Climate-gate

[milkweed]

"So, in answer to the question, the warming rates for all 4 periods are similar and not statistically significantly different from each other."

in discussion of a single dataset - nothing sensational there

So your claim of a single dataset is false.

Here is the question:
A - Do you agree that according to the global temperature record used by the IPCC, the rates of global warming from 1860-1880, 1910-1940 and 1975-1998 were identical?

"So, in answer to the question, the warming rates for all 4 periods are similar and not statistically significantly different from each other."

Now you are claiming the IPCC (the Holy Grail of all things AGW, the Keeper of the Gate, the UN round table of the Science of Climate Change) uses one dataset?

2 reasons:

1. watching the anti-climate change cranks deal with the data would be like expecting your hairdresser to diagnose multiple sclerosis - a complete waste of time, and not something that they would in fact have any interest at all in honestly dealing with.

2. there's plenty more available data where than came from that supports the trend

Interesting. I shall translate this into your position is the science is not as important as the politics.

as for #2, its circular.

If the homogenization/normalizing procedures have not been open to falsification and the "plenty more available data" is based on the above, the trend being supported is all based on data that has not been open to scientifically falsification.

That Phil Jones has behaved very badly is beyond question - that climate change science in general remains solid also remains beyond question.

Phil Jones is a piece of the whole behaving badly.

Me thinks your not a sceptical as you believe yourself to be.

 

[Trippy]

If the homogenization/normalizing procedures have not been open to falsification and the "plenty more available data" is based on the above, the trend being supported is all based on data that has not been open to scientifically falsification.

And yet, The Russians managed to write a paper complaining about the CRU's treatment of Russian climate data.

:/

Incidentally, the data that the russians mention (as being rejected) in that specific paper is also the noisiest data sets.

 

[milkweed]

And yet, The Russians managed to write a paper complaining about the CRU's treatment of Russian climate data.

:/

Incidentally, the data that the russians mention (as being rejected) in that specific paper is also the noisiest data sets.

Trippy, I am aware you are a beeleever in AGW.

Its all noisy data....

But that has nothing to do with the codes used to 'normalize' being beyond the reach of falsification does it?

Not available for the US data, not available for Asia, africa, s. am.......
'tis truely a global phenomenon!
Documented.
Verifiable.
but alas, still beyond the reach of falsifiable.

Unprecedented Indeed!

 

[milkweed]

The Russians[/url] managed to write a paper complaining about the CRU's treatment of Russian climate data.

:/

Incidentally, the data that the russians mention (as being rejected) in that specific paper is also the noisiest data sets.

Just curious.

Are you implying that Phil Jones is now working for the russians?



Because I cant figure out why you brought up the russian report (which I havent read)

 

[Trippy]

Trippy, I am aware you are a beeleever in AGW.

No, you only think I am, the truth of the matter (of my opinions) is apparently too complex for you grasp.

Its all noisy data....

No shit sherlock.
However, some data is noisier than others, and if you had bothered looking at the paper, you would have seen that the data sets that were rejected were consistently noisier than the data sets that were accepted.

But that has nothing to do with the codes used to 'normalize' being beyond the reach of falsification does it?

Proof please?
No, seriously.
You've missed the point entirely.
Let me repeat it for you:

The Russians reproduced the work of HADCRU.
The Russian complaint is that the data that HADCRU uses over emphasizes the trend.
The Russians contend that HADCRU consistently used the data that gave the most positive trend, and so obtained a bogus result.
The thing that I had noticed, reviewing the data presented in the Russian paper was that the data rejected by HADCRU was consistently the noisiest data.
Re-read what I said.

Incidentally, the data that the russians mention (as being rejected) in that specific paper is also the noisiest data sets.

I said noisiest, not noisey.
You've set up a strawman argument, then made a poor attempt at demolishing it.

Not available for the US data, not available for Asia, africa, s. am.......
'tis truely a global phenomenon!
Documented.
Verifiable.
but alas, still beyond the reach of falsifiable.

Unprecedented Indeed!

I don't know what to say here other than :LOL:

 

[Trippy]

Just curious.

Are you implying that Phil Jones is now working for the russians?



Because I cant figure out why you brought up the russian report (which I havent read)

No, I'm bringing up the Russian paper (which it's obvious you haven't even glanced at) because it directly contradicts your assertion that the HADCRU results are irreproducable. They can, and have been reproduced.

 

[milkweed]

No, you only think I am, the truth of the matter (of my opinions) is apparently too complex for you grasp.

You just cant help being venomous can you?

well... here we go.

Some other considerations when using these files
Gridding
The gridding process used in Brohan et al.. (2006) and earlier publications assigns each station to the 5 degree latitude/longitude box within which it is located. The gridding then simply averages all available station temperatures (as anomalies from 1961-90) within each grid box for each month from 1851. No account is taken of the station's elevation or location within the grid box (anomalies show little consistent dependence on altitude). A more up-to-date location for a station is not important for the gridding, unless a site change were to move the station to an adjacent grid box.

http://www.cru.uea.ac.uk/cru/data/landstations/

What are the basic raw data used?
Dataset Terminology
CRUTEM3 land air temperature anomalies on a 5° by 5° grid-box basis

Soooo No. The closest thing to RAW data CRU offers is a host of averaged temps.

I win
you lose.

 

[milkweed]

No, I'm bringing up the Russian paper (which it's obvious you haven't even glanced at) because it directly contradicts your assertion that the HADCRU results are irreproducable. They can, and have been reproduced.

I said falsifable.

I can reproduce the mistakes made if I am only given the flawed data.

 

[Trippy]

You just cant help being venomous can you?

Statement of fact.
My position is not one that is easily explained.
You have misunderstood my position.

QED.

No venom.

well... here we go.

Some other considerations when using these files
Gridding
The gridding process used in Brohan et al.. (2006) and earlier publications assigns each station to the 5 degree latitude/longitude box within which it is located. The gridding then simply averages all available station temperatures (as anomalies from 1961-90) within each grid box for each month from 1851. No account is taken of the station's elevation or location within the grid box (anomalies show little consistent dependence on altitude). A more up-to-date location for a station is not important for the gridding, unless a site change were to move the station to an adjacent grid box.

http://www.cru.uea.ac.uk/cru/data/landstations/

What are the basic raw data used?
Dataset Terminology
CRUTEM3 land air temperature anomalies on a 5° by 5° grid-box basis

Soooo No. The closest thing to RAW data CRU offers is a host of averaged temps.

I win
you lose.

Actually, no.
As an example, if you look at the russian paper that I have already linked to, you would see that they list the stations that HADCRU used, provide you with a link to the Russian equivalent of NOAA, so that you can go there and download the data for yourself.

Just because the raw data isn't explicitly or specifically listed in the paper, doesn't mean it isn't available.

Incidentaly, this actually provides a more robust measure of review/verifiability, because it means that rather than having to simply take their word that the data they present in their paper is accurate, I can go to NOAA, or wherever, download the data for myself, preform the calculations as detailed for myself, and I should get the same result.

 

[Trippy]

http://www.cru.uea.ac.uk/cru/data/temperature/#datdow

 

[milkweed]

No venom.

You did not link to a russian report, you linked to a blog.

I did find the report translation:

http://climateaudit.files.wordpress.com/2009/12/iea1.pdf

I think I know where you intend to go with this, and presume your focus will be the very last graph.

So to be sure there is no confusion about my closing, Here is a quote from the PDF.

QUOTE FROM REPORT Pg 21. para2.
"At the same time, this consideration is very conservative, since for the calculations of temperatures across Russia, all data contained in the RosHydroMet database was used – without any selection based on the data’s contents, as well as without any necessary corrections, for example, for the effect of cities’ temperatures."

So what is obvious to me because of the last graph, HadCRU applies NO UHI effect to their temperature releases in the latter half for sure.

Not sure why the russian data diverges so much pre-1940 and is so much warmer than HadCRU temp data. I could wonder if maybe.. maybe they do manipulate the data to show an increased warming in the latter half of the century.

Lies, damn lies, and statistics.

 

[Trippy]

You did not link to a russian report, you linked to a blog.

Which, has at the bottom of it, a link to the report (which of course you knew, having read the blog in its entirety).

I did find the report translation:

http://climateaudit.files.wordpress.com/2009/12/iea1.pdf

I think I know where you intend to go with this, and presume your focus will be the very last graph.

You like making assumptions don't you?

So to be sure there is no confusion about my closing, Here is a quote from the PDF.

QUOTE FROM REPORT Pg 21. para2.
"At the same time, this consideration is very conservative, since for the calculations of temperatures across Russia, all data contained in the RosHydroMet database was used – without any selection based on the data’s contents, as well as without any necessary corrections, for example, for the effect of cities’ temperatures."

So what is obvious to me because of the last graph, HadCRU applies NO UHI effect to their temperature releases in the latter half for sure.

Not sure why the russian data diverges so much pre-1940 and is so much warmer than HadCRU temp data. I could wonder if maybe.. maybe they do manipulate the data to show an increased warming in the latter half of the century.

Lies, damn lies, and statistics.

They're addressing the data set that they generated based on the data that they think that HCRU should have used, and they are discussing the difference between what HCRU actually used, and what they think HCRU should have used.

However, their calculations include data that is noisier than the data that HCRU used, and noisey data obscures trends - in other words their 'less pronounced positive trend' that they claim may simply be a reflection of the statistical noise they've added to their data sets.

The perfect example of which I've posted here: http://www.sciforums.com/showthread.php?t=99625

 

[Trippy]

Incidentally, were you aware that the IEA (demonstrably) have a vested interest in promulgating anti-AGW science?

The IEA portfolio is one of free trade.
Most (all?) of the current sloutions dealing with AGW involve some level of redistribution of wealth, which goes (generally) against the principles of free trade.

 

[milkweed]

Which, has at the bottom of it, a link to the report (which of course you knew, having read the blog in its entirety).

Apparently you didnt follow the link at the bottom. Might not have even read the whole blog being as it clearly states the link is in russian (which I dont read)

Here is the link in the blog to the russian report:
http://www.iea.ru/article/kioto_order/15.12.2009.pdf

Its IN FUCKING RUSSIAN!

I found the link elsewhere thankyouverymuch.

You like making assumptions don't you?

See above

They're addressing the data set that they generated based on the data that they think that HCRU should have used, and they are discussing the difference between what HCRU actually used, and what they think HCRU should have used.

Thats your interpretation of intent.

I see it showing when you use the entire available data, you see quite a divergence from what is presented by HadCRU as fact.

I also noted the tendency of HadCRU (assuming the russian report was accurate on these percentages) of them dumping longer records, gravitating to urban areas and leaving grids out of the data, even when rural stations exist.

I saw clear comparisons between urban (HadCru selected) and rural stations (hadCRU drops) which did show higher temps in urban, lower temps in rural, within the same grid area.

However, their calculations include data that is noisier than the data that HCRU used, and noisey data obscures trends - in other words their 'less pronounced positive trend' that they claim may simply be a reflection of the statistical noise they've added to their data sets.

Thats your allegation. Support it with evidence. I find it impossible to believe a rural station would be noisier than an urban station.

It must have been really noisy in the 1930s because thats some remarkable divergence RAW vs HadCRU.

You completely avoided that. How in the world did the RAW data diverge so greatly yet come together magically with the adjusted HadCRU in celebration of the alleged UNPRECEDENTED warming?

Snippet from Phil Jones interview:

A - Do you agree that according to the global temperature record used by the IPCC, the rates of global warming from 1860-1880, 1910-1940 and 1975-1998 were identical?

"So, in answer to the question, the warming rates for all 4 periods are similar and not statistically significantly different from each other."

excuse me but I must snicker for a bit....

I dont think you will be able to support this (noisy rural areas) other than MAYBE a HadCRU cut and paste:

"We use only the finest temp selections based on our secret sauce (formulas) which you (general population) are just too ignorant to understand so you gotta trust us."

 

[milkweed]

Incidentally, were you aware that the IEA (demonstrably) have a vested interest in promulgating anti-AGW science?

No more so than Al Gore, Pachuri, Jones, Mann, US gov, UK gov et all and the BILLIONS in funding, higher taxes and redistribution.

Actually, with the snippets of info out there, seems the pro-AGW side stands to lose a whole lot more 'vested interest' and talk about getting your fingers into the cookie jar (free market).

 

[Trippy]

Apparently you didnt follow the link at the bottom. Might not have even read the whole blog being as it clearly states the link is in russian (which I dont read)

Here is the link in the blog to the russian report:
http://www.iea.ru/article/kioto_order/15.12.2009.pdf

Its IN FUCKING RUSSIAN!

I found the link elsewhere thankyouverymuch.

Apparently you're wrong.
1. How do you know I don't read Russian?
2. How do you know I wasn't able to find an online translator, for example google translate http://translate.google.com/#

I much prefer to get my translations first hand.

And please, temper temper, I've been civil to you, the least you can do is return the favour.

Thats your interpretation of intent.

I see it showing when you use the entire available data, you see quite a divergence from what is presented by HadCRU as fact.

Show me again where I questioned this?

I also explaine dthat this was because of the increased noise in the extra data sets.

I also noted the tendency of HadCRU (assuming the russian report was accurate on these percentages) of them dumping longer records, gravitating to urban areas and leaving grids out of the data, even when rural stations exist.

I saw clear comparisons between urban (HadCru selected) and rural stations (hadCRU drops) which did show higher temps in urban, lower temps in rural, within the same grid area.

See above.

Thats your allegation. Support it with evidence. I find it impossible to believe a rural station would be noisier than an urban station.

I don't need to, it's all right there in the IEA paper.
Take a look for yourself.
Or do you not know what R[sup]2[/sup] is?
It's the correlation coefficient.
It measures how well a measured trend fits the data.
The closer to zero the correlation coefficient, the worse the fit.
A completely random data set has a correlation coefficient of zero.
I don't need to prove anything.
YOU need to learn more about the topic you're arguing about.
And as for this statemen:
"I find it impossible to believe a rural station would be noisier than an urban station"
It's a logical fallacy, it's an argument from personal incredulity, which is generally classed as an argument from ignorance.

It must have been really noisy in the 1930s because thats some remarkable divergence RAW vs HadCRU.

And?
You think that the historical records should neccessarily be as accurate as todays records?

You completely avoided that. How in the world did the RAW data diverge so greatly yet come together magically with the adjusted HadCRU in celebration of the alleged UNPRECEDENTED warming?

See above.

Snippet from Phil Jones interview:

A - Do you agree that according to the global temperature record used by the IPCC, the rates of global warming from 1860-1880, 1910-1940 and 1975-1998 were identical?

"So, in answer to the question, the warming rates for all 4 periods are similar and not statistically significantly different from each other."

excuse me but I must snicker for a bit....

Quite, although, I notice you've quoted the interviewers question, and the interviewer paraphrasing the interviewee's response, without quoting the interviewee's response itself.
That smacks of intellectual dishonesty, please excuse me while I puke a little bit.

I dont think you will be able to support this (noisy rural areas) other than MAYBE a HadCRU cut and paste:

"We use only the finest temp selections based on our secret sauce (formulas) which you (general population) are just too ignorant to understand so you gotta trust us."

More bullshit. It's not my fault you don't understand basic descriptive statistics, perhaps in future you might do some research for yourself before mouthing off.

 

[milkweed]

Apparently you're wrong.
1. How do you know I don't read Russian?
2. How do you know I wasn't able to find an online translator, for example google translate http://translate.google.com/#

Your just playing word games.

So I will counter with,

If you were engaging in an honest discussion, you would have noted the link was to the russian version.

Snip opinion

You mean like this for coefficient?

From page 14:

It turns out that the data not included in the HadCRUT sample and not used to calculate the global temperature are systematically much more extensive than the used data. The temperature series with the data series completeness coefficient exceeding 90% are used less than in the quarter of all cases (23%). Instead, the half-empty data series (with completeness coefficient below 50%) are used for two-thirds (66.7%).

Moreover, after Russian data are processed the HadCRUT experts, they sometimes suffer some losses that are difficult to explain. For example, the Hadley Centre has cut the temperature series for the Sortavala station as provided by Roshydromet, see figures 5 and 6.

Additionally, its not a completely random data set.

Again, I will ask you to support your statement of noisy rural RAW data with HadCRUs documentation on this, the russian dataset.

You ASSUME it was dropped for noise. You put your faith in the secret sauces mixed up by hadCRu.

You havent posted any proof. You have posted rants which are easily dismissed with a simple copy/paste.

Its no wonder Phil Jones is starting to come clean.

 

[Trippy]

Your just playing word games.

So I will counter with,

If you were engaging in an honest discussion, you would have noted the link was to the russian version.

That's all you've got?
Bullshit.

You mean like this for coefficient?

From page 14:

It turns out that the data not included in the HadCRUT sample and not used to calculate the global temperature are systematically much more extensive than the used data. The temperature series with the data series completeness coefficient exceeding 90% are used less than in the quarter of all cases (23%). Instead, the half-empty data series (with completeness coefficient below 50%) are used for two-thirds (66.7%).

Moreover, after Russian data are processed the HadCRUT experts, they sometimes suffer some losses that are difficult to explain. For example, the Hadley Centre has cut the temperature series for the Sortavala station as provided by Roshydromet, see figures 5 and 6.

Completness coefficient is not the same thing as correlation coefficient (which is what R[sup]2[/sup] means, and the fact that you would bring this nonsense up only serves to demonstrates you don't have the foggiest clue what you're going on about).

Additionally, its not a completely random data set.

Didn't say it had to be.

Again, I will ask you to support your statement of noisy rural RAW data with HadCRUs documentation on this, the russian dataset.

You ASSUME it was dropped for noise. You put your faith in the secret sauces mixed up by hadCRu.

You havent posted any proof. You have posted rants which are easily dismissed with a simple copy/paste.

Look for you self.
Here's one of the figures fomr the ORIGINAL paper.

The Red data set is the set that HCRU used (Makhachkala (Махачкала)).
The Blue data set is the set the russians think that HCRU should have used(Buinaksk (Буйнакск)).
The Red data set has a correlation coefficient of 0.085.
The blue data set has a correlation coefficient of 0.001

This is the consistent pattern through the entire fucking paper. HCRU consistently chose the data sets with the higher correlation coefficient, and by definition of what the correlation coefficient means, chose the least noisey data sets.

Did I EVER ONCE fucking state that this was WHY HCRU chose the data?
No, I didn't.
I only ever presnted it as an observation, and POTENTIAL explanation you moron.

Oh, and for the record? The ONLY thing that I have put faith in is that the Russians have done their calculations PROPERLY, and represented the HCRU data set ACCURATELY.

 

[milkweed]

That's all you've got?
Bullshit.

Thats all you got? One graph with an R2 out of 21 pages?

The paper states in the very beginning they were not applying manipulations and working with the RAW data.

The RAW data will have a lower R[sup]2[/sup]. I think its a pretty standard process to note it as such when its being compared to numbers with R[sup]2[/sup] factored.

And it doesnt explain why HadCRU rejected the rural setting for the 500K plus urban area does it?

Again, its not a random dataset.

From page 14:

It turns out that the data not included in the HadCRUT sample and not used to calculate the global temperature are systematically much more extensive than the used data. The temperature series with the data series completeness coefficient exceeding 90% are used less than in the quarter of all cases (23%). Instead, the half-empty data series (with completeness coefficient below 50%) are used for two-thirds (66.7%).

Moreover, after Russian data are processed the HadCRUT experts, they sometimes suffer some losses that are difficult to explain. For example, the Hadley Centre has cut the temperature series for the Sortavala station as provided by Roshydromet, see figures 5 and 6.


You seen a doctor yet about your anger issues? I seriously thought of ignoring you when you introduced this paper into a portion of the discussion revolving around Phil Jones.

No wonder Phil Jones starting to come clean.

 

[Trippy]

unbelievable.

Thats all you got? One graph with an R2[/sup] out of 21 pages?

The paper states in the very beginning they were not applying manipulations and working with the RAW data.

Look douchebag - BOTH of the data sets are the RAW DATA.
The Red set and the Blue set, the higher correlation coefficient is because the data has a better fit with the fitted trend, not because one set has been adjusted, and the other hasn't.

The RAW data will have a lower R[sup]2[/sup]. I think its a pretty standard process to note it as such when its being compared to numbers with R[sup]2[/sup] factored.

Technically true, but completely irrelevant in the context of the discussion you're avoiding (see above).

And it doesnt explain why HadCRU rejected the rural setting for the 500K plus urban area does it?

Yes, it does.
Because the Urban Area gave the higher correlation coefficient. Do I have to write it in Russian? How about simplified chinese? Would that be helpful? Should I get the statement translated into a non english language? Go on, name your language!

Again, its not a random dataset.

Completely AND TOTALLY fucking IRRELEVANT!

You seen a doctor yet about your anger issues?

You're the one that started the shit slinging, and yes, he said I have an allergy to morons, moronic statements, and idiots, that has the untreatable result that when I hear an ignorant moron going on about something they provably know nothing about, I have an overwhelming urge to strangle the nearest living thing. The Dr suspects that there may be complex-partial epileptic seizures involved, but I'm waiting on the tests.

I seriously thought of ignoring you when you introduced this paper into a portion of the discussion revolving around Phil Jones.

No wonder Phil Jones starting to come clean.

Hey, why not, you've done nothing but ignore verifiable facts so far.

The point of bringing the paper into the discussion, as I believe I have already outlined was that it demonstrates that your accusation that the raw data isn't available, and the HCRU results are unrepeatable, is FALSE.