deals with this problem? Well, they only use the Darwin data after 1963, a fine way of neatly avoiding the question ⦠and also a fine way to throw away all of the inconveniently colder data prior to 1941

It seems that Eschenbach got his graph in a twist!

I spotted this while foolishly trying and failing to get a Denidiot to see that Eschenbach was lying.

Thanks to Tim for highlighting and unravelling this web of deceit.

Bruce

Eli said it was parody and I believed him. My only crime was failure to raise the 'parody' flag, but as my credibility here is close to zero it doesn't really matter.

We are moving back into a cool PDO and so we can expect more winter snow in Europe for the next 20 years.

>*but as my credibility here is close to zero it doesn't really matter.*

A level of credibility that el gordo has done [much to earn](http://scienceblogs.com/deltoid/2009/12/russian_analysis_confirms_20th…)!

Close to zero?? You flatter yourself. Try non-existent.

Eschenbach may have got it wrong, but we won't resolve it here and now with a definitive answer. Anyway, the real action is elsewhere.

In the mid 1990's there were some in the environment movement who wanted to de-emphasize the link between AGW and extreme weather because it left them vulnerable.

But the movement failed to take their own advice and all extreme weather events were offered as further proof of global warming.

This New Year's Eve a return of blizzards to North East US and the UK are predicted, which everyone will agree has nothing to do with global warming, but everything to do with climate change. Or will I be told that it's weather and not climate?

335 El Gordo,

nothing to do with global warming, but everything to do with climate change

What gibberish is that?

Anyway, it's weather, not climate. It will still be weather if the UK is covered in 10 feet of snow for a month.

>...but as my credibility here is close to zero it doesn't really matter.

Au contraire, Gordito, your credibility is a large negative number, quite far from zero.

Sub zero? The mini ice age commeth.

>*Sub zero? The mini ice age commeth.*

Thanks for re-emphasizing [the point](http://scienceblogs.com/deltoid/2009/12/willis_eschenbach_caught_lying…).

Joe Bastardi just made this forecast. Should we take it seriously?

The major population centers of the northern hemisphere are facing a cold spell not seen in a quarter century. 'The first 15 days of the opening of the New Year will be the coldest, population weighted, north of 30 north world wide in over 25 years in my opinion'.

A 15 day weather forcast? Keep cherry picking el gordo!

[This is](http://www.woodfortrees.org/plot/gistemp/plot/gistemp/mean:180/plot/gis…) what 'we' should take seriously.

What if this winter in the UK is the coldest since 1962-63, is that weather or climate?

>*What if this winter in the UK is the coldest since 1962-63, is that weather or climate?*

What if one season in one country is the coldest since 1962? I'll call your puny regional weather, and puny 1962 record and raise you with last month, the globally hottest November on record! A record set during an extended solar minima!

That el gordo, is cherry picking weather. This is the [global climate trend!](http://www.woodfortrees.org/plot/gistemp/plot/gistemp/mean:180/plot/gis…)

The NAO and AO tend (like the PDO) to be predominantly positive for some decades and negative during others.

The NAO and AO went negative at the time of the Copenhagen Conference and may remain in this mode for some time. The PDO is also fully cool.

These oscillations determine climate, not CO2.

>*These oscillations determine climate, not CO2.*

Now all you need it evidence to match and [surpass this](http://ipcc-wg1.ucar.edu/wg1/wg1-report.html). Oh and the evidence that [keeps mounting](http://www.skepticalscience.com/What-happened-to-the-evidence-for-man-m…)!

Here at 39º41'3.85" North, 2º50'48.60" East and 142m over the sea, we are having a rather soft end of december: 19 °C, come clouds, wind SW 26km/h, humidity 70%

Palma?

342 el gordo,

Weather, of course. Why do you keep repeating similar questions?

"Palma?"
Very near, about 22km in the island of Majorca. Should have been colder, but somehow siberian air is been contained in northern latitudes of Europe.

¿[Binissalem](http://ca.wikipedia.org/wiki/Denominaci%C3%B3_d%27Origen_Binissalem), no?

346, 347,

That's where I make it.

Let me get this straight, gordo.

2009 started with a la nina, and one of the colder winters in recent years. We're in a negative PDO, have gone from neutral to negative NAO and AO, are a couple years into a major extended solar minimum. We are ending the year with an el nino, but it isn't very strong. And yet, despite all these conditions trending to colder temperatures and the only 'warming' condition being relatively weak, 2009 is going to end up being one of the few hottest years on record.

This is all weather, of course, but it is striking that even your very own cherry-picked weather contradicts the climate claims you are making, gordo.

"Oh look, here's the Australian Bureau of Meteorology's high quality climate data for Darwin aiport."

I looked and what did I see? According to the chart of the high quality data, Darwin Airport was founded within a year or so of the Wright Brothers first flight. Does this make Darwin the first airport in the world? How many other airports were built before 1910?

Even the AMB only rate Darwin data as "fair". Perhaps that is because it is now classified as "urban" whereas, unless the airport was built on top of existing buildings, the site was most unlikely to have been urban previously. Could this explain the significant discontinuities in 1930 and later in 1940?

How much of the obvious temperature increase post-1940 can be traced to increased air travel. Would it not be interesting to see aircraft movements at Darwin Aiport plotted against the temperature anomaly histograms? Or would that be considered a spurious correlation?

Nothing is straight forward, that's why I raised a question earlier about the hard winter of 1780. The sun was fully active and innocent, so what was the cause?

A strong La Nina was in place at the time and now we have a weak El Nino, so where's the link?

As a member of the general public it would be nice to get an informed response, instead of 'just the usual pedantic certainty'.

el gordo writes:

>*it would be nice to get an informed response, instead of 'just the usual pedantic certainty'.*

el gordo science provides evidence, you provide the [usual arrogant certainty](http://scienceblogs.com/deltoid/2009/12/willis_eschenbach_caught_lying…).

According to the chart of the high quality data, Darwin Airport was founded within a year or so of the Wright Brothers first flight.

No, Solomon Green, the chart indicates that there was a thermometer at or about that location.

It doesn't say anything about the history of powered flight.

What a pathetic note for you to end the year on.

Seems like the data manipulation goes on and on and on and on!

All manipulated up of course

www.twawki.com

In other news, black is white and hot snow falls up.

Twawki should be renamed twitti. His/her web site peddles the usual anti-scientific contrarian crap; it is no wonder that every 'article' has no replies to it. Go away, twawki.

I am sorry to rain on your parade but from what I can see it is the Australian BOM that has a credibility problem, not Willis. As long as the methods and justification used to make the changes are unavailable for review the temperature reconstructions will have to be considered invalid because science does not allow for a 'trust us, our data and methods are OK,' position. This should be an easy argument to settle. Make all data, metadata, and methods available for independent review and let the claims stand up to indepenent verification.

Vangel, the methods and justification are [here](http://www.ncdc.noaa.gov/oa/climate/ghcn-monthly/images/ghcn_temp_overv…) and [here](http://reg.bom.gov.au/amm/docs/2004/dellamarta.pdf). Eschenbach noted them, then ignored them. I can't be arsed to track the metadata down for you, but given that Eschenbach used it in his post that's a pretty big clue that it's publically available. That old line won't wash here.

Science may not ideally operate on trust alone (although I've yet to meet the layman who wants to peer-review his own coronary bypass operation in situ), but neither can it function under an atmosphere of poison and baseless allegations of fraud. If we had to discard every single old set of historical records because someone didn't want them to show what they showed and had a hissy fit over a missing notebook here or there then nothing would ever get done.

Eschenbach would have you believe that it is more likely that he has uncovered bogus data tampering than an outlier station who's homogenisation has a much greater warming affect than average. Why are you taking him on trust?

Vangel @360: What was the BOM's reply when you asked them for "the methods and justification used to make the changes"? How about when you requested all the "data, metadata, and methods"?
And what would you do with them if you got them?

Many on this thread have apparently not read what Eschenbach posted. It is especially troubling that some have been so irresponsible that they called him a liar based on a complete distortion of what he wrote. Even the starting premise of the discussion is questionable.

Anyone who spends any time reading blogs and internet posts must realize by now that you can't trust anything anyone writes, even if you desperately want to agree with it. Please, to avoid pointless conflict in the future, do your due diligence, or don't post.

CarlNC,

I read what Eschenbach posted. I read what you posted, You just presented empty opinion devoid of factual content.

to jakerman:

Thank you for your comment. You are correct, I provided no facts because I thought the facts were apparent, and in any case everyone should do their own research and reach their own conclusions. Question everything, even if you are disposed to believe it.

To expand on my first post:

There were several posts by W.E. about the Darwin records. It is necessary to read them all in order to fairly judge them.

The graph "GHCN Raw and Adj. Temperatures Darwin Airport" was presented by W.E. to show the degree of adjustments made. He did not maintain that the unadjusted data was correct, as Lambert implies, thus the premise of this discussion appears to be false. In his (W.E.'s) discussion, he agreed that there was a step change made and an adjustment required. He questioned the reason for all the other adjustments. Perhaps you missed that post, which as I recall was a follow-up. There were other statements made by Lambert that need a line by line fact checking, but I'll leave that to you.

Question everything, even if you are disposed to believe it.

See, this attitude of waltzing in pretending to be an independent thinker in a sea of sheep is going to do you no favours whatsoever here. Most people here have read the post in question, are well aware of what Eschenbach was trying to do and why he was wrong. If he has since corrected the errors in this post and apologised for the baseless accusation of fraud, feel free to link to it.

Lambert made no implication as to Eschenbach's belief about the unadjusted data. He accurately recounts Eschenbach's belief that the data was adjusted to create false warming. W.E. noted the reasons for homogenisation early on in the 'smoking gun' post, then promptly forgot about them. His suggested asjustment has no justification for it at all. He switched from absolute temps to anomolies half-way through his post so as to prevent any reasonable like-for-like comparison between data pre and post homogenisation. He offers no reasonable comparison with other stations available to him, suggesting strongly that he cherry-picked the station that would show the most radical upward adjustment (see Nick Stokes @57 to see effects on other stations).

And his conclusion after all this? Numbers were deliberately bumped up to show warming. This conclusion has absolutely no evidence whatsoever to support it. The post was an absolute joke.

CarlNC writes:

>*The graph "GHCN Raw and Adj. Temperatures Darwin Airport" was presented by W.E. to show the degree of adjustments made. He did not maintain that the unadjusted data was correct, as Lambert implies, thus the premise of this discussion appears to be false.*

Did Tim imply that Willis Eschenbach "maintain[ed] that the unadjusted data was correct" and was this the premise of Tim's post?

No, the premise was that Eschenbach claims that for Darwin "the trend has been artificially increased to give a false warming where the raw data shows cooling".

And that: Eschenbach, simply declares the NOAA's adjustments "blatantly bogus" that created a "false warming".*

Hence I do not accept the characterization of your criticism as you currently state it.

I never know whom to believe. So I will just continue trying to do what I can to help keep the earth from heating up.

>*Oh man. It's things like this that discredit the entire field of climate research.*

You mean Willis Eschenbach's erroneous allegations, pushed by Watts' denial machine discredit climate research?

Only in the eyes of those vulnerable to false propaganda. Unfortunately that number is not small.

It now appears to have been accepted that Willis was right. See, for example, http://diggingintheclay.wordpress.com/2011/02/28/willis-was-right-and-n…

Willis writes:

>*Those, dear friends, are the clumsy fingerprints of someone messing with the data Egyptian style ⦠they are indisputable evidence that the âhomogenizedâ data has been changed to fit someoneâs preconceptions about whether the earth is warming.*

Julian can you show us where this claim is justified?

This is associated with an attempt to claim the global warming is an artifact of bad homogenization. Yet global warming from GHCN v2 is from a -0.3 deg anomaly (1880) to a +0.7 deg anomaly at present. This is virtually the same warming as GHCN v3. You [have to go to 3 decimal places](http://rankexploits.com/musings/2010/ghcn-version-3-beta/) to find a difference over this 130 year period.

Willis like Watts can't see the wood for the trees. They are focusing on possible imperfections that don't significantly change the magnitude of warming.

And because the data does not fit their narrative they over reach can make unsupportable claims about a unjustified systematic bias that is where the data is supposedly:

>*changed to fit someoneâs preconceptions about whether the earth is warming.*

This narrative is unsupported and it is dishonest.

Most people buy Movies, because we would like to enjoy them all.

You may find this interesting,

Who is Willis Eschenbach?

http://www.populartechnology.net/2013/10/who-is-willis-eschenbach.html

As of 2012 Mr. Eschenbach has been employed as a House Carpenter.

He is not a "computer modeler", he is not an "engineer" and he is certainly not a "scientist" (despite all ridiculous claims to the contrary).

"A final question, one asked on Judith Curry's blog a year ago by a real scientist, Willis Eschenbach..."