CERN: Climate Models will need to be revised

[waldo]

Yes. It is called Real Climate and it was set up in 2003 with the specific purpose of trashing McIntyre.

oh my! This is simply too good to pass on... ya, ya, TimG, it's allllllll about McIntyre! (note: disciple TimG bowing to McIntyre's gawd complex (/snarc))

[waldo]

So you're saying at least four years of warming is weather because it was under an article with "weather"in the title? Give me a break.

... as I said, I simply defer to your linked references title... the one that speaks to Nov 1922 Weather... ya, that one! In any case, yes... most certainly, most emphatically, anything in the timeframe of your suggested "4 years"... is weather.

Their analysis is only as good as their data. There was nothing sophisticated about gathering data at that time. By 1918, the captain said there had been great change throughout the seasons. According to the graph, this didn't happen.

clearly, feel free to revel in your anecdotal reference... certainly, why would your ilk consider any semblance of scientific data gathering/processing/analysis, when you can simply fall back to an anecdotal account of "a supposed" testimonial of some sea-faring individual... notwithstanding no scrutiny applied to the regional/localized nature of said anecdote, or scrutiny of any kind, for that matter.

GISST is what was used prior to HadISST. Before satellites came into the picture, data collection was a bit primitive.

do tell... flushing you out has it's rewards - in relation to my clearly highlighted association to HadISST, you felt it prudent to, in your scientific anecdotal prowess, presume to make alternate reference. Again, do tell.....

[TimG]

Multivariate regression methods are insensitive to the sign of predictors.

I rebutted this statement earlier:

1) The algorithm that Mann uses is insensitive to sign.

Reply: It is true that Mann uses an algorithm that does not care if a proxy is positively or negatively correlated but the problem in this case is caused by the bad data in the correlation period that is used to determine whether a proxy positively or negatively correlated. This bad data causes the algorithm to choose the wrong sign for the lake sediment proxies.

Mann is basically trying baffle people with bullshit. There is simply no way to calibrate those lake sediments to temperature using any algorithms. The fact that the his algorithm ended up reversing the of the proxies proves that his algorithm is nothing but a statistical meat grinder that produces no meaningful results.

Here is McIntrye's reply to the Mann 2008 SI reference you posted:

http://climateaudit.org/2010/08/01/the-no-dendro-illusion/

Neither of the figures you reference support your argument because they either leave the tree rings in or leave the sediments in. What Mann needs to do is publish a correction to his paper that acknowledges that removing the lake sediments as he should results in reconstruction that depends on the tree rings - a direct contradiction of one of major 'findings' of the paper.

Edited September 15, 2011 by TimG

[waldo]

Michael, your inference presumes upon an actual error... again, per the (now many times linked) MannBradleyHughes comment reply:

Multivariate regression methods are insensitive to the sign of predictors. Screening, when used, employed one-sided tests only when a definite sign could be a priori reasoned on physical grounds. Potential nonclimatic influences on the Tiljander and other proxies were discussed in the SI, which showed that none of our central conclusions relied on their use.

I rebutted this statement earlier:

For reference, when waldo gets back his arguments will be:

1) The algorithm that Mann uses is insensitive to sign.

Reply: It is true that Mann uses an algorithm that does not care if a proxy is positively or negatively correlated but the problem in this case is caused by the bad data in the correlation period that is used to determine whether a proxy positively or negatively correlated. This bad data causes the algorithm to choose the wrong sign for the lake sediment proxies.

Mann is basically trying baffle people with bullshit. There is simply no way to calibrate those lake sediments to temperature using any algorithms. The fact that the his algorithm ended up reversing the of the proxies proves that his algorithm is nothing but a statistical meat grinder that produces no meaningful results.

the bullshit you speak to is yours. As close to this issue as you one-sidedly are, you must be selectively and purposely blind to the separate (multiple persons) analysis done on the transparently available Mann2008 matlab code... code analysis that reflects upon both of the separate processes referenced within that MBH comment reply (i.e., regression and screening, both of which are directly responding to the initial MM comment... and both of which, per the code, are factually correct). Again:

Multivariate regression methods are insensitive to the sign of predictors.

Screening, when used, employed one-sided tests only when a definite sign could be a priori reasoned on physical grounds.

your echoing the oft repeated McIntyre labeling of the algorithm as a "meat grinder", is simply further qualification of your lapdog presence, nothing more, nothing less! Don't bother offering me further ClimateFraudit links... there is nothing to be realized/gained from reading/referencing that tripe. By the way, what's stopping your "vaunted hero" McIntyre from issuing a follow-up... his own paper even... it's been a couple of years... what's he waiting for? What's stopping McIntyre from moving to bring forward his own reconstruction summary account... he has access to all the same proxies, he has representative code... he presumes to talk a code game... what's holding him back? Where's McIntyre's formal contribution to the science... why does he perpetuate his shell game within the bounds/sanctity of his hallowed sandbox. McIntyre... the blogging "scientist", who isn't a scientist, does no science/research, writes no papers, yet revels in perpetuating his "never-ending audit" - huckster extraordinaire!

[TimG]

As close to this issue as you one-sidedly are, you must be selectively and purposely blind to the separate (multiple persons) analysis done on the transparently available Mann2008 matlab code.

And that code uses the lake sediment proxies upside side down because the data in the calibration period is contaminated by bridge building. Simply repeating the irrelevant excuses used by Mann and co does not refute my point.

You trouble is you don't understand this stuff. The only thing to can do is mindlessly parrot stuff you pulled of an alarmist blog. An example of your total cluelessness can be found with your quote here:

Screening, when used, employed one-sided tests only when a definite sign could be a priori reasoned on physical grounds.

This statement is true but it has absolutely NOTHING to do with the argument that the lake sediment data has the wrong sign applied because contaminated data was used by the algorithm to determine the sign and, as a result, the algorithm picked the wrong sign.

Lets put this another way: lets say you had some data which was non-linearly related to temperature. For the first half of the record the value went up as the temperature increases. For the second half of the record the value went down as temperature increases.

Can you use multivariate regression methods on this data where the calibration period only uses data from the last half of the record? Please explain your answer.

Edited September 15, 2011 by TimG

[Pliny]

waldo, your concern seems to be that your cottage by the lake might get flooded out? Although that would be devastating for you the big picture isn't all about you. The issue is gradually being buried by important events that have more relevancy to the vast majority of people's lives rather than this propped up concern for our developmental progression and over-consumption, and it seems to be increasingly apparent that if current political solutions were implemented they would be more devastating than doing nothing. The probability of the loss of your cottage by the lake may be in graver danger from politicians than flooding.

The CERN experiment is simply more evidence that the science is not, as claimed, settled.

[waldo]

the bullshit you speak to is yours. As close to this issue as you one-sidedly are, you must be selectively and purposely blind to the separate (multiple persons) analysis done on the transparently available Mann2008 matlab code... code analysis that reflects upon both of the separate processes referenced within that MBH comment reply (i.e., regression and screening, both of which are directly responding to the initial MM comment... and both of which, per the code, are factually correct). Again:

Multivariate regression methods are insensitive to the sign of predictors.

Screening, when used, employed one-sided tests only when a definite sign could be a priori reasoned on physical grounds.

And that code uses the lake sediment proxies upside side down because the data in the calibration period is contaminated by bridge building. Simply repeating the irrelevant excuses used by Mann and co does not refute my point.

You trouble is you don't understand this stuff. The only thing to can do is mindlessly parrot stuff you pulled of an alarmist blog. An example of your total cluelessness can be found with your quote here:

Can you use multivariate regression methods on this data where the calibration period only uses data from the last half of the record? Please explain your answer.

as for not, 'understanding the stuff', you're the king... you initially brought up the Kaufman study and acted the clown over Kaufman's issued correction... words to the effect, "that if Kaufman could recognize his error with the Tiljander proxies and issue a correction"... so should Mann et al (2008) have. What you don't understand, where you wrongly/falsely parrot, is that the Kaufman calibration method (unlike that used within Mann2008), did care about the sign of the proxies (whether pre, post, modern period)... hence, the Kaufman correction. Again, the Mann2008 regression algorithm is blind to proxy orientation, whether EIV or CPS methodologies ("Multivariate regression methods are insensitive to the sign of predictors")... screening is relevant for one method, not the other ("Screening, when used, employed one-sided tests only when a definite sign could be a priori reasoned on physical grounds").

[waldo]

waldo, your concern seems to be that your cottage by the lake might get flooded out?

how trite! Let me know when you're ready to step-up your game... please... bring back Mr. Wizard Pliny!!!

Looks like Al Gore's world is flooding.

The CERN experiment is simply more evidence that the science is not, as claimed, settled.

another canard, one you beak-off on quite regularly... again, "the science is never settled"; confidence and uncertainty associated with certain facets of the science, may presume towards a "settling" - one that is, per norm, always subject to new advances/understandings. You subtlety throw out wide-sweeping implications of the initial CERN/CLOUD results... but you won't extend yourself to actually narrow those implications in terms of actual results or extrapolations therein. You're sitting in the same holding pattern as your stated "johnny-be-good" cohort and TimG - he who refuses to substantiate his claims that CERN/CLOUD provides an attribution study, one presenting an alternate causal link for today's enhanced/accelerated warming. C'mon Pliny... bring out Mr. Wizard Pliny... unleash the Wizard!!!

the only relevant/accurate thing you've had to say reflects upon models... but then, per your way/misunderstandings, you presume that model enhancements aren't something scientists, quite naturally/regularly, work for/towards. I stated from the onset of this (and the other threads CERN/CLOUD reference):

in fact... this is a most significant first release from CERN/CLOUD... one that's been anxiously waited upon these last few years by real legitimate scientists, particularly in regards to the science and modelling relative to atmospheric aerosols.

you suffer from a standard denier myopic that presumes climate scientists aren't actively engaged in model development/refinement - the reality is anything but that... there are organizations dedicated to model progression... there are regular world-wide expert meetings and workshops aimed towards improvements and refinements in methodology and underlying processes. The present state of atmospheric aerosol modeling is known and recognized to be lacking/uncertain... hence, the long-standing desire for a vehicle, like CERN/CLOUD, to help bring forward new learnings/understandings relative to atmospheric aerosols. Your title's attempt to cast aspersion towards climate models is misplaced... the "will need to be modified" is simply a reflection of anticipated results relative to aerosols - results that are eagerly awaited with presumptions towards bettering the models.

[TimG]

Again, the Mann2008 regression algorithm is blind to proxy orientation, whether EIV or CPS methodologies.

It does not make any difference how many times you repeat this nonsense it does not address my criticisms.

So tell me: what would you get if you feed a proxy that is negatively correlated for the first half of the record and positivity correlated for the last half of the record into Mann's algorithm? Do you think the results would have meaning? Show me where Mann claims that his algorithm can handle proxies that change sign half way through. Being 'insensitive to sign' does not mean that. Show me any reference you can find that says that multivariate regression of the type Mann did can handle data with those characteristics.

Edited September 16, 2011 by TimG

[TimG]

----

Edited September 16, 2011 by TimG

[waldo]

... again, per the (now many times linked) MannBradleyHughes comment reply:

Multivariate regression methods are insensitive to the sign of predictors. Screening, when used, employed one-sided tests only when a definite sign could be a priori reasoned on physical grounds. Potential nonclimatic influences on the Tiljander and other proxies were discussed in the SI, which showed that none of our central conclusions relied on their use.

as for not, 'understanding the stuff', you're the king... you initially brought up the Kaufman study and acted the clown over Kaufman's issued correction... words to the effect, "that if Kaufman could recognize his error with the Tiljander proxies and issue a correction"... so should Mann et al (2008) have. What you don't understand, where you wrongly/falsely parrot, is that the Kaufman calibration method (unlike that used within Mann2008), did care about the sign of the proxies (whether pre, post, modern period)... hence, the Kaufman correction. Again, the Mann2008 regression algorithm is blind to proxy orientation, whether EIV or CPS methodologies ("Multivariate regression methods are insensitive to the sign of predictors")... screening is relevant for one method, not the other ("Screening, when used, employed one-sided tests only when a definite sign could be a priori reasoned on physical grounds").

It does not make any difference how many times you repeat this nonsense it does not address my criticisms.

So tell me: what would you get if you feed a proxy that is negatively correlated for the first half of the record and positivity correlated for the last half of the record into Mann's algorithm? Do you think the results would have meaning? Show me where Mann claims that his algorithm can handle proxies that change sign half way through. Being 'insensitive to sign' does not mean that. Show me any reference you can find that says that multivariate regression of the type Mann did can handle data with those characteristics.

your parroted criticisms are the nonsense of shyte-throwing monkeys... making mountains out of molehills... perpetuating the charade of the "never-ending auditor", McIntyre!

read it again: "Screening, when used, employed one-sided tests only when a definite sign could be a priori reasoned on physical grounds". I quoted you, verbatim, the lengthy Mann2008 reference that speaks directly to, as titled, "Potential Data Quality Problems"... offering a full acknowledgment that, relative to the 4 proxy records in question, per the proxy author(s), "human effects over the past few centuries unrelated to climate might impact records." Please note my bold emphasis on the word, "might"... might as in... possibly... as in possibly not qualifying a definite sign on temperature correlation (i.e., "one that could be a priori reasoned on physical grounds"). So, in the case of the 4 Tiljander proxy records, Mann2008 chose not to associate a priori; rather, they chose to keep, to include, the 4 proxy records; presumably, given the uncertain "might" qualification of the proxy author(s) themselves. You seem hell-bent to extend upon that "might" qualification of the proxy author(s), to presume on some unsubstantiated claim of a definite (or... "what if") shifting correlation within the proxy, one you describe, theoretically, as, "negatively correlated for the first half of the record and positivity correlated for the last half of the record". So... what? Yes... so what?

Mann2008 chose to keep the 4 Tiljander proxy records and, (1) fully qualified and acknowledged uncertainties over the data quality of these 4 records and, (2) per the paper's Supporting Information SI - Fig. S9., presented plots of these 4 proxy records (amongst the greater grouping of plots of all of the proxy records that passed screening) and, (3) as previously stated, included comparative graphic reconstruction (SI - Figure S8), that included and excluded the, "potentially problematic series discussed"... comparative results that showed - "no significant difference" whether the 4 Tiljander proxy records were included, or not!

But don't let any of that stop the parroted criticisms of shyte-throwing monkeys... making mountains out of molehills... perpetuating the charade of the "never-ending auditor", McIntyre!

... more pointedly, Mann2008 and its SI clearly addresses what you presume to state was, "corrected and buried within the SI of another paper - Mann2009-SI". Specifically,

Mann2008 and its SI - Figure S7 & S8

:

=> Mann2008-SI - Figure S8: Comparison of long-term CPS NH land (S8a) and EIV NH land plus ocean (S8b) reconstructions (full global proxy network)

both with and without the seven potentially problematic series discussed

.

[TimG]

one-sided tests only when a definite sign could be a priori reasoned on physical grounds

Again - that provides no refutation of my arguments.

to presume on some unsubstantiated claim of a definite (or... "what if") shifting correlation within the proxy, one you describe, theoretically, as, "negatively correlated for the first half of the record and positivity correlated for the last half of the record". So... what? Yes... so what?

The author of the data says that the last 200 years is not data that can be used if 'special consideration' is taken. It is bad data. Period. The author of the data always says the data prior to 200 years has a physical correlation with temperatures that is opposite the apparent correlation between the bad data and temperatures. IOW, my example is not hypothetical. It is a description of what the Tilandjer data really is. Mann algorithm screwed up because it cannot deal with data where the sign switches half way through the time series. That is why he used it upside down.

Here are some quotes from Tjilander:

Fig. 5. Description of the sediment sequence. Two parallel cores are correlated according to magnetic susceptibility measurements. Relative X-ray density variation, varve thickness, magnetic susceptibility and loss-on ignition (LOI) variation within the past 3000 years are presented. High X-ray density corresponds to high amount of mineral matter (light grey value tints in X-ray film) and low X-ray density corresponds to dark grey values caused by a higher proportion of organic matter.

Note the last 200 years is a period of extremely low proportions of organic matter.

Periods rich in organic matter indicate favourable climate conditions, when less snow accumulates in winter by diminished precipitation and/or increased thawing, causing weaker spring flow and formation of a thin mineral layer.

So high organic matter indicates a warmer climate - but hold it - Fig 5 says the last 200 years has extremely low organic matter content!

However, the gradually increasing varve thickness during the whole 20th century probably originates from the accelerating agricultural use of the area around the lake.

Note the wording. Tjilander is not saying the data *might* be contaminated. Tjilander is saying it *IS* contaminated and the switch in sign is *probably* due to accelerating agricultural use. Mann is either deliberately lying or simply incompetent when he suggests that this data can be used during the last 200 years. http://www.climateaudit.info/pdf/others/Tiljanderetal.pdf Edited September 16, 2011 by TimG

[Pliny]

another canard, one you beak-off on quite regularly... again, "the science is never settled"; confidence and uncertainty associated with certain facets of the science, may presume towards a "settling" - one that is, per norm, always subject to new advances/understandings. You subtlety throw out wide-sweeping implications of the initial CERN/CLOUD results... but you won't extend yourself to actually narrow those implications in terms of actual results or extrapolations therein. You're sitting in the same holding pattern as your stated "johnny-be-good" cohort and TimG - he who refuses to substantiate his claims that CERN/CLOUD provides an attribution study, one presenting an alternate causal link for today's enhanced/accelerated warming. C'mon Pliny... bring out Mr. Wizard Pliny... unleash the Wizard!!!

The science is never settled. But waldo and Gore would like it to be. Once politicians are settled we get laws. I guess they have decided the science is settled.

You like it both ways, waldo. You would like the question to be settled yet it will never be settled.

That is the great skeptic advantage. It is settled, there is no god. But the science is never settled so it is never wrong or for that matter right. It is just you "beaking off" worried about your ocean front property - I guess Gore figures he has got the issue settled or did he sell his newly acquired ocean front property already?

the only relevant/accurate thing you've had to say reflects upon models...

You mean the temperature hasn't risen 1.5 degrees fahrenheit over the last century? Am I wrong about that?

The present state of atmospheric aerosol modeling is known and recognized to be lacking/uncertain...

I can't believe what I am reading. Did you just say that, waldo?? It doesn't seem to waver your certainty.

hence, the long-standing desire for a vehicle, like CERN/CLOUD, to help bring forward new learnings/understandings relative to atmospheric aerosols. Your title's attempt to cast aspersion towards climate models is misplaced... the "will need to be modified" is simply a reflection of anticipated results relative to aerosols - results that are eagerly awaited with presumptions towards bettering the models.

You agree with me then. Climate models will need to be revised.

The latest from the US political administration is a demonstration of it's inability to even address an issue let alone understand it. That would br the 535 million dollars they wasted on Solyndra. The green jobs plan also has other investments that are more than likely to tank. I think we are seeing the tip of the iceberg. People will nonetheless be less willing to have their taxpayer money wasted supporting political solutions to what may or may not be an anthropogenic problem, that hasn't been settled. I'm all for that.

Just a note on the post regarding the Petermann glacier. A quick glance at the website posted here on that would undoubtedly get some action from politicians who cursorily view the data and deduce that all that ice melted in two years - wow! Something must be done.

To his credit the scientist involved didn't claim that the ice had melted. He just claimed that he was flabbergasted at the change, which was that part of the glacier had broken off and floated away.

However, the presentation of the site could have fooled a lot of politicians into thinking that much of the glacier had melted in less than two years - something that I know from my grade eight geography class could never happen. A public education wasn't quite as useless and propagandistic in those days.

But in those days we were being made afraid of glaciers coming down upon us. Today we are being made afraid that they are melting away - so you see, as you say, climate models are always changing and the science is never settled. Just tell those politicians to simmer down a bit - they are posing a greater threat to our well-being than a rise in temperature of 1.5 degrees fahrenheit over a century ever did.

As you know I don't oppose the science I oppose those willing to sell us down the river by telling us the science is settled, pretty much.. well.. it never is...well.. sometimes...we know the earth is not flat for instance and are 99.99% certain....but we can never really know...if we lose the technology, like we did that of such things as the construction of the pyramids and stonehenge and other strange things....we may have to revert to the idea that the earth is flat.

Are you a politician? You seem so easily convinced.

[waldo]

read it again: "Screening, when used, employed one-sided tests only when a definite sign could be a priori reasoned on physical grounds". I quoted you, verbatim, the lengthy Mann2008 reference that speaks directly to, as titled, "Potential Data Quality Problems"... offering a full acknowledgment that, relative to the 4 proxy records in question, per the proxy author(s), "human effects over the past few centuries unrelated to climate might impact records." Please note my bold emphasis on the word, "might"... might as in... possibly... as in possibly not qualifying a definite sign on temperature correlation (i.e., "one that could be a priori reasoned on physical grounds"). So, in the case of the 4 Tiljander proxy records, Mann2008 chose not to associate a priori; rather, they chose to keep, to include, the 4 proxy records; presumably, given the uncertain "might" qualification of the proxy author(s) themselves. You seem hell-bent to extend upon that "might" qualification of the proxy author(s), to presume on some unsubstantiated claim of a definite (or... "what if") shifting correlation within the proxy, one you describe, theoretically, as, "negatively correlated for the first half of the record and positivity correlated for the last half of the record". So... what? Yes... so what?

Mann2008 chose to keep the 4 Tiljander proxy records and, (1) fully qualified and acknowledged uncertainties over the data quality of these 4 records and, (2) per the paper's Supporting Information SI - Fig. S9., presented plots of these 4 proxy records (amongst the greater grouping of plots of all of the proxy records that passed screening) and, (3) as previously stated, included comparative graphic reconstruction (SI - Figure S8), that included and excluded the, "potentially problematic series discussed"... comparative results that showed - "no significant difference" whether the 4 Tiljander proxy records were included, or not!

Again - that provides no refutation of my arguments.

whatever (parroted) "arguments" you think you've provided... none have merit. Do I need to extract the graphics within the SI - Figure S8 & S9 and pointedly link to them to force you to actually look at them? Again, SI - Figure S9 shows the plotted pre-1800 influence (as in no/minimal influence) of the Tiljander proxies... post-1800 influence is negligible to the overall reconstruction, per, again, SI - Figure S8 shows the Tiljander proxies have no influence on the reconstruction.

your references to Tiljander(2003) do nothing except highlight you haven't a clue about any of this... again, for the umpteenth time, problems with the proxies were fully understood and acknowledged within the Mann2008 paper (again, I quoted you, verbatim, the section on "Potential Data Quality Problems"). Again, Mann2008 made the decision to include the proxies and, given the raised profile on their possible contamination, went the extra step to show that they had no impact on the reconstruction... in or out, the Tiljander proxies have no affect on the reconstruction. Do you get that... do you acknowledge that?

again, many have run the transparently available Mann2008 code to realize the same published results... I would expect that if McIntyre and his lappers could show otherwise, they should move to publish something that could contradict those results. Otherwise, it's simply more of the same, "shyte-throwing monkeys... making mountains out of molehills... perpetuating the charade of the "never-ending auditor", McIntyre!"

[TimG]

SI - Figure S8[/url][/b] shows the Tiljander proxies have no influence on the reconstruction.

As I have said before: Mann 2008 makes the following claim which it considers to be important enough to put in the abstract:

Recent warmth appears anomalous for at least the past 1,300 years whether or not tree-ring data are used.

http://www.pnas.org/content/early/2008/09/02/0805721105.full.pdf+html

This claim is not true if Tiljander proxies are removed from the recontructions.

You keep posting figures that supposedly show they have no effect but all they show is you are either an idiot or liar since the figures you provide either remove the Tiljander proxies and keep the tree rings or remove the tree rings and keep the Tiljander proxies. If you remove BOTH Tiljander and tree rings you get a reconstruction that is NOT valid prior to 1500 - a fact which directly contradicts the claims made in Mann 2008.

the Tiljander proxies have no affect on the reconstruction. Do you get that... do you acknowledge that?

You are arguing a strawman. I have consistently said that the reconstruction only falls apart when BOTH tree ring and Tiljander are removed. I have consistently said that the issue could be addressed bu Mann submitting a correction to paper that withdraws the claim that "Recent warmth appears anomalous for at least the past 1,300 years whether or not tree-ring data are used.".

If you really think this is a molehill then why doesn't Mann do what any scientist with an ounce of integrity would do: withdraw claims that have been shown to be false.

Ironically, Mann is turning what should be a molehill into a mountain by refusing to correct his error because the fact that he can get away with it shows that the climate science community consists of a bunch of sniveling cowards who are too afraid to stand up for good science.

Edited September 16, 2011 by TimG

[TimG]

http://ignoranceanduncertainty.wordpress.com/2011/09/14/the-stapel-case-and-data-fabrication/

Some words on problems with statistical analysis which form the basis of a lot of climate science:

Let me speak to means and opportunity first. Attempts to more strictly regulate the conduct of scientific research are very unlikely to prevent data fakery, for the simple reason that it’s extremely easy to do in a manner that is extraordinarily difficult to detect. Many of us “fake data” on a regular basis when we run simulations. Indeed, simulating from the posterior distribution is part and parcel of Bayesian statistical inference. It would be (and probably has been) child’s play to add fake cases to one’s data by simulating from the posterior and then jittering them randomly to ensure that the false cases look like real data. Or, if you want to fake data from scratch, there is plenty of freely available code for randomly generating multivariate data with user-chosen probability distributions, means, standard deviations, and correlational structure. So, the means and opportunities are on hand for virtually all of us. They are the very same means that underpin a great deal of (honest) research. It is impossible to prevent data fraud by these means through conventional regulatory mechanisms.

Edited September 16, 2011 by TimG

[waldo]

too rich... after all your nonsense over regression processing, over the select tiljander proxy correlation, over the "meat-grinder" algorithm, over the pre-screening, over calibration, over contamination, over recognized priori, etc.,; you now want to accept the 08 paper claims and presume to refute them with... what exactly? Please... please... let the irony meter register off-scale! What exactly?

(as an aside, I have yet to, as you have now done (several times to me), called you a liar. I have no trouble labeling you for exactly what you are... none of which has, as I recall, ever entered into the realm of calling you a liar).

[waldo]

http://ignoranceanduncertainty.wordpress.com/2011/09/14/the-stapel-case-and-data-fabrication/

Some words on problems with statistical analysis which form the basis of these proxy reconstructions:

you pompous ass... what analysis and what basis presumes to have you make said associations, particularly with the slimy implications you're fronting. Spit it out, man!

[waldo]

As you know I don't oppose the science I oppose those willing to sell us down the river by telling us the science is settled, pretty much.. well.. it never is...well.. sometimes...we know the earth is not flat for instance and are 99.99% certain....but we can never really know...if we lose the technology, like we did that of such things as the construction of the pyramids and stonehenge and other strange things....we may have to revert to the idea that the earth is flat.

see:

another canard, one you beak-off on quite regularly... again, "the science is never settled"; confidence and uncertainty associated with certain facets of the science, may presume towards a "settling" - one that is, per norm, always subject to new advances/understandings.

You agree with me then. Climate models will need to be revised.

see:

you suffer from a standard denier myopic that presumes climate scientists aren't actively engaged in model development/refinement - the reality is anything but that... there are organizations dedicated to model progression... there are regular world-wide expert meetings and workshops aimed towards improvements and refinements in methodology and underlying processes. The present state of atmospheric aerosol modeling is known and recognized to be lacking/uncertain... hence, the long-standing desire for a vehicle, like CERN/CLOUD, to help bring forward new learnings/understandings relative to atmospheric aerosols. Your title's attempt to cast aspersion towards climate models is misplaced... the "will need to be modified" is simply a reflection of anticipated results relative to aerosols - results that are eagerly awaited with presumptions towards

bettering the models

.

The latest from the US political administration is a demonstration of it's inability to even address an issue let alone understand it. That would br the 535 million dollars they wasted on Solyndra.

see:

Don’t Be Fooled By the Solyndra Bankruptcy Circus — Solar Is Booming

Just a note on the post regarding the Petermann glacier.

see:

Petermann Glacier before-after-photos 2010-2011 - From Byrd Polar Research Center

... similar to/extension of earlier link provided... as stated, "implications to inland ice acceleration and draw-down of the ice sheet remains to be seen, but will be revealed by the GPS data recovered." If you are questioning the actual extent of Greenland ice-sheet melt, proper, please advise.

[TimG]

(as an aside, I have yet to, as you have now done (several times to me), called you a liar. I have no trouble labeling you for exactly what you are... none of which has, as I recall, ever entered into the realm of calling you a liar).

What? You don't like it when people treat you like you treat them? Hypocrite. In any case, I am merely expressing my frustration at your absolute refusal to address my actual arguments instead of the ones you make up. Aside: I do not read your ad homs (which sometimes make up 100% of your posts) - I skip over them looking for any actual substance in your posts. So I would not know if you used the word liar or not.

In any case, I am more that willing to have a civil debate. Why don't you lead by example?

I take it from your response that you are:

1) Unable to provide a credible justification for using the Tiljander proxies in Mann 2008.

2) Unwilling to admit that the Tiljander proxies do "matter" since one of the key claims of Mann 2008 (the no-tree ring reconstruction one) depends on using Tiljander.

Edited September 16, 2011 by TimG

[waldo]

(as an aside, I have yet to, as you have now done (several times to me), called you a liar. I have no trouble labeling you for exactly what you are... none of which has, as I recall, ever entered into the realm of calling you a liar)

What? You don't like it when people treat you like you treat them? Hypocrite. In any case, I am merely expressing my frustration at your absolute refusal to address my actual arguments instead of the ones you make up. Aside: I do no read your ad homs - I skip over them looking for any actual substance in your posts. So I would not know and don't really care if you used the word liar or not.

hypocrite? As I stated, I have not, as I recall, ever stooped to your level of calling another MLW member a liar. You do it with regularity. Your arguments? What arguments? All I read is regurgitated McIntyre swill... you're a fine parrot! While you're presuming to make an actual argument, don't forget to answer the following, "what exactly" questioning... again, "what exactly", are you presuming to use to refute the 08 paper claim - exactly!

too rich... after all your nonsense over regression processing, over the select tiljander proxy correlation, over the "meat-grinder" algorithm, over the pre-screening, over calibration, over contamination, over recognized priori, etc.,; you now want to accept the 08 paper claims and presume to refute them with... what

exactly

? Please... please... let the irony meter register off-scale! What

exactly

?

I take it from your response that you are:

1) Unable to provide a credible justification for using the Tiljander proxies in Mann 2008.

[waldo: I'm not required to provide a justification... ask the paper's authors. They are a part of the proxy mix (4 of the ~1200 proxies in total... if I recall correctly). Your parroting has you questioning their usage - good on ya, bully.

Lets recap: of these 4 proxies that has you, "flying shyte", the proxy author did not construct a temperature series - none exists. As I'm aware, no chemical analysis of the 4 proxies has ever been done. Certainly, given the proxy authors paper, questions exist over possible contamination of the proxies in most recent years (related to human infrastructure related activity). The Mann08 paper thoroughly acknowledged questions concerning data quality... proxies passed the papers screening processing and were calibrated, accordingly. The only question to arise is whether or not the calibration is on/close across the full range... one aspect of testing the sensitivity of this is to, quite obviously, check the significance of the overall reconstruction with the proxies in compared to their exclusion... this was done within the Mann08 paper, as repeatedly stated to you, over and over. There was no significant affect on the overall reconstruction if the 4 proxies were left in... or removed. Again, no significant affect - no appreciable difference.]

2) Unwilling to admit that the Tiljander proxies do "matter" since one of the key claims of Mann 2008 (the no-tree ring reconstruction one) depends on using Tiljander.

[waldo: duh! The "doesn't matter" context is in terms of the overall reconstruction impact... the 4 proxies in versus the 4 proxies out... what discernible difference can be shown in comparing the in vs. out. As stated, now too, too, many times, Mann2008 - SI - Fig. S8 shows that comparison; i.e., no discernible difference to the reconstruction.]

[waldo]

http://ignoranceanduncertainty.wordpress.com/2011/09/14/the-stapel-case-and-data-fabrication/

Some words on problems with statistical analysis which form the basis of these proxy reconstructions:

you pompous ass... what analysis and what basis presumes to have you make said associations, particularly with the slimy implications you're fronting. Spit it out, man!

hey now... no need to edit back-peddle on my account!

http://ignoranceanduncertainty.wordpress.com/2011/09/14/the-stapel-case-and-data-fabrication/

Some words on problems with statistical analysis which form the basis

of a lot of climate science

:

why such a broad-sweeping accusation... surely you have much more to say, surely you must want to target your aspersion, fabrication, fraud, etc., implications. Surely! Clearly it's time for another round of manufactured doubt and uncertainty, ala your fevered past Hackergate fixations. Surely... isn't there more... from you?

[TimG]

Your arguments? What arguments? All I read is regurgitated McIntyre swill... you're a fine parrot!

I can tell you never even read what McIntyre says so you are hardly qualified on comment on what I am repeating and what I am expanding on to deal with the arguments presented in thread.

"what exactly" questioning... again, "what exactly", are you presuming to use to refute the 08 paper claim - exactly!

I posted this a long time ago.

http://www.mapleleafweb.com/forums//index.php?showtopic=19451&view=findpost&p=707870

From the Mann 2009 SI - a graph that only appeared because of repeated criticisms by McIntrye.

Figure S8: Sensitivity of NH mean reconstruction to exclusion of selected proxy

record. Reconstructions are shown based on “all proxy” network (red, with two standard

error region shown in yellow) proxy network with all tree-ring records removed (blue),

proxy network with a group of 7 long-term proxy with greater uncertainties and/or

potential biases as discussed in ref. S1 (brown) and both tree-ring data and the group of 7

records removed (green; dashed before AD 1500 indicates reconstruction no longer

passes validation).

Read those words "reconstruction no longer passes validation". This is Mann admitting that, despite using 1200 proxies as input, the entire reconstruction depends on tree rings and Tiljander. Edited September 16, 2011 by TimG

[waldo]

excellent! We have the "exactly"... you want to (presume) to refute Mann08 with Mann09... and you have no qualms, none whatsoever, in doing this. Has McIntyre a challenging reconstruction to Mann08... a challenging paper? No, he doesn't... he lives in a trumped up denier blogfest world, one you frequent with abandon! Where's the formal McIntyre (or other) challenge? After 2+ full years... what's he waiting for?

[TimG]

Has McIntyre a challenging reconstruction to Mann08

Yawn. Here we go again! Instead of actually addressing arguments you try to change the topic. McIntyre submitted a comment to PNAS explaining the problem. Mann pretended there was no problem and the incompetent editor allowed him to get away with it. McIntyre did what was expected. The scientific journal failed.

Here is Gavin Smith acknowledging that Tijlander matters comments on RC:

Gavin, So just to be clear with regard to your response to 525. Under either method (CPS or EIV) it is not possible to get a validated reconstruction to before 1500 without the use of tree rings, or the Tijlander sediments. I understand, of course, that as you remove proxies that the ability to project backward will naturally diminish.

[Response: That appears to be the case with the Mann et al 2008 network. Whether you can say more general things about medieval times using these and other proxies (cf osborn and briffa 2006) is another question. -gavin]

So are you ready to admit that the conclusions of Mann 2008 depend on Tijlander and that Mann should be issuing a correction? Edited September 16, 2011 by TimG